A Guide To Marketplaces 2nd Edition 03012018

User Manual:

Open the PDF directly: View PDF .
Page Count: 65

Download
Open PDF In Browser	View PDF

Contents
Introduction to Marketplaces

6

Marketplace types .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  . 8
Selecting the right market.   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   . 9
Success factors: how to win against the incumbent . . . . . . . . . . . . . . . 12
Marketplaces and network effects .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  . 15
Seeding, Growing, and Scaling a Marketplace

18

Stage 1: Seeding a Marketplace. . . . . . . . . . . . . . . . . . . . . . . . 18
Stage 2: Growing a Marketplace...How to spark the virtuous cycle .   .   .   .   .   .   .   . 21
Stage 3: Scaling a Marketplace .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .  22
Finding the Right Business Model for your Marketplace

29

Transaction fees vs. listing fees .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  . 29
Enhanced seller services . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
What to do when services are delivered offline?.  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  . 32
Finding the right pricing strategy.   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .  34
New Marketplace Types

36

On-demand marketplaces .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .  36
Managed marketplaces .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   . 38
Community-driven marketplaces.   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .  40
SaaS-enabled marketplaces.   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   . 40
Decentralized marketplaces.  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  . 42
Marketplace Metrics

44

Marketplace Tools

49

Working with Investors

51

What do investors look for in a marketplace?.   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .  51
Our Marketplace Funding Napkin.   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .  53
How do investors determine valuation multiples for marketplaces? .  .  .  .  .  .  .  . 55
Marketplace Exits

58

Conclusion

60

Resources

61

PREFACE TO THE SECOND EDITION
It is now over two years since we published the first edition of “A Guide to Marketplaces.” It
was our first experience with this kind of project and we weren’t quite sure what to expect.
We wondered if anyone would read or find the book at all. But we knew that marketplaces
were an important part of our portfolio and investment thesis, and felt that the existing
content geared toward marketplace startups was fragmented and sometimes inconsistent.
We were surprised by the book’s incredible reception. It has been downloaded more than
25,000 times since publication and was translated into Japanese. And, we have received
countless thank you emails from founders and others — we have appreciated each and
every note.
Two years is a long time in the tech world. Since we first published the book, millions of
dollars have poured into blockchain projects and cryptocurrencies. We have also seen
marketplaces evolve from products, to services, to information. At the same time, our work
with founders and startups has continued to inform our understanding of the challenges and
opportunities that marketplaces face. As such, we realized it was time to refresh the guide.
We have retained much of the previous organization, and have aimed to keep it a quick and
easy read. Major additions include an entirely revised section on decentralized marketplaces
as well as marketplace exits. And throughout the book, examples have been updated as
needed. We hope you find the second edition as useful as the first.

INTRODUCTION
The web changes everything. In just 25 years, the Internet has become so ubiquitous
that it’s hard to remember how things were done before.
Nowhere is this more true than with how goods and services are exchanged. Not too long
ago, the marketplace was a purely local affair. Artisans and farmers brought their goods to the
local marketplace to be sold. But when the marketplace moved online, it shattered the very
notion of local and global. You’re now able to buy anything and everything with just a few
clicks, and vendors have unprecedented access to a global audience.
Let’s take books as an example. Used and rare books were once limited to tiny backstreet
sellers. If they were lucky, book-lovers might happen upon a prized out-of-print edition
while browsing the aisles. Moving the process online brought about a radical shift. You
just need to type in an author’s name and title, and scores of choices will pop up from
booksellers all over the world. If there’s a book out there, it can be found. You can buy a
hard-to-find copy of a used hardback from someone in San Francisco, read it, then sell it
to someone else living in Dublin.
Boris Wertz saw this transition first hand. Back in Germany in 1999, he launched an online
marketplace for books, JustBooks. In 2002, JustBooks was acquired by AbeBooks and by
2008, AbeBooks was doing several hundred million dollars a year in transactions between
buyers looking for rare prints, limited runs, and used textbooks and the sellers who had
them. AbeBooks was subsequently acquired by Amazon in 2008.
The online marketplace is a true example of a green-field opportunity. Whether it’s rare
books or home repair services, an online marketplace brings buyers and sellers together
in ways that were never possible before. It creates new buying and selling opportunities.
And, unlike an e-commerce site, anyone can build a marketplace with minimal capital
requirements since there’s no inventory to buy, build, or manage.

As a result, the marketplace business model has attracted countless entrepreneurs. When a
marketplace works, it works really well. There’s high potential for long-term profits at scale.
And once a marketplace achieves liquidity, it’s hard for others to compete. That’s why you
probably can name only one online auction site.
But, as many entrepreneurs find out, it’s hard to get the virtuous cycle of supply and demand
to reach liquidity. You need to ramp up both sides in near unison. If there are buyers but
few sellers, the buyers leave. Yet if there are no buyers, how do you convince the sellers to
sign on?
After AbeBooks was acquired by Amazon, Boris moved into investing, first as an angel
and now with Version One Ventures, with Angela Tran Kingyens joining in 2013. Over the
past decade, Version One has invested in great marketplace companies, such as Drover,
Dwellable, Headout, Indiegogo, Kinnek, Shippo, Top Hat and VarageSale. We have learned
from and helped many passionate entrepreneurs looking to break down the walls for how
goods are bought and sold in their particular space.
What follows here are some of the insights we’ve learned while observing, funding, and
building marketplaces for more than two decades. There’s no single way to build and scale a
marketplace, but we hope this handbook provides some insight to help you in your journey.
Now let’s get to work.

— Boris and Angela,
March 1, 2018

1
INTRODUCTION TO MARKETPLACES

	 |
Introduction to Marketplaces

WHAT’S A MARKETPLACE?
An online marketplace is a type of e-commerce site that connects those looking to provide a
product or service (sellers) with those looking to buy that product or service (buyers). These
buyers and sellers may have had trouble finding each other before, and thus the marketplace
creates efficiency in an otherwise inefficient market.
eBay’s launch in 1995 sparked the first wave of product-focused marketplaces, and Uber’s
success in 2010 brought a second wave of marketplaces for services, and combinations of
product and service marketplaces (such as food delivery). And today, we’re approaching a
third wave of information marketplaces fueled by blockchain technology.
With the explosion of marketplaces, there are now sites to connect bands with music venues,
homeowners with snow plowing services, and students with used textbooks.
While these marketplaces spread across diverse market segments, there are two common
elements to every online marketplace.

1. Marketplaces aggregate many sellers and their inventory
What sets a marketplace apart from a standard e-commerce site is that the goods and services
are supplied by a third party. In most cases, the marketplace platform acts as a type of digital
middleman.
The independence of suppliers is an increasingly grey area when it comes to service
marketplaces. Most on-demand service marketplaces like Uber, Handy, and Postmates stress
the independence of their workforce, while other startups like Munchery in San Francisco and
MyClean in New York switched to full-time employees, in order to gain more control over the
customer experience.

2. Marketplaces include a transaction element
A marketplace differs from a listing site by taking care of at least part of the transaction between
buyer and seller. A true marketplace like eBay and Uber manages the entire transaction, from
listing to payment processing (with the service and goods delivered offline).

7

	 |
Introduction to Marketplaces

For the purpose of this handbook, we’re also including those marketplaces where transactions
are initiated but not processed on the platform — for example, lead generation sites like
Thumbtack where providers send quotes through the platform. By contrast, a review site like
Yelp is not a marketplace, since there’s generally no transactional element involved.

MARKETPLACE TYPES
A marketplace can be categorized in terms of its participants (on both the supply and
demand side):
•

P2P (peer-to-peer): This is a grassroots transaction where private individuals
sell their goods to peers. For example, the early days of eBay and Airbnb were
dominated by private individuals.

•

B2C (business-to-consumer): As a marketplace gains traction, its seller base
inevitably professionalizes as two things happen: 1) existing businesses see the
opportunity and sign on to the platform and 2) private individuals find success on
the marketplace and formalize their activities into a business.

•

B2B (business-to-business): While B2B marketplaces have lagged behind P2P and
B2C, this is now a fast-growing segment. Alibaba is the biggest player in the group,
and we’re seeing the emergence of a new wave of B2B startups, such as Kinnek,
focused on making procurement and supply chains more efficient for the long-tail
SMB market.

FAST FACTS: TOP MARKETPLACES
Alibaba

Etsy

488m buyers (Q3 2017)

1,891,000 active sellers and 31,680,000
active buyers (Q3 2017)

eBay
170m buyers (Q4 2017)

Uber

Airbnb

40m monthly active riders (Q3 2016)

Nearly 120m guest arrivals
(first half of 2017)

Amazon
2MM+ third-party sellers (Q2 2016)

8

	 |
Introduction to Marketplaces

TWO METRICS FOR ANY MARKETPLACE
We’ll discuss marketplace metrics in detail in Chapter 5, but the two key metrics for any
marketplace startup to understand are:

1. GMV (Gross Merchandise Value):
GMV is simply the total dollar value of everything sold through the marketplace in a given
time period. This is a different and much bigger number than the platform’s revenue.
Net merchandise value is GMV minus the value of returns (products) or cancellations
(services).
•

Accurately calculating GMV can be tricky for marketplaces where transactions
are initiated on the platform but completed offline.

2. Take Rate:
Take rate is the percentage of transaction value that the marketplace collects in fees. Setting
the take rate is typically the cornerstone of a marketplace’s monetization and business model.
•

The more value a marketplace provides, the more it’s able to charge in fees. For
example, if a marketplace enables suppliers to find new buyers, then suppliers
are more likely to tolerate higher fees.

•

The more fragmented the overall market, the higher a marketplace’s take rate
can be. That’s because high fragmentation reduces the negotiation power of
marketplace participants.

•

When transactions are concluded offline, it’s hard for a marketplace to track
those transactions and monetize based on take rate. In Chapter 3, we’ll discuss
other monetization strategies for these situations.

SELECTING THE RIGHT MARKET
Bill Gurley of Benchmark Capital wrote the definitive post on which markets are the best
fit for marketplaces: “All Markets Are Not Created Equal: 10 Factors To Consider When
Evaluating Digital Marketplaces.”

9

	 |
10
Introduction to Marketplaces

While Bill laid out ten factors, we believe six to be the most essential:

1. High fragmentation
It’s much easier to start a marketplace when there are a lot of suppliers and buyers. When
there are just a few suppliers, they’ll likely fight the arrival of a new intermediary in their
market and won’t want to share in the economics (meaning a low take rate for your
marketplace). In addition, a marketplace can create the most value in a highly fragmented
market, since there’s a true need to help buyers and sellers find one another.

2. The buyer/seller relationship: monogamy vs. playing the field
In those markets where buyers are fiercely loyal and use the same supplier every time, the
value of a marketplace is reduced. For example, once consumers find a doctor or cleaning
service they like and trust, it’s easier to stick with the known provider rather than take a
chance on, or spend time searching for, someone new.
Once consumers are comfortable with a certain provider, there’s little need to visit or use
the marketplace again. The same is also true for commoditized products where you source
products from the same supplier (e.g. diapers on the consumer side and raw materials for
businesses).

3. Higher frequency
Bill Gurley said, “All things being equal, a higher frequency is obviously better. GrubHub,
OpenTable, 1stdibs (for the designer) and Uber are all high frequency use cases, where
consumers can rely on the marketplace as a utility. Many failed marketplaces attack
purchasing cycles that are simply way too infrequent, which makes it much more difficult
to build brand awareness and word-of-mouth customer growth.”
When evaluating frequency, it’s important to differentiate between how often a buyer users
a service and how often they use the marketplace. For example, you might need a babysitter
every day, but you only look for a new babysitter every two years (back to point #2).

4. Total Available Market (TAM)
As with any new business opportunity, a proper TAM analysis is imperative: how big is the
total available market and how much can you capture? Keep in mind that the best online
marketplaces create new value; therefore, the current market size does not necessarily paint
an accurate picture of the opportunity.

	 |
11
Introduction to Marketplaces

For example, Uber’s convenience has greatly expanded the car service market to more users
and geographic locations. Gurley gives an excellent analysis on Uber’s impact on the overall
market size of the car-for-hire transportation market.

5. Transactional: being part of the payment flow
When buyers and sellers need the marketplace to exchange money, the potential take rate
is much higher than when transactions happen offline. Think about it this way: When a
marketplace handles the transaction, it pays the supplier their portion of the revenue with
the fees automatically deducted, instead of having to send them a bill at a later date. This
makes fees much more palatable to most sellers.

6. The ability to expand the overall market and create value
Virtually all marketplaces create efficiency: they make it easier for buyers and sellers to find
each other and complete a transaction. But, not all marketplaces create value equally. For
example, marketplaces may connect schools with substitute teachers, visiting doctors with
hospitals, commercial landlords with retailers interested in setting up pop-up shops.
These all certainly increase efficiency, but they might not necessarily expand the market.
The likely reason is because there’s finite supply and finite demand. In other words, a new
marketplace won’t create a new crop of teacher or doctors, or school and hospitals. And
in the absence of a new online marketplace, both sides would probably find each other
eventually, albeit in a less efficient way.
The best marketplaces can tap into new groups of participants and grow the overall number
of transactions. The sharing economy is a great example of creating value. People who never
thought they’d be landlords or hotel proprietors are now hosts on Airbnb. And, people who
never thought they’d transport people for money are now drivers on Uber. These platforms
empower whole new groups of people to become sellers. More sellers = more participation
on the platform = more liquidity = more transactions = more overall value.
You don’t need to hit all six criteria (or all ten on Bill’s list) to build a good marketplace. But
the better you score, the higher the overall potential of the marketplace. And underlying all
these six factors we consider the magnitude of potential value creation.

	 |
12
Introduction to Marketplaces

SUCCESS FACTORS: HOW TO WIN
AGAINST THE INCUMBENT
The narrative of the little guy going up against the giant is baked into the culture of Silicon Valley.
There’s always an incumbent marketplace (even if it’s Craisglist), so finding the right business
model to break into an existing market is critical for any new marketplace.
Any newcomer should focus on the incumbent’s most important weakness and that weakness will
be different for each situation. Some of the common strategies are:

1. Lower the take rate
According to Albert Wenger of Union Square Ventures:
“I don’t see take rates of 20% or more as sustainable in the long run. Why? Because
they (a) impose too heavy a tax on marketplace activity and (b) produce too much
profitability for the marketplace operator. The first is bad from the perspective of
efficiency as it will crowd out some transactions that would benefit from taking place
in the market. The second will provide the incentive for competing marketplaces to be
created.”
For example, Craigslist offered free listings to compete with the paid listings in the traditional
classified sections of newspapers. Applying Clayton Christensen’s theory of the Innovator’s
Dilemma, most incumbents find it hard to follow down the market. They brush off an upstart
as being too small or too low end to matter. By the time the incumbent finally gets around to
addressing the new innovator, the upstart has already found traction.
However, unseating an incumbent based on lower take rate alone is difficult, since the network
effects of the existing marketplace will make it hard for a newcomer to build any momentum.
You’ll typically need to differentiate in other ways in addition to lowering the take rate.

2. Go vertical
A startup can unbundle a generalized marketplace and focus on creating the best product for a
specific vertical. Many marketplaces have struck gold by picking one thing and doing it extremely
well. For example: StubHub vs. Craigslist and HotelTonight vs. Expedia.

	 |
13
Introduction to Marketplaces

This is a strategy that has been spectacularly effective for many verticals and the market map
that venture capitalist Andrew Parker’s created back in 2010 has become the blueprint for
tracking the different companies that have chosen to carve out niches from Craigslist. David
Haber offered an updated chart at the end of 2012 to demonstrate just how effective the vertical
strategy can be.
In some cases, marketplace businesses build traction in smaller verticals before expanding their
reach. Geographical marketplaces often nail one location before expanding into new territories.
Indiegogo first went after the indie film market before opening up to other categories. Expanding
outward from the niche often happens organically as sellers start listing items in new categories.
For example, Etsy sellers began listing craft supplies and tools and this category has become one
of Etsy’s largest today.
Keep in mind that category expansion doesn’t always work as planned. When Boris was COO
at AbeBooks, he thought they could expand the site into new books. Despite a strong selection
at competitive prices, the new books category never lived up to expectations and most of the
buyers stayed with their existing retailers (mainly Amazon).

3. Develop a 10x better product
Another strategy is to build a product that’s ten times better than the incumbent’s. This has
been the key to VarageSale’s success (one of our portfolio companies that has done a great
job building mobile-first, social, and safe local marketplaces) vs. Craigslist. Or Uber’s success
against the taxi industry. Or think about how Airbnb’s fortune changed when they started using
professional photography.
Keep in mind that sometimes a 10x better product itself isn’t enough. Benchmark’s Sarah
Tavel wrote that multi-billion dollar businesses build a 10x better product while also recasting
incumbent cost structures. In other words, a better product might not be enough if only the
very wealthy will be willing or able to afford the service or convenience. She wrote that “a 10x
product and save people money” is the undercurrent running through dozens of breakout
companies, including Uber, Airbnb, and WhatsApp.
Lastly, these kinds of opportunities are rare and hard to pull off, because the incumbent’s
network effects often outweigh any product improvements.

	 |
14
Introduction to Marketplaces

GO VERTICAL: THE SPAWN OF CRAIGSLIST

Source: The Gong Show, Andrew Parker, 2010

Source: David Haber, TechCrunch, 2012

	 |
15
Introduction to Marketplaces

4. Unique inventory
Last but not least, a marketplace startup can focus on the supply side, and identify
opportunities to bring unique inventory to underserved markets. This strategy certainly
helped Airbnb against HomeAway, as well as Etsy vs. eBay.
In some cases, you can even create supply that didn’t exist before. Prior to Airbnb’s success,
there wasn’t much incentive for people to post their unused space. Not many people would
be comfortable advertising a room in their apartment on Craigslist. Now, Airbnb is a primary
source of income, or at least a major supplemental source of income, for a lot of folks.

MARKETPLACES AND NETWORK EFFECTS
One of the main reasons why investors and entrepreneurs love marketplaces is the power
of their network effects and how it creates a high degree of defensibility for startups. Keep
in mind that “network effects” means more than just a large number of users — network
effects kick in when the value of a product depends on how many other users there are.
When a new user/member is added to the network, it increases the value of the product or
service to all other users. This increased value can come in the form of cost reduction (in
user acquisition as an example), higher liquidity (in a marketplace), stronger community or
deeper relationships (in social networks), etc.
Generally speaking, network effects are categorized into two types: direct and indirect.

Direct Network Effects
The simplest network effects are direct: an increase in usage leads to a direct increase in
value for all other users. Social sites like Facebook, Twitter, and Wattpad have a direct
network effect. If you join Facebook and start sharing content I’m interested in, Facebook
becomes more valuable to me
When we think about direct network effects, we consider the roles of the participants. In
a marketplace, users are buyers, sellers, or sometimes both. Likewise, in a community or
social platform, users can be content creators, content consumers, or both.

	 |
16
Introduction to Marketplaces

In a marketplace, the roles of a buyer and seller are typically distinct — effectively creating
two-sided network effects. When there’s overlap between buyers and sellers, it’s easier to
get your network off the ground at the beginning as you’ll achieve liquidity more quickly.
For example, in Etsy, Craigslist, and Airbnb, we see some overlap between buyers and
sellers (or hosts and guests in Airbnb’s case).
While overlap helps a marketplace at the beginning, two-sided network effects add to
the defensibility of your marketplace over time (since there’s more fragmentation). For
example, with Uber, there’s little overlap between drivers and riders.
We also need to consider how network effects fall on the local-global spectrum:
•

Local network effects: The value-add of a product may not increase with the overall
size of its user base, but is dependent on the size of a smaller subset. In this case, a
user’s experience might be influenced directly by the number of people and actions
of a small subset of users (i.e. connections through the user’s personal/professional
network or neighborhood). We see local network effects at play with instant
messaging, social networks, and the on-demand economy.

•

Global network effects: These take place with a larger-sized network with few barriers
in payment or language. We see global network effects on a global marketplace like
eBay. Keep in mind that some marketplaces, like Etsy, will fall between local and
global network effects, as there can be shipping and language limitations.

Indirect Network Effects
Network effects can also be indirect. In these cases, when more people use a product or
network, it sparks the production of complementary products and goods — thus increasing
the value of the original product. A common example is with hardware and software: the
more people that use a hardware product, the more likely it is that developers will build
software and apps for that hardware.
Indirect network effects can also apply to marketplaces. Developers often build applications
and products on top of APIs of existing marketplaces and platforms. For example, Shopify
has a very active developer/value-add app ecosystem. In addition, countless services sprout
up to support popular platforms — like Airbnb management services (Guestly) and driver
rental car options for Uber/Lyft (Drover).

	 |
17
Introduction to Marketplaces

How do network effects differ from virality?
When information can be shared rapidly and widely from one user to another, or when
the rate of adoption increases with adoption, we have virality. In other words, the product
grows faster as more users adopt it (to a certain limit).
Network effects and virality often go hand-in-hand, but not all network effect products are
viral, and not all viral products have network effects. For instance:
•

Marketplaces often have low virality, even though they have network effects. While
you can achieve some virality through incentives, the two-sided network effect does
not lend itself to adoption of users on one side to come on-board with invitations/
interactions from the other.

•

News outlets, gaming, and communication products can be highly viral products
without strong network effects.

As you build your marketplace, don’t fall into the trap of confusing network effects
with virality.
Marketplaces have low virality, but network effects are critical to scaling and sustaining a
marketplace. Think about how users benefit each time someone else participates, and in
what way: directly or indirectly. Understanding how network effects impact your startup
brings you one step closer to building a more defensible product.
In addition, we can’t underestimate the importance of data network effects to the success
of a marketplace. Tophatter (a mobile e-commerce marketplace with no prior data
infrastructure or data background) has been refining its marketplace by making decisions
about what products go into the marketplace (e.g. seller-side analytics to determine
which items and sellers are good). Their goal is to build a personalized marketplace for
each buyer. They built the right data infrastructure to generate and evaluate personalized
recommendations for users on the platform.
The result has been a 30% increase in clicks. The company is expecting to make $100
million in revenue in 2017 (compared with nearly $40 million in 2016). Much of this growth
is due to how data-driven the company is, especially when it comes to recommendations.

	 |
18
Seeding, Growing and Scaling a Marketplace

2
SEEDING, GROWING, AND
SCALING A MARKETPLACE

STAGE 1: SEEDING A MARKETPLACE
The earliest days in a marketplace are a tricky time. There’s a chicken and egg problem when it
comes to supply and demand: customers need supply, and suppliers need customers. But it’s
nearly impossible to ramp up supply and demand in lock step.
We’ve found that in most cases it’s best to focus on building up the supply first. That’s because
there’s more incentive for sellers to invest their time in the early days. There’s zero motivation
for customers to stick around without any inventory.
So how do you go about seeding supply? Here are five common strategies:

1. Identify unique inventory
One of the best types of suppliers to target are those people who don’t already sell online. For
example, Etsy went after the producers of hand-made goods. These designers and crafters
found their unique goods buried within eBay’s massive inventory and typically resorted to
selling their items locally at craft shows, farmers’ markets, etc. Etsy came along and brought
these local experiences online.
The advantage of focusing on unique inventory is that it won’t take much effort to persuade
providers to join once they’re made aware of your site. In addition, these types of sellers will
typically bring along some of their own buyers – helping you ramp up your customer base at
the same time.

	 |
19
Seeding, Growing and Scaling a Marketplace

2. Convince existing sellers to list on your platform
If your ideal suppliers already have an online outlet, you’ll need to be creative and exert
more energy to convince them to join your young marketplace. Airbnb is a classic example.
Airbnb’s target customers and suppliers were using Craigslist to buy and sell lodging outside
of the standard hotel experience. Airbnb offered a more personal and trustworthy experience
than the free-for-all on Craigslist, but Craigslist had one thing Airbnb didn’t: a massive user
base and inventory.
As detailed in a Growth Hackers case study, Airbnb found some interesting ways to leverage
the existing supply on Craigslist, including emailing Craigslist listers and encouraging them
to check out Airbnb. While one could argue this approach was pretty close to spam, it most
likely enabled Airbnb to cheaply reach tens of thousands of their targeted suppliers each day.

3. Bring customers to a provider
With service marketplaces, one of the easiest ways to build early liquidity is to approach a
provider with a customer opportunity already in hand. A ready-to-go, paying customer is
always much more compelling than talk about future potential.
For example, some restaurant delivery startups started out by scraping the web for menus
from local area restaurants. Then when a customer places an order on their site, the startup
goes to the restaurant to get the order fulfilled.
There are two challenges with this approach. The first is that the provider’s information (e.g.
restaurant menu) needs to be available and up-to-date. Then, there’s always a risk that the
selected provider won’t convert and you’ll end up leaving your customer high and dry. Even
if a different provider fulfills the order, you may not be delivering a perfect experience to
the customer.

4. Pay for inventory
During the early days at Boris’ startup JustBooks, their first growth hack was buying books
to list so they didn’t have an empty site. When Uber launched in Seattle, they paid town car
drivers to idle. This generated the supply, and once the customers and money started rolling
in, they switched those drivers over to commission.

	 |
20
Seeding, Growing and Scaling a Marketplace

Buying inventory artificially creates supply, improving the way the marketplace works at the
outset. It’s an easy way to get inventory in the early stage, but it’s not scalable as it obviously
gets expensive to both buy and manage inventory yourself.

5. Aggregate readily accessible inventory
The last strategy is to aggregate inventory that’s already listed somewhere, for example
through affiliate programs. This approach will provide the initial scale that you’re looking
for, but there are several key drawbacks. First, while you may have a lot of inventory, none
of it will be unique: why should buyers come to your site instead of the other sites you’re
pulling inventory from?
The second problem is that when you aggregate existing inventory, you run the risk of
becoming a cross-platform utility rather than your own marketplace with lots of highly
engaged users. Fred Wilson discusses this issue with regard to networks, but his words are
equally relevant to marketplaces:
“If the initial utility of an app is to connect to a bunch of networks, collect
information, present it, and then let the user engage with one or more of those
networks, what incentive is there for the user to engage directly with other users
of the app and help build a network inside of it?”
With that said, aggregating existing inventory can be an effective way to kick-start your
marketplace early on. That’s what our portfolio company Dwellable did before quickly
transitioning to their own inventory.
No matter which approach you use, it’s easier to build a strong community if there’s a large
overlap between buyers and sellers. For example, think about the number of Airbnb hosts
who stay with other hosts during their own travels. Likewise, sellers on Etsy also make
purchases since they appreciate handcrafted goods. Peer-to-peer marketplaces are more
likely to have sellers that double as customers and vice-versa.
The bottom line is this: if you are building a marketplace, you need to devise an effective
strategy for ramping up supply. The next stage is getting to that virtuous cycle of supply
and demand.

	 |
21
Seeding, Growing and Scaling a Marketplace

STAGE 2: GROWING A MARKETPLACE...HOW TO
SPARK THE VIRTUOUS CYCLE
The virtuous cycle is the holy grail for online marketplaces. In this positive feedback loop, a
high number of quality suppliers attract more customers; then more customers attract more
suppliers to join. This cycle continues as a self-sustaining growth engine until both supply and
demand reach critical mass to be “winner takes all.”
But how does a marketplace create the momentum for a virtuous cycle in the first place?

Identify and double down on the hot spots
One of the most important things you can do is identify and double down on the things that
work in your marketplace. As you start scaling, there will be many matches and transactions
between buyers and sellers. But not all matches are created equal. Identify where things are
clicking on both the supply and demand side. This could be in certain geographies, audience
segments, price points, and user behavior.
Then you’ll want to double down on these hot spots, often following Paul Graham’s advice
to do things that don’t scale. Don’t worry that the absolute numbers hardly seem worth the
effort. At this point you’re just trying to get the virtuous cycle going. If the market exists, you
can recruit your suppliers and customers manually, then switch over to less manual (more
scalable) methods.
“At some point, there was a very noticeable change in how Stripe felt. It tipped
from being this boulder we had to push to being a train car that in fact had its own
momentum.” Patrick Collison, Stripe
Airbnb is a classic example. Founders Brian Chesky, Nathan Blecharczyk, and Joe Gebbia
frequently traveled to NYC to acquire their early users. Then when they realized that high
quality listing photos were key to attracting customers (and differentiating the experience from
Craigslist), they rented a $5,000 camera and went door to door, taking professional pictures
of as many New York listings as possible. This approach led to two to three times as many
bookings on their New York listings.

	 |
22
Seeding, Growing and Scaling a Marketplace

“When New York took off, we flew back every weekend. We went door to door with
cameras taking pictures of all these apartments to put them online. I lived in their
living rooms. And home by home, block by block, communities started growing. And
people would visit New York and bring the idea back with them to their city.”
– Brian Chesky, for The Atlantic

Be Patient: marketplaces take time
With a typical SaaS or e-commerce startup, you probably should reassess your market or
model if you don’t see signs of traction after six to nine months. However, this timetable is
way too accelerated for marketplaces. Considering you need to establish both buyer and seller
communities, you will need more time to prove your business. It can take three years for a
marketplace to get going.
Looking at a Version One portfolio company, the crowdfunding platform Indiegogo was
founded in 2007, but its breakout year didn’t come until four years later.
This means that founders need to believe in their idea even when no one else does. But that
doesn’t mean turning a blind eye to the market. You’ve got to continually look for small signals
that you’re on the right track – such as increased word of mouth from early adopters, increased
repeat usage from buyers, increased listings from sellers, and positive user feedback.
Just as important, investors need to stay patient: no two-sided marketplace is built overnight.

STAGE 3: SCALING A MARKETPLACE
As a marketplace figures out how to bring buyers and sellers together and builds liquidity,
the virtuous cycle starts kicking in. Growth quickly accelerates. The ultimate goal is that
a successful startup grows into a big company. But as you begin to scale, you’ll need new
strategies to keep the forward progress going.

	 |
23
Seeding, Growing and Scaling a Marketplace

1. Foster trust and safety
A certain level of trust is required for any transaction to take place. During the early days
of your marketplace, you may have a small and passionate community. As such, trust and
safety are not significant concerns. However, as a marketplace grows and gains popularity, it
will inevitably attract some bad actors. As one TechCrunch commenter wrote back in 2008,
responding to an article on Airbnb:
“If this ever becomes mainstream, the whole thing will come crashing down. The
kind of people that *DO* rob, abuse, rape and murder people will start using these
systems, and that will spread legitimate fear, corrupting the whole thing. Pray that
this remains underground, that’s the only way it can survive.
Your job as a marketplace is to fight these bad actors proactively, so both buyers and sellers
have confidence in the platform.
Transparency is one of the most effective ways to establish trust and credibility. This can be
done with a rating system, user reviews, or testimonials. Some platforms act on this data to
weed out the bad users (both buyers or sellers). For example, you’ll never see an Uber driver
with a rating lower than three stars. Uber filters these drivers out, so customers don’t have
to sift through driver reviews before getting a ride. On the flip side, riders are also rated, so
drivers can feel safe with their pickups too.
Providing some level of guarantee — be it for service quality, delivery time, or payment — is
critical for instilling trust on both sides. If your marketing claims you’ll be at your customer’s
doorstep in 15 minutes, you need to make sure you can deliver on that promise. If a buyer
never receives their item, you’ll need to be ready with a money-back guarantee. And,
whenever the terms of the transaction are broken (on either side), you need to follow up
with a very personal response.

	 |
24
Seeding, Growing and Scaling a Marketplace

CASE STUDY: AIRBNB’S TRUST & SAFETY TEAM
The following is extracted from Growth Hackers case study,
Airbnb: The growth story you didn’t know.
Airbnb has managed to accomplish the once unimaginable: make people feel safe
and secure when opening up their home to strangers or staying in someone else’s
place. There have been some hiccups, company missteps, and security issues
over the years, as can be expected with a platform that has more than 200 million
guests since 2008 and 4 million properties listed worldwide (as of August 2017).
Here are some of the key ways that Airbnb has managed to overcome the inherent
trust issue and instill confidence within their community:
1. Professional photographs did a lot to inspire trust on one side of the
equation, as it helped to ensure that listings weren’t complete dumps, and also
verified addresses of listed properties.
2. In 2011, Airbnb introduced Airbnb Social Connections, which
leverages Facebook Connect. As a result, guests weren’t necessarily staying with
total strangers, but friends of friends, people from the same school, or hosts that
were recommended by friends.
3. In response to some specific incidents in 2011 (when host homes were
ransacked and robbed), Airbnb created a dedicated Trust and Safety
team that’s on call 24 hours a day, as well as a neighborhood hotline where
neighbors can report any questions or concerns.
4. In 2012, Airbnb partnered with Lloyds of London to expand their
guarantee to cover damages to host property from vandalism or theft. They
now cover every booking with their $1,000,000 Host Guarantee (with some
limitations and exceptions).

	 |
25
Seeding, Growing and Scaling a Marketplace

2. Support your power sellers
In addition to solving trust issues, the next step to scaling a marketplace is to support your
power sellers — those sellers who earn a living off your marketplace. This approach may seem
counter-intuitive for some startups that are typically used to pleasing end users (the buyers).
But to be successful at scale, a marketplace needs good supply.
Anand Iyer, former Head of Product at Threadflip, stated that Threadflip initially focused
more on the buyer experience but shifted strategy when the net promoter scores indicated that
sellers needed more support. He describes the importance of focusing on supply/sellers:
“But for buyers, very little is actually out of the ordinary. Most people are used to
buying or even ordering services online, either from Amazon or eBay. On the flipside,
fewer people are used to shipping products, providing taxi service, or turning their
home over to other occupants. For all these reasons, it’s crucial that marketplaces
focus first on devising an elegant, instructive and — above all, easy — experience
for suppliers.”

What are some of the ways a marketplace can support their top sellers? Here
are some real-world examples:
eBay offers an official PowerSeller program where qualified sellers get priority customer
support, unpaid item protection, a “top-rated Seller” designation, and other promotional
offers. eBay does a great job organizing its local eBay On Location events to connect its sellers,
provide additional learning/selling resources, and build a sense of community and identity
around being an eBay seller.
Threadflip sends its sellers shipping supplies like boxes, and even free mannequins to help
them display used clothes (after they discovered that people are most likely to buy used clothes
they see on a mannequin rather than worn on a real person). In addition, Anand mentioned
how Threadflip’s leadership team has regular one-on-one calls with top sellers to find out what
they like about the experience and what Threadflip could be doing differently.

3. Develop an ecosystem
Think about add-on products, services, and experiences that give sellers a deeper connection
with your marketplace. These are incredibly effective ways to differentiate the selling
experience on your marketplace from other sites, and lock sellers into your site.

	 |
26
Seeding, Growing and Scaling a Marketplace

The first approach is to develop or offer these add-on products and services yourself. For
example, Uber connects its drivers with exclusive leasing and financing offers. And Uber
drivers who complete at least 100 trips each month can get a partner Fuel Card for gas and car
maintenance purchases. The card offers a minimum 1.5% discount at participating gas stations/
mechanics and any charges are deducted from one’s Uber earnings. Other benefits are that the
card doesn’t require a credit check and won’t impact the driver’s personal credit.
The second approach is to support the third-party services that spring up around your site.
Marketplaces have a tremendous opportunity to scale into a platform with an entire ecosystem of
value-added services and startups.
The success of Airbnb has sparked an ecosystem of value-added services, with startups offering
property management services, guest screening, catering, and guest hop services for hosts. We
saw a similar ecosystem emerge around eBay, with companies forming to take care of shipping,
photos, insurance, and consignment.
Shopify, a SaaS provider of e-commerce stores for SMBs, has done an excellent job of
understanding and formalizing the role that third-party developers play in enhancing Shopify’s
overall value to sellers; they even brought on a Head of Apps and Third Party Developer
Ecosystem. In the Shopify Apps Store, sellers can find third-party apps for selling, marketing,
inventory management, customer support, shipping, reporting/analytics, and more.

4. Prevent leakage
With many marketplaces, there’s a real risk that buyers and sellers will settle the transaction
off platform, preventing the marketplace from capturing any revenue from the transaction it
worked so hard to put together. This is particularly true with “monogamous”, non-commoditized
services. For example, once you find a good babysitter or doctor, you won’t want to look around
for a different provider.
Charles Hudson provided a great analysis on the role of a well-designed rating system to combat
leakage. In order for a rating system to have an impact, suppliers need to value their rating and
buyers need to trust the quality of the rating system. For example, with Uber, drivers and riders are
matched partly by their ratings and scores. This creates a strong incentive for both the provider and
buyer to care about their score and keep the transaction on the platform.

	 |
27
Seeding, Growing and Scaling a Marketplace

5. Build a moat
As any marketplace gains popularity, its supply inevitably becomes less and less unique. From
Boris’ first-hand experience, booksellers once were exclusive to AbeBooks, but then started to
list with Amazon after it launched its marketplace. Uber and Lyft drivers typically begin with
one service, but often end up driving for both.
And marketplace uniqueness can be diluted even when suppliers don’t intentionally leave. For
example, Etsy struggles with design copying, including complaints of major retailers massproducing products that look extremely similar to its handcrafted artisan designs.
When building a marketplace, it’s safe to assume that the uniqueness of your supply will fade
over time as your suppliers seek out opportunities on other marketplaces and competitors look
to grab a piece of the pie that you discovered.
You can adopt two strategies to minimize the impact of these competitive dynamics: protect
the supply and protect buyer mindshare.
•

Protect supply: You can’t keep your supply unique when sellers list on other sites. Don’t
give your sellers any reason to seek out other marketplaces. For example, you can lower
listing/transaction fees for unique inventory, tie sellers to your site through reviews
(which cannot be transferred to other marketplaces), or find some other innovative
model like Uber’s leasing model.

•

Win buyer mindshare: While strategies like Uber’s leasing model can slow down the
drain of your unique supply to other marketplaces, they won’t stop it altogether. It’s
much more effective, and important, to win buyer mindshare.

For example, Etsy has done an incredible job driving loyalty and repeat usage. As reported in its
2016 Annual Report, a large and growing number of Etsy buyers have made purchases on two or
more days within the year.

	 |
28
Seeding, Growing and Scaling a Marketplace

Etsy’s Active Buyers by Purchase Type

Multiple Purchase
Days
One Purchase
Day

Source: Etsy, 2016 Annual Report
Uber and Lyft have been able to lock in customer mindshare by becoming a home screen
app on customer’s mobile phones. In the age of mobile where long-tail discovery (whether
paid or organic) is less important, becoming a home screen app is critical. This trend is
even stronger in those countries that have skipped the desktop altogether and gone straight
to mobile.
If buyer mindshare is the key to building a moat against the competition, you need to have
the right product mix to become a frequent destination for your customers. Then employ an
aggressive mobile strategy to get on the home screen of as many buyers in your target group
as possible.

3
FINDING THE RIGHT BUSINESS
MODEL FOR YOUR MARKETPLACE
Many entrepreneurs come up with a great idea for a marketplace, but then fail to find the right
business model that will translate their activity into profits. The challenging part of monetizing
a marketplace is that buyers and sellers are looking to complete transactions with one another,
and fees introduce friction to the process. Whichever party is charged a fee will naturally try to
conduct the transaction off-platform whenever possible.
Finding the right business model depends entirely on the characteristics of your market:
•

•

If customers are comfortable buying products/services over the marketplace with little
back and forth, then seller transaction fees (complemented with additional seller
services) are most likely the way to go.
If the marketplace acts purely like the classifieds (no transactions), listing fees make
more sense.

In general most marketplaces take fees from their sellers — either through listing fees, lead
generation fees, or transaction fees. The primary reason for monetizing the seller is simple:
after you’ve seeded the marketplace with initial supply, demand becomes the limiting factor
for marketplace growth. Charging buyers creates friction and limits their participation.
In this chapter, we’ll look at several of the most common monetization models: transaction
fees, listing fees, end-user/buyer subscriptions, and seller services. We’ll also discuss how to
find the right pricing/rake scheme.

TRANSACTION FEES VS. LISTING FEES
Let’s start with the transaction fee, where the marketplace takes a cut of each transaction
generated through the platform. It’s the fairest monetization model for suppliers and vendors,
as they only pay a fee if and when they sell something.

	 |
30
Finding the Right Business Model for your Marketplace

This model encourages more suppliers to join the platform and thus increases the liquidity
of the marketplace’s supply for two reasons: first, when you take away the upfront fee it’s
easier to join, and second, if you only charge when a sale is made, you decrease the supplier’s
risk of losing money. A transaction fee model also scales nicely: the more sales your platform
generates, the more revenue you bring in.
By contrast, listing fees can have the opposite effect. Charging suppliers to list on your site
will inevitably discourage some from listing. You might find your marketplace soon hits a
ceiling where you can’t get any new suppliers to join. In addition, this model is less fair than
transaction fees as listing fees hit all vendors equally — no matter how many sales they end
up making on the site.
However, there are upsides to listing fees. They assure some level of quality control for a better
end user/buyer experience. When sellers need to pay to list, they are more likely to list items
with a high chance of selling, as well as invest time in each listing. In other words, you won’t
have to worry about providers flooding your marketplace with low quality products and listings.

ENHANCED SELLER SERVICES
A recent trend has been for marketplaces to monetize by charging for optional seller services.
In this case, you can make the base listing free and charge for enhanced services like better
placement. The low entry cost encourages more suppliers to list (increasing liquidity), but
revenue can be hard to come by since only a small percentage of suppliers will choose to pay
for the better placement.
Such a freemium approach works best for companies that serve a huge market. Even if just a
small percentage of suppliers pay for additional seller services, their base is so big that they
can build a large business (e.g. Yelp). However, if you are tackling a niche market, you’ll never
reach the scale necessary for a freemium model to work as the core source of revenue.
If we analyze Etsy’s 2016 Annual Report, we see that the company’s premium seller services
are growing much faster than base transaction fees. Seller services cater toward the
marketplace’s power sellers: promoted listings, direct checkout, shipping labels, etc. Revenue
from these seller services overtook general marketplace transaction fees in 2015.

	 |
31
Finding the Right Business Model for your Marketplace

Since the average Etsy seller is too small to worry about fulfillment efficiency, these enhanced
services take a higher rate from Etsy’s power users. This approach works very well for the
company and community. By relying less on monetizing the smaller sellers, Etsy ensures that
small sellers can afford to stay on the platform and contribute their unique inventory that
buyers want. Etsy then takes a higher rate from the big sellers that can most afford it due to
their scale.
This strategy means that Etsy doesn’t have to choose between unique inventory and
monetization. They keep competitors at bay (because there’s no reason for the small sellers
to leave) while growing the company’s take rate and revenues.

Etsy’s Annual Revenue Breakdown:
Seller Services Revenue Overtakes Transaction Fees in 2015

Markets (transaction fees)

Source: Etsy, 2016 Annual Report

Seller Services

Other

	 |
32
Finding the Right Business Model for your Marketplace

WHAT TO DO WHEN SERVICES ARE DELIVERED OFFLINE?
A business model based on transaction fees seems like the simplest and most effective
approach for any marketplace. However, it’s hard to pull off when the service is delivered
offline.
For example, Thumbtack is a platform that helps people find service professionals
(photographers, painters, home contractors, movers, etc.). They started with a transaction
fee model. But when a plumber goes to a customer’s house to fix a leaky faucet, it’s hard
for Thumbtack to know when the service was performed and for how much. That makes
charging by transaction fees tricky.
Furthermore, when a marketplace connects the buyer and seller before facilitating the
transaction, it weakens the marketplace’s ability to capture value.
Thumbtack pivoted to a lead-based model: they make money by charging suppliers to
contact customers. Vendors see all the details for a particular job (what, when, where) and if
they feel they are a good fit, they pay Thumbtack to be introduced to compete for the job.
Keep in mind that the lead-based model only works with new connections (i.e. discovery).
It’s ineffective in those cases where there’s a lot of repeat purchases from the same vendor.
In addition to a lead generation revenue model, there are some other strategies for service
marketplaces. One approach is to stay as close to the transaction as possible. Elance offers
hourly/project tracking and billing solutions that encourage both the provider (freelancer)
and the customer to stay on the platform throughout the transaction.
The other approach is to productize the service as much as possible, with boxed offerings
that include pre-defined scope, duration, pricing, deliverables, etc. This simplifies the end
user experience, enabling them to complete a transaction in just a few clicks. The less
back and forth between provider and customer, the more likely they are to complete the
transaction on platform.

	 |
33
Finding the Right Business Model for your Marketplace

SUBSCRIPTION FEES
While most marketplaces extract a fee from their sellers, a few charge subscription fees to their
buyers. Angie’s List and Care.com are the most notable examples.
Angie’s List charges buyers a membership fee in order to search the directory and view ratings.
They also charge advertising fees to the seller.
Care.com has a freemium model, with value-added services on both sides of the marketplace.
Families (buyers) need to pay a subscription fee to message a caregiver through the platform.
Until you pay, identities are masked. Paying caregivers can move up in the search results,
provide a preliminary background check, and get “priority notification” of opportunities.
Care.com’s IPO prospectus offers a detailed look at their free and premium services. The
starred features represent paid/premium services.

Care.com: Breakdown of free and paid/premium services,
Source: Care.com IPO Prospectus

	 |
34
Finding the Right Business Model for your Marketplace

There are a few risks with the subscription model:
•

Charging buyers can limit the number of potential customers who join your site.
Your value proposition needs to be very strong to entice consumers to pay a
subscription in order to buy a product or service.

•

There’s also the question of how long any buyer will remain a paying subscriber.
Unlike a SaaS app which can be continually useful, the lifespan of a marketplace’s
utility can be rather limited. Once a subscriber finds a provider they like (whether
it’s a doctor or babysitter), they may not look for another provider for years.

Care.com expressed this challenge in their prospectus:
“Currently, most of our paid memberships are monthly memberships, and
the average paid membership length for our consumer matching solutions is
approximately seven months. As a result, we must regularly replace paying
members who allow their membership to lapse with new paying members either
by converting existing non-paying members or by attracting new members to our
service.”

FINDING THE RIGHT PRICING STRATEGY
Bill Gurley’s article “A Rake Too Far: Optimal Platform Pricing Strategy” is recommended
reading for any startup trying to determine their pricing strategy. He begins by saying:
“It may seem tautological that a higher rake is always better – that charging more
would be better than charging less. But in fact, the opposite may often be true. The
most dangerous strategy for any platform company is to price too high – to charge
a greedy and overzealous rake that could serve to undermine the whole point of
having a platform in the first place.”
What’s the problem with levying too high a fee? First, as Gurley explains, if you charge an
excessive rate, that cost ultimately gets passed along to the consumer and the “pricing of
items in your marketplace are now unnaturally high.” Second, high rakes give suppliers a
reason to look elsewhere and you become vulnerable to competitors from below.

	 |
35
Finding the Right Business Model for your Marketplace

Here’s a telling example: in the late 1990s companies like Expedia and Travelocity were the
online travel leaders in Europe, and they took a rake over 30% with certain packages. This left
an opening for Booking.com (part of Priceline Group) to come in with a lower 10% rake and
sign up nearly every small hotel in Europe. This resulted in greater supply and selection for
consumers – and Booking.com is the undisputed leader today.
What’s even more interesting is that the average rake at Booking.com is higher today since
merchants bid up their rake for better placement, similar to the Google Ads model.
Just like with Etsy’s Premium Seller Services model, this approach offers the best of both
worlds: low starting fees that encourage broad supplier adoption (and supply liquidity)
along with added revenue from those suppliers who want more exposure, or whatever other
enhanced services you offer. You won’t turn anyone away due to high fees, but can still boost
your take as suppliers compete with one another.

4
NEW MARKETPLACE TYPES
At their core, every marketplace is similar. On one side, there’s a seller (supply), on the other
side a buyer (demand). The marketplace acts as an intermediary to bring these two sides
together. But there can be a lot of innovation in how a marketplace handles these transactions,
takes care of its seller and buyer communities, and approaches monetization.
As the industry matures, we’re seeing new marketplace models emerge, including on-demand,
managed, community-driven, SaaS-enabled, and decentralized. Here’s a quick look at each kind.

ON-DEMAND MARKETPLACES (E.G. UBER, LYFT, ZEEL)
The widespread use of smartphones has helped create a consumer culture that’s conditioned
to expect anything and everything on-demand. With the “uberification” of many different
markets and services, the smartphone has turned into a remote control that can summon
anything from cookies to legal services.
These on-demand marketplaces match jobs with independent contractors on the fly, supplying
labor service or products on demand. For example, there’s Instacart to buy and deliver
groceries, TaskRabbit for odd jobs, and the list continues.
In many ways, today’s on-demand economy is a continuation of the sharing economy, or
collaborative consumption — exemplified by Airbnb, where people turn their underused assets
(i.e. home, bed, car, etc.) into a source of revenue. Now, with on-demand services, people
aren’t renting out their assets, but instead their spare time.
While entrepreneurs and investors were once excited by the massive opportunity, it’s
important to realize that the on-demand service model that lifted Uber’s valuation to a high
of $70B won’t work for every vertical. In many cases, the hype once driving on-demand
marketplaces has come and gone.

	 |
37
New Marketplace Types

The service marketplaces that have been able to break through and continue to scale follow
three key principles:

1. Underlying Commoditized Services
When it comes to hiring a ride, most of us are happy as long as a driver brings us from Point
A to Point B in a clean car without getting lost. This makes us pretty flexible in terms of who
delivers the service. For more complex services — e.g. haircuts, babysitting, and legal/medical
advice — it’s harder for customers to accept the idea that somebody at random will show up
each time. We develop preferences, and like to stick with one provider once they earn our trust.

2. High purchase frequency
The best marketplaces have high purchase frequency and regular usage. For city dwellers and
frequent travelers, taxis are used on a daily, or at least weekly, basis. Few other services have
such a high purchase frequency. On top of this, Uber also enjoys significant spillover effects as
travelers move from one location to another.
With high frequency use cases, customers fall into the habit of using the same service as long
as they’re satisfied. This makes it easy for a startup to become the “home screen app” for that
particular use case. By contrast, it is much harder to retain customer mindshare with lower
purchase frequencies. If customers need a yard cleanup just two times a year, they’re more
likely to begin the research process over again each time.

3. True on-demand use case
Many of the services that fall into the on-demand mobile services category aren’t actually “ondemand.” In most cases you don’t need a cleaning service or house painter to show up within
minutes, or even the same day. But, taxis are a different story.
A true on-demand marketplace requires sufficient liquidity on the supply side. Without enough
available drivers in a car service marketplace, customers will be left waiting on the curb.
This creates a large barrier to enter the market, since a new competitor needs to launch with
hundreds of providers, not just a handful.
By contrast, when services can be delivered with more flexible timing, it’s easier for
competitors to enter a vertical or new location and there’s less of a winner takes all dynamic.
As such, we can expect just one or two major players for a true on-demand service, while less
time-sensitive markets will be crowded with smaller companies.

	 |
38
New Marketplace Types

MANAGED MARKETPLACES
Over the past few years, we’ve seen the emergence of managed marketplaces. These are
marketplaces that don’t just connect buyers and sellers, but take on additional parts of
the value chain to deliver a better overall experience. With startups like thredUP, Luxe,
Opendoor, and Real Real, managed marketplaces have been a hot category for VC investment.
These types of marketplaces can be game-changers in their categories, particularly when
dealing with high-value ticket items. By adding managed services, marketplaces can create
new demand by helping buyers overcome the trust issues associated with most peer-to-peer
marketplaces. There’s a big difference in trust levels when looking for elder-care services on
a managed marketplace like Honor versus searching on Craiglist. As Anand Iyer wrote, trust,
safety, and risk management are basically tables stakes now for a managed marketplace: “what
draws most consumers to a managed marketplace is how the marketplace manages trust.”
Borrowing from Ezra Galston’s “Anatomy of a Managed Marketplace” and Anand Iyer’s
“The Evolution of Managed Marketplaces,” we believe that managed marketplaces share
three key characteristics:
•

The managed marketplace takes on additional parts of the value chain (beyond
just connnecting buyers and sellers) to provide a superior experience versus more
traditional peer-to-peer marketplaces. In many cases, the end user experience is a
core differentiator — where end users can be buyers, sellers, or both. For example,
by adding concierge and logistics support, a marketplace can entice sellers of luxury
items who may not have been inclined to go through the time and trouble to list their
items on a marketplace previously.

•

A managed marketplace can involve considerable risk involved — for example, should
you guarantee the sale and take on sellers’ inventory. If economic conditions worsen
rapidly, such as the “overnight recession” in 2008, you’ll end up sitting on inventory
that’s impossible to move.

•

Lastly, a managed marketplace can have a higher take rate than a pure marketplace
that just connects buyers and sellers. This can offset the premium service offerings
and added risk.

	 |
39
New Marketplace Types

thredUP offers an interesting case study in finding the right managed model. thredUP works
with a lot of children’s and mom’s clothing — not exactly big ticket items. It’s a great market
opportunity for re-sale, because as any parent knows, it’s easy to over-spend on kids clothing
and items are worn for only one season (or less). Moms were taking photos and cataloging their
clothing, but soon they didn’t want to do it any more. Photographing clothes is a hassle.
The company considered taking on these tasks to help out their sellers, but were advised
against taking any inventory. It was too risky, complex, and expensive for such small ticket
items. However, thredUP understood how much their sellers needed these service and went
ahead with a managed model. People can order a bag, throw their things in, and thredUP takes
care of the rest. After switching to a managed model, their business took off again. Inventory
quality and variety has improved, making the marketplace more attractive to buyers.
The takeaway? Sometimes it’s okay to deviate from conventional wisdom. thredUP listened to
what their users needed and delivered on those needs (and only those needs — you don’t want
to add costs by taking on more than what’s necessary).
The core question is: how much value can you unlock by managing parts of the marketplace
value without raising the risk profile too much?

Managed Marketplaces: Who sets the pricing?
Another element of a managed marketplace is when the platform itself sets the pricing, instead
of the sellers. In a traditional marketplace, sellers define their own prices. A great example
is Sidecar vs. Uber/Lyft where these ride-sharing companies adopted different pricing
strategies to target a different consumer segment.
Uber and Lyft standardized their pricing and service offerings. Consumers simply push a
button and a car shows up. Every Uber driver will charge you the same rate to take you from
Point A to Point B. By contrast, Sidecar drivers set their own rates. You had a better chance of
finding a cheaper ride with Sidecar, but you needed to spend time browsing through a
lot of choices (lowest fees, larger cars, etc.) in order to find what you want. Ultimately, Sidecar’s
“Wild West” model failed — letting supply set a price for a commoditized, undifferentiated
service just doesn’t make sense.

	 |
40
New Marketplace Types

COMMUNITY-DRIVEN MARKETPLACES
Some marketplaces leverage the community of buyers and sellers to create a unique
differentiator. Here, buyers and sellers visit the marketplace not just to complete a transaction,
but also for a sense of identity and belonging. They meet and form communities with others
who have similar interests.
Etsy is a prime example of a marketplace that has built a business around community and
culture. From Etsy’s 10th anniversary blog: “Over time, Etsy has come to represent something
even more powerful: an alternative to traditional commerce and a different, people-centered
model for doing business.” Etsy calls itself more than a marketplace: “we’re a community of
artists, creators, collectors, thinkers and doers.”
To support their community, Etsy added team functionality to its marketplace, letting users
(sellers and buyers) create and join teams around shared interests. They also host official craft
nights in Brooklyn and other local meetups.
Lyft has thus far been able to hold its own by building a strong culture and sense of identity
around its drivers. Driver meetups and community rallies are a longstanding Lyft tradition.
There’s also a private Facebook group (the Driver Lounge) where Lyft drivers can meet other
local drivers, share stories, connect over common interests, and organize local events. Such
efforts pay off in the form of better brand loyalty for the drivers, and a more unique culture for
riders.
And at Boris’ old company, AbeBooks, the seller community was incredibly active and the
marketplace gave them a place to come together as booklovers. To help foster this sense of
community, AbeBooks created online community forums, which evolved into book clubs,
science fiction groups, collector groups, help with forgotten authors, titles, etc.

SAAS-ENABLED MARKETPLACES
Today we’re seeing a new segment of companies that offer SaaS tools with the ultimate plan
of building a marketplace. Such companies attract users (either customers or sellers) with a
useful and typically free tool, then encourage them to participate in the marketplace. Chris
Dixon described this approach, “come for the tool, stay for the network.”

	 |
41
New Marketplace Types

SaaS-enabled marketplaces can be either B2B, B2C or C2C and the SaaS component can be
offered to the demand and/or supply side. Point Nine Capital mapped out a high-level overview
of the SaaS-enabled marketplace ecosystem.

Source: Point Nine Capital
OpenTable is a prime example. They offer various front-of-house software tools for
restaurants, like seating management. While OpenTable may be a leader in online reservations
today, these kinds of business tools attracted restaurants and gave them an easy way to list and
manage their incoming reservations (thus helping OpenTable build liquidity).
Zenefits is another example. They offer free benefits software for HR/business, and make their
money off commissions when someone buys insurance from their marketplace. Free software
can be a strong go-to-market strategy, as Tomasz Tunguz described:
“Free good software spreads quickly, enabling rapid customer acquisition. In
competitive markets or when pursuing customers that have been expensive to
acquire, free software creates a clear, and often defensible, competitive distribution
advantage.”

	 |
42
New Marketplace Types

However, developing a SaaS-enabled marketplace requires certain factors to be aligned. First,
you need to make sure that the consumers or businesses that sign up for your tool will be just
as willing to join and use your marketplace. This strategy will only work if the marketplace is
complementary to the tool. Second, you may have to contend with some sellers who want to
use your tool, but aren’t too eager to join a marketplace that puts them in competition with
others.
And third, market dynamics require ongoing use of the marketplace. OpenTable has
succeeded because people are always looking for a new restaurant to try, so the marketplace
provides continuous value. This approach won’t work in those cases where consumers prefer
a “monogamous relationship” like with doctors.

DECENTRALIZED MARKETPLACES
Over the past few years, we have been closely following the evolution of blockchain
technology and looking at the potential impact that blockchain will have on the marketplace
landscape.
In terms of decentralized marketplaces today, OpenBazaar is really the only game in town
for buying/selling physical products. It’s an open source project to create a peer-to-peer
commerce network using Bitcoin. With OpenBazaar, there’s no middle man: the platform
connects buyers and sellers directly. And since there’s no one in the middle of a transaction,
there are no fees and no restrictions on what goods can be listed and sold. OpenBazaar
realizes blockchain’s promise to bring power back to the edge (aka the users).
A decentralized marketplace is an interesting proposition and we are eagerly watching how
new marketplaces based on blockchain may disrupt industries. However, the big question
is if a decentralized platform will be able to create the same great customer experience that
we’ve grown accustomed to on centralized marketplaces.
Can a decentralized platform offer the same kind of UX, conflict resolution tools, and
customer service? And how much do these things this matter? Are customers willing to pay a
fee for a smoother, sleeker, more trustworthy experience?
The other question is will decentralized marketplaces be pushed to the fringe and just attract
buyers and sellers for illicit goods? If so, just how scalable are these kinds of use cases?

	 |
43
New Marketplace Types

These are important questions and we’re still not certain in what ways a decentralized
marketplace can disrupt a well-functioning centralized marketplace. Will a decentralized
marketplace replace a centralized one, or will blockchain technology become a part of an
existing marketplace?
While there may be challenges associated with replacing well-established marketplaces,
tremendous opportunities exist to enable something new and create a brand new generation
of marketplaces. Prime candidates are marketplaces for digital assets that allow anyone in the
world to conduct a transaction with any other participant without having a central instance
manage the transaction (e.g. payment settlement, etc.). Examples include:
•

•
•
•

Prediction markets like Augur or Gnosis that enable completely new financial
products (buying and selling options; insurances/betting on potentially any “event”
in the world)
Decentralized storage (Filecoin, Sia) and computing markets (Golem) that can
leverage underused computing power and replicate the Airbnb model for digital assets
Decentralized exchanges (0x) that allow for trading the ever rising amount of tokens
and currencies supporting the buying and selling of this new asset class
Data marketplaces (Ocean Protocol, OpenMined) that are decentralized
marketplaces around the collection and distribution of data

In these decentralized marketplaces, transactions between marketplace participants are
settled using the marketplace token. As the marketplace’s usage grows, those tokens will
hopefully further appreciate, making participation from the supply side even more attractive.
This will solve the chicken and egg dynamic that can pose such a problem for young
marketplaces. It will also create strong network effects as the platform grows, resulting in a
winner-takes-all where the biggest player is way bigger than its competitors.
This space is rapidly evolving and no doubt, the players, models and trends will be vastly
different just a few months from now. It’s an area we’ll be watching closely.

5
MARKETPLACE METRICS
As you grow, it’s important to understand how the business is performing: if your model is
financially viable, what’s working well, and what needs to be changed. But traditional business
metrics and KPI dashboards don’t necessarily capture the key factors that drive marketplace
performance and health.
Christoph Janz at Point Nine Capital created a KPI dashboard for early-stage SaaS. Inspired by
this, we have put together a KPI template for marketplaces.
You can access the Marketplace KPI Template via the Google doc here (make a copy of it and
then you can edit away).
Special thanks to Vijay Nagappan of MHS Capital for bringing the KPI template to life
and visualizing these KPIs using Plot.ly and Keen.io dashboards. Vijay created four starter
dashboards for summary, supplier side, demand side, and then a mashup of supplier side and
demand side data to see if there are any interesting trends. If you’re interested in creating your
own visualizations and dashboards, check out Vijay’s post — which includes links to Keen.io’s
templates, Plot.ly’s Web GUI, Plot.ly’s Python APIs and much more.

Samples of KPI dashboards created by Vijay Nagappan, MHS Capital, using Plot.ly and Keen.io

	 |
45
Marketplace Metrics

Our Marketplace KPI Dashboard, Available for download here
The dashboard is separated into three dimensions of your business’ efficacy: 1) overall
marketplace metrics, 2) seller/supplier metrics, and 3) buyer metrics.

1. Overall Marketplace Metrics

Gross merchandise volume (GMV) is the total sales dollar value for goods sold or
services purchased through the marketplace over a certain time. Given that GMV is one of
the most important marketplace KPIs, founders should track its growth rate on a monthly and
yearly basis, and understand its makeup by customer acquisition channel. With GMV and the
total number of transactions, we can compute the average order value (AOV) since:
GMV = # of Transactions * AOV

	 |
46
Marketplace Metrics

Revenue is the income that the company receives from facilitating connections in the
marketplace. It comes in the form of transaction fees, listing fees, and/or the offering of
premium seller/supplier services. With revenue and GMV, we can calculate take rate via:
Revenue = GMV * Take Rate
In addition to take rate, we can evaluate business efficacy by calculating the total customer
acquisition cost (CAC) of buyers and sellers/suppliers as a percentage of revenue.

2. Seller/Supplier Metrics
Supplier metrics are categorized into general and engagement KPIs, for example:

General KPIs
––

Number of suppliers

––

Suppliers growth rate

––

Number of listings (in a services marketplace, this is not applicable so perhaps
consider the number of active suppliers at a specific time instead).

––

Listings growth rate

––

Average listing price

––

CAC

Engagement KPIs on the supplier side
Engagement KPIs are most important to the supplier side, so make sure you track these at the
very least:
––

Cohort analysis: percentage of suppliers still active 1 month and/or 1 year after
signing up

––

GMV retention: average percentage of Month 1 GMV generated by suppliers in
Month 12

––

Concentration: percentage of revenue generated by the top 20% suppliers

––

Net promoter score (NPS)

	 |
47
Marketplace Metrics

3. Buyer Metrics
Similar to supplier metrics, we can categorize these into general and engagement:

General KPIs
––

Number of buyers

––

Buyer growth rate

––

Average dollar amount purchased per buyer

––

Average number of orders per buyer

––

Average order growth per buyer

––

CAC

Engagement KPIs on the buyer side
––

Repeat buyer contribution: percentage of buyers who have purchased more than
once; and percentage of GMV generated from buyers in previous months (note:
in terms of measuring network effects, it can also be valuable to track the time
between first-time purchase and second-time purchase)

––

GMV retention: average percentage of Month 1 GMV generated by buyers in
Month 12

––

Concentration: percentage of revenue generated by top 20% buyers

––

Cross pollination (if applicable): percentage of buyers whose second purchase is
in a different category

––

NPS

4. Measuring Liquidity
The first marketplace in a category to reach liquidity wins (it’s winner takes all: we’ll discuss
this further in Chapter 7, Working with Investors). Marketplaces strengthen with scale and
scale comes from liquidity. Until you reach liquidity, you’re vulnerable. After, you have the
opportunity for dominance.

	 |
48
Marketplace Metrics

Liquidity is the reasonable expectation of selling something you list or finding what you’re
looking for. Liquidity indicates how successful the marketplace is for sellers and buyers.
How long does it take to make a transaction? What percentage of goods or services are
purchased daily?
So how do you measure liquidity? Consider two numbers: supply liquidity and buyer liquidity.
•

Supply Liquidity (measured as a percentage): Sangeet Paul Choudary defines
supply liquidity as “the percentage of listings that lead to transactions within a
certain time period.” Sharetribe’s Marketplace Academy outlined some examples
for how supply liquidity can be calculated: On Etsy, it is the proportion of total
stock sold in a given time period (probably one month). On Airbnb, it’s measured
daily: how large a proportion of rooms are booked each night. And on Uber, you
might need to measure it hourly: how big a percentage of drivers that are on duty
are driving a customer at any given hour (how many logged-on drivers in a given
time period get rides).
Note: The percentage of listings that lead to transactions within a certain time
period serves as a proxy for the efficiency of the marketplace. Merely increasing
the number of buyer and seller sign-ups doesn’t serve a purpose unless this metric
starts rising.

•

Customer Liquidity: This is the probability of a visit leading to a transaction.
Simon Rothman believes that a conversion range of 30-60% is a reasonable
expectation (not counting people who bounce). You can calculate how many visits
you get in a given period and how many transactions you get in the same period, i.e.
percentage of visiting customers who buy something. Keep in mind that the higher
the average selling price of products on your marketplace, the lower the conversion
rate may be.

6
MARKETPLACE TOOLS
As you’re building your marketplace, consider which tools and technology to leverage. There’s
no reason to reinvent the wheel as new infrastructure solutions are emerging to help facilitate
transactions on your platform, including: payment software, background checks, shipping and
fulfillment solutions, and data privacy.
With help from the Marketplaces Facebook Group, we have curated a list of recommended
B2B tools by marketplace founders.

	 |
50
Marketplace Tools

You’ll see that most of these tools are relevant and useful to any software startup, and not just
marketplace companies. However, a marketplace is unique in that it deals with distributed
supply (and demand) and there aren’t many marketplace-specific tools out there yet to help
with this.
With that said, some of the products on our list are designed to help third parties provide
services and sell products: Shippo (a Version One investment), ShareTribe and Near Me,
Stripe Connect and previously, Balanced (which has since shut down but most customers
have moved over to Stripe).
Here’s the bottom line: leverage good tools wherever you can, so you can focus on more
valuable tasks like making connections and refining your business model.

7
WORKING WITH INVESTORS
Two of the most frequently asked questions at any startup event or investor panel, are “What
do investors look for in a marketplace startup?” and “How to investors value a startup?” While
answering these questions is often more art than science, we’ll shed some light on how we
evaluate companies at Version One.

WHAT DO INVESTORS LOOK FOR IN A MARKETPLACE?
As mentioned earlier, Bill Gurley outlined a list of 10 factors to consider when evaluating
marketplaces. If you haven’t read his post already, we strongly encourage you to do so.
And, as we mentioned in Chapter 1, the six most important factors for us are:
––

High fragmentation

––

Regular frequency of use/purchase

––

Discovery/new buyer-seller relationships (vs. monogamy)

––

Total Available Market

––

Transactional (being part of the payment flow)

––

Ability to expand the overall market

These aren’t the only factors that lead to success. For instance, other VCs feel that a
marketplace with less frequency of purchase can be offset by a high AOV. However, for
us, these six points create the foundation of our thesis and we’re more likely to dive into a
startup’s data if we’re aligned at this higher level.

What type of traction do we want to see in order to invest?
The answer is, “it depends.” Every startup is different and we conduct our due diligence on a
case-by-case basis. Yet, with that said, there is a general set of questions we use to evaluate the
dynamics of a marketplace and assess a startup’s product-market fit.

	 |
52
Working With Investors

Using our portfolio company Headout as an example, here’s a sampling of questions we may
ask a potential marketplace startup. You’ll see that many of these questions are inspired by
the metrics/KPI data we discussed in Chapter 5.
While some of the specific details may vary based on market (Headout is a travel/experience
oriented marketplace), any founder can use this as a primer for building out his or her own
metrics. As you can imagine, there are many more questions to ask but these serve as a good
conversation starter between Version One and the entrepreneurs we meet, helping us develop
a stronger thesis faster.

On buyers (e.g. travellers/tourists):
––

How do you define engagement? What does your engagement pyramid or
conversion funnel look like (e.g. % of users who download app, browse your app,
purchase based on overall signups or relative to previous activity)?

––

How do you identify your most engaged users? Have you been able to identify
common traits of power users (e.g. perhaps a demographic breakdown)?

––

What is the average time spent on the app on the whole, and on a per-activity basis?

––

Do you have a weekly or monthly cohort analysis that you can share? What is your
DAU and/or MAU?

––

What is the average spend per user? Are users sensitive to price?

––

What is the average time to conversion? For some companies, this is the number
one KPI. A company like Uber will focus on converting within a minute of the
app opening, while a company like Airbnb wants a visitor to convert over multiple
visits (but within a few days). The target conversion time for most marketplaces
will probably fall somewhere in the middle. Hopefully, your time to conversion
improves over time as you identify ways to reduce friction in the process.

––

How many users have purchased more than once? What is the time between first
and second time of purchase? Have users purchased in more than one city?

––

What is the average time between booking and experiencing the purchase?

On suppliers (i.e. vendors):
––

What percentage of vendors’ total inventory is made available on your marketplace?

––

How often do suppliers list new inventory?

	 |
53
Working With Investors

––

What percentage of listings are purchased?

––

How long is something posted before it is purchased?

––

What is the average price point? What is the most popular price for purchases?

––

What is the average discount offered by vendors (if any)?

––

What are the common characteristics of the most successful vendors?

––

What percentage of vendors do you need to sign up (in a particular area) for
critical mass or starting liquidity?

In general:
––

To get a greater sense of product-market fit: what is the total number of
downloads to date, number of users (suppliers and buyers), etc.?

––

Are there any overlaps between suppliers and buyers? If so, what percent? (Note:
this question won’t be applicable to every marketplace)

––

What is your GMV? What is your take rate?

––

What are your sources of traffic for users?

––

What is your CAC by channel?

Wherever applicable, we ask for average values (i.e. spend, price, buyers, suppliers, time) and
absolute numbers and distributions.

OUR MARKETPLACE FUNDING NAPKIN
Together with Point Nine Capital we put together a framework that links milestones with
fundraising stages for marketplace companies. You can download a copy as a Google Sheet
or in stylized Marketplace Napkin form. Note that Point Nine has also created a great chart
for SaaS companies as well.
As always, there are exceptions to every rule. Some startups can raise a big round early on
(for example, when the founding team already has large exits under their belts), while
other startups feel the frustration of struggling to raise money even though they’re hitting
every milestone on the chart. But we hope this chart can be a useful framework for founders
to understand what benchmarks we as marketplace investors look for, and when we look
for them.

	 |
54
Working With Investors

Marketplace Funding in 2018

Source: Point Nine Capital and Version One, Available as a Google Sheet or Marketplace Napkin

	 |
55
Working With Investors

Ultimately, we invest in smart founders — ones who are incredibly passionate, ambitious,
and talented. But these founders are also data-driven, no matter how early the company is.
We want to know that an entrepreneur has a good hold on all KPIs, i.e. the knobs and levers
that he or she can turn and pull in order to engage users and scale the business. In fact, the
founders we have been most impressed with have been able to present their data quickly and
communicate insights clearly.

HOW DO INVESTORS DETERMINE VALUATION
MULTIPLES FOR MARKETPLACES?
The main multiple we like to use for marketplace businesses is GMV. Our rule of thumb is
that marketplaces at scale are valued at roughly 1x annualized GMV (typically about 6-8x
annual revenue). These are for marketplaces that are growing fast and are category leaders.
Our assumptions for this valuation:
––

Scale: > $1b GMV

––

YoY growth of at least 30%

––

Take rate of about 10-15%

Taking Etsy as an example, after it went public...
––

Consensus estimate of approximately $270M for 2015 revenue

––

$2.4B for 2015 GMV

––

Take rate of about 11%

––

YoY growth for most recent quarter (ending Q1 2015): 45%

	 |
56
Working With Investors

Source: Bessemer Venture Partners, SlideShare “Valuations: What is happening and does it matter?”
Etsy had a market cap of about $1.86B as of July 13, 2015 with a revenue multiple of 6.8 and
a GMV multiple of 0.76. But note that its GMV multiple was as high as 1.7 immediately
following its IPO.
Bessemer offers a good overview of current valuations for different business models: SaaS,
marketplaces, consumer, and e-commerce.
You can see in their chart above that marketplaces get some of the highest revenue multiples
because of their operational leverage and high defensibility at scale.
While there’s certainly a lot of enthusiasm surrounding marketplace valuations, it’s important
to realize that there’s a large (and potentially sobering) disconnect between the valuations
for public and private consumer marketplaces. This fact was recently pointed out by Mahesh

	 |
57
Working With Investors

Vellanki of Redpoint Ventures, after he took a deep look into six metrics for marketplaces,
including enterprise value, revenue multiples, and revenue growth. His conclusion:
“What I see are large scale, profitable and fast growing marketplace businesses in
exciting markets that are worth shockingly little relative to private comps. I’m
both excited and scared to see how things will shake out.”

SO WHAT ABOUT EARLY-STAGE MARKETPLACES?
These calculations apply to marketplaces that have already solved the chicken-andegg problem, reached liquidity, and have become a category leader. But you might be
wondering how to value your early-stage marketplace startup.
Early-stage companies are valued very differently. In this case, metrics don’t count- we’re
evaluating the team, idea, and vision. Then as the marketplace starts to scale fast, the
multiples are often very high because growth is high as well. But you need to understand
that your marketplace will ultimately be valued at 1x GMV.
If you’re a founder of an early-stage marketplace, you should focus on two things:
1. Growing GMV
2. Proving out take-rate
Sometimes we see entrepreneurs who pitch impressive GMV numbers, but haven’t proven
that they can ultimately get to a significant take rate. For example, a marketplace that
generates leads instead of being part of the transaction might have a take rate as low as
2-3%. In this case, the GMV needs to be 5x bigger than a comparable marketplace with a
10-15% take rate.

The bottom line? Don’t wait too long to prove out your monetization.

8
MARKETPLACE EXITS

It’s typically a binary outcome for marketplace start-ups at the beginning: either you
figure out the chicken-and-egg problem or you don’t. But, marketplaces can face a binary
outcome at the other end too: either you scale your marketplace into a large, stand-alone
company that can go public, or you find yourself with few other exit options.
This reality is quite different than the exit opportunities for SaaS companies. It’s common
for SaaS start-ups to be bought by large enterprises for the technology/product. The
acquiring company can just plug that product into their existing distribution channel to
help round out their current product suite and grab more market share.
However, in the case of marketplaces, it’s much harder to leverage existing distribution or
find synergies between the acquirer and acquired company. A marketplace’s technology/
product is often custom and proprietary, making it tough to replace with existing
technology and save costs.
It’s also hard to transition buyers and sellers over to the acquirer’s marketplace, as Boris
has seen first hand. When Amazon acquired Bibliofind (a competitor to his company
AbeBooks) in 1999, they expected to switch all of the Bibliofind sellers over to Amazon.
Instead, most of the sellers left the platform and signed up with AbeBooks, giving AbeBooks
a huge boost.
So, if an acquiring company cannot fundamentally change the economics of the
marketplace they acquire and if it’s tough to switch over the new buyers and sellers, that
leaves little incentive to acquire a marketplace.

	 |
59
Marketplace Exits

In fact, the only reason you’d ever acquire a marketplace is if it were a large, standalone
unit that could grow on its own. eBay was relatively successful when it acquired
marketplaces like Rent.com (2005) and StubHub (2007). But, the universe of eBay-like
buyers is limited, as is the number of marketplaces that build platforms with several
billions of GMV.
This means that an IPO is the exit option for the most successful marketplaces. Over the
past few years, we’ve seen this with Alibaba, Lending Club, and Etsy. And, we will most
certainly see this with Airbnb, Uber, and Lyft.
In short, it’s a binary outcome for most marketplace start-ups at both the beginning and
end. Be sure to set your expectations and business model accordingly.

CONCLUSION
Marketplaces are tough to build. But once they reach liquidity, they can be
even tougher to kill.
Some of the most truly revolutionary companies born over the past decades have been
marketplace companies, and we believe there’s a lot of room for new entrants. Some of the
world’s next billion dollar unicorns will be marketplace startups.
Buying and selling is an integral part of daily life, and there’s still great opportunity for
innovation in new verticals, models, and monetization strategies.
We also see how online marketplaces are changing participants’ lives. For example, Airbnb
created a whole new consumer behavior and way of travel… and 72% of hosts in NYC
depend on their Airbnb earnings to pay their rent or mortgage. Etsy enables crafters and
hobbyists to make money doing what they love: 35% of Etsy sellers started their business
without much capital investment (compared with 21% for small business owners in
general).
One third of Version One portfolio companies are marketplace companies, and with each
company, we’re learning more about marketplace success strategies and how to overcome
the challenges associated with liquidity and scaling. We’re excited to be active investors in
this space and are looking forward to the journey ahead.

RESOURCES
Chapter 1
“Alibaba is just the beginning: How B2B will thrive (for real, this time).” Boris Wertz,
GigaOm. June 30, 2013. https://gigaom.com/2013/06/30/alibaba-is-just-the-beginning-howb2b-marketplaces/
“All Markets are Not Created Equal: 10 Factors to Consider When Evaluating Digital
Marketplaces.” Bill Gurley, Above the Crowd. Nov. 13, 2012. http://abovethecrowd.
com/2012/11/13/all-markets-are-not-created-equal-10-factors-to-consider-when-evaluatingdigital-marketplaces/
“How to miss by a mile: An alternative look at Uber’s potential market size.” Bill Gurley,
Above the Crowd. July 11, 2014. http://abovethecrowd.com/2014/07/11/how-to-miss-by-amile-an-alternative-look-at-ubers-potential-market-size/
“Undoing incumbent marketplaces: Lowering the take rate.” Albert Wegner, Continuations.
January 26, 2015. http://continuations.com/post/109215849455/undoing-incumbentmarketplaces-lowering-the-take
“As vertical markets rise, Craigslist faces its demise.” Jeremy Levine, TechCrunch. April 10,
2015. http://techcrunch.com/2015/04/10/as-vertical-marketplaces-rise-craigslist-faces-itsdemise/
“How to build an enduring, multi-billion dollar business.” Sarah Tavel, Medium. February
15, 2016. https://medium.com/@sarahtavel/how-to-build-an-enduring-multi-billion-dollarbusiness-hint-create-a-10x-product-recast-3527df2b8fcb
“The spawn of Craigslist.” Andrew Parker, The Gong Show. January 21, 2010. http://
thegongshow.tumblr.com/post/345941486/the-spawn-of-craigslist-like-most-vcs-that-focus
“The Magic of Liquidity: Web Marketplaces Still Have A Long Way To Go.” David Haber,
TechCrunch. December 1, 2012. https://techcrunch.com/2012/12/01/the-future-of-onlinemarketplaces/

	 |
62
Resources

“Zero to One in Data Science: The Tophatter Story.” Angela Tran Kingyens, Version One Blog.
August 3, 2017. http://versionone.vc/data-zero-one/#ixzz53mLO26tO

Chapter 2
“5 tips for building a two-sided marketplace.” Boris Wertz, Forbes. Feb. 7, 2013. http://www.
forbes.com/sites/ciocentral/2013/02/07/5-tips-for-building-a-two-sided-online-marketplace/
“Airbnb: The growth story you didn’t know.” Growth Hackers Case Study.
https://growthhackers.com/companies/airbnb/
“Cross Network Utility and Networks.” Fred Wilson, AVC.com. September 16, 2012.
http://avc.com/2012/09/cross-network-utility-and-networks/
“Do things that don’t scale.” Paul Graham, Paul Graham,com. July 2013. http://paulgraham.
com/ds.html
“Airbnb CEO Brian Chesky on building a company and starting a sharing revolution.”
Derek Thompson, The Atlantic. August 13, 2013. http://www.theatlantic.com/business/
archive/2013/08/airbnb-ceo-brian-chesky-on-building-a-company-and-starting-a-sharingrevolution/278635/
“How modern marketplaces like Uber and Airbnb build trust to achieve liquidity.” Anand
Iyer. First Round Capital. http://firstround.com/review/How-Modern-Marketplaces-LikeUber-Airbnb-Build-Trust-to-Hit-Liquidity/
“Is the freemium model the future of marketplaces?” Boris Wertz, Version One Blog, March
31, 2015. http://versionone.vc/freemium-model-future-marketplaces/
“Airbnb blooms an entire ecosystem of startups.” Jeremiah Owyang. April 21, 2014. http://
www.web-strategist.com/blog/2014/04/21/airbnb-blooms-an-entire-ecosystem-of-startups/
“Marketplaces, Rating Systems and Leakage.” Charles Hudson, Charles Hudson’s Blog.
February 4, 2014. http://www.charleshudson.net/marketplaces-rating-systems-and-leakage
Etsy, 2016 Annual Report. https://investors.etsy.com/~/media/Files/E/Etsy-IR/annual-reportproxy-materials/etsy-ar2016.pdf

	 |
63
Resources

“Why India’s Flipkart abandoned its mobile website.” Rajiv Rao, ZDNet. March 27, 2015. http://
www.zdnet.com/article/why-indias-flipkart-abandoned-its-mobile-website/

Chapter 3
Etsy, 2016 Annual Report. https://investors.etsy.com/~/media/Files/E/Etsy-IR/annual-reportproxy-materials/etsy-ar2016.pdf
“A rake too far: optimal platform pricing strategy.” Bill Gurley. Above the Crowd. April 18, 2013.
http://abovethecrowd.com/2013/04/18/a-rake-too-far-optimal-platformpricing-strategy/

Chapter 4
“The Evolution of Managed Marketplaces.” Anand Iyer, Medium. June 12, 2017. https://medium.
com/@ai/the-evolution-of-managed-marketplaces-3382290963b2
“Anatomy of a managed marketplace.” Ezra Galston. TechCrunch. May 25, 2017. https://
techcrunch.com/2017/05/25/anatomy-of-a-managed-marketplace/
“Come for the tool, stay for the network.” Chris Dixon. CDIXON Blog. January 31, 2015.
http://cdixon.org/2015/01/31/come-for-the-tool-stay-for-the-network/
“An overview of the Growing “SaaS Enabled Marketplace” Ecosystem.” Point Nine Capital,
Medium. October 11, 2016. https://medium.com/point-nine-news/an-overview-of-the-growingsaas-enabled-marketplace-ecosystem-b4f5314356e5
“Free SaaS-enabled marketplaces — a novel go-to-market for software startups.” Tomasz
Tunguz. http://tomtunguz.com/software-enabled-marketplaces/

Chapter 5
“A KPI Dashboard for early-stage SaaS startups.” Christoph Janz, Point Nine Capital. April 11,
2013. http://christophjanz.blogspot.com/2013/04/a-kpi-dashboard-for-early-stage-saas.html
Version One Marketplace KPI Spreadsheet. https://docs.google.com/spreadsheets/d/1K5E1Nzjq4
KEvGVv6EysGswTuQm5u21Sa2ugSZewrZqE/edit#gid=1569059121

	 |
64
Resources

“KPIs & Dashboards: Must-haves for Early Stage Startups.” Vijay Nagappan, Batting399.com.
August 17, 2015. http://www.batting399.com/post/126915896445/kpis-dashboards-must-havesfor-early-stage
“How to win with marketplaces: The three success factors.” Sangeet Paul Choudary,
TheNextWeb. May 4, 2013. https://thenextweb.com/entrepreneur/2013/05/04/how-to-win-withmarketplaces-the-three-success-factors/
“How to measure your success: The key marketplace metrics.” Juho Makkonen, Marketplace
Academy. June 9, 2016. https://www.sharetribe.com/academy/measure-your-success-keymarketplace-metrics/
“How To Structure A Marketplace.” Simon Rothman, TechCrunch. August 19, 2012. https://
techcrunch.com/2012/08/19/how-to-structure-a-marketplace/

Chapter 7
“All Markets are Not Created Equal: 10 Factors to Consider When Evaluating Digital
Marketplaces.” Bill Gurley, Above the Crowd. Nov. 13, 2012. http://abovethecrowd.
com/2012/11/13/all-markets-are-not-created-equal-10-factors-to-consider-when-evaluatingdigital-marketplaces/
“Marketplace Funding in 2016.” Version One and Point Nine Capital. Spreadsheet: https://docs.
google.com/spreadsheets/d/1DaRrcgYXKJIzjgTK_rgGfaRLIU3LTdVTa3kfMEmCCK8/edit#gid=0
Image: http://versionone.vc/wp-content/uploads/2016/07/Marketplace-Napkin.pdf
“Building your user engagement pyramid.” Version One. December 15, 2014. http://versionone.
vc/engagement-pyramid/
“Valuations: What is happening and does it matter.” Bessemer Venture Partners, SlideShare
Feb. 2015. http://www.slideshare.net/taliagold/valuation-presentation-blog-version
“This Is How Public Consumer Marketplaces Are Valued.” Mahesh Vellanki. October 7, 2015.
http://www.mahesh-vc.com/blog/this-is-how-public-consumer-marketplaces-are-valued

Vancouver Office
boris@versionone.vc
@bwertz

San Francisco Office
angela@versionone.vc
@atkingyens

Pub Date: March 2018. ©2018 Version One Ventures. All rights reserved. All trademarks are the property
of their respective owners.



---

Source Exif Data:

File Type                       : PDF
File Type Extension             : pdf
MIME Type                       : application/pdf
PDF Version                     : 1.5
Linearized                      : No
Language                        : en-US
XMP Toolkit                     : Adobe XMP Core 5.6-c137 79.159768, 2016/08/11-13:24:42
Format                          : application/pdf
Title                           : A Guide to Marketplaces
Creator                         : Version One
Create Date                     : 2018:02:28 20:10:17-05:00
Metadata Date                   : 2018:02:28 20:10:22-05:00
Modify Date                     : 2018:02:28 20:10:22-05:00
Creator Tool                    : Adobe InDesign CC 2017 (Macintosh)
Instance ID                     : uuid:c73f5a3b-054f-8d4e-9318-dbc9eda0ada3
Original Document ID            : xmp.did:ED7F1174072068119457A2EF1D94667E
Document ID                     : xmp.id:5852dd68-2eee-436e-9bc1-b8a1d7e6f871
Rendition Class                 : proof:pdf
Derived From Instance ID        : xmp.iid:6a2d3306-7942-4699-b6ba-4754ca53fa03
Derived From Document ID        : xmp.did:08d774b3-cd47-400a-9785-56b3327a55fc
Derived From Original Document ID: xmp.did:ED7F1174072068119457A2EF1D94667E
Derived From Rendition Class    : default
History Action                  : converted
History Parameters              : from application/x-indesign to application/pdf
History Software Agent          : Adobe InDesign CC 2017 (Macintosh)
History Changed                 : /
History When                    : 2018:02:28 20:10:17-05:00
Producer                        : Adobe PDF Library 15.0
Trapped                         : False
Page Count                      : 65
Author                          : Version One

EXIF Metadata provided by EXIF.tools