A Comparison of Take Rates: Drivers of Value
Every investor and entrepreneur knows that pricing is one of the most important factors that can make or break a business. If it’s too high, you trade off growth (sometimes). If it’s too low, you trade off profitability. For marketplace businesses, the pricing equivalent is the take rate — the percentage that the marketplace keeps for every dollar that transacts on the platform. Also referred in other contexts as a “commission”, “fee”, “rake”, and more.
Having evaluated many marketplace businesses as an investor at both FJ Labs (VC firm that specializes in marketplaces) and Carlyle (PE firm that invests in a variety of large-cap businesses), as well as having spent a significant portion of my time at HBS founding my own marketplace business, I’ve spent uncountable hours thinking about take rates. In particular, I want to address the question — why are some marketplaces able to charge a 20%+ take rate while others can only 2%?
Take rate comps
Let’s look at the take rates across some of the most successful marketplaces sorted by magnitude. Right away, you’ll see it’s a huge range, from 0% to 85%.
Note: If you have any companies that should be added to this list, or if you have any updates to these numbers, please let me know. There is a lack of B2B companies listed, as I haven’t come across any where that information is public, and additionally, many B2B marketplaces charge as a subscription model instead of on a transaction basis (which in itself could be another blog post!).
Right away, you’ll notice patterns in the data. Here are my thoughts on why some marketplaces are only able to charge 0–5% whereas others are able to charge 20%+ (and with the photography marketplaces upwards of 70–80%!).
1) The next best alternative for sellers / buyers is extremely inconvenient
Willy Braun says it best here: “Take rate is a good indicator of the power of negotiation: a high take rate usually means that the marketplace proposition value is high and somehow exclusive in a given segment.” This point is incredibly important. Like any other business, pricing is a negotiation, and as any good negotiator knows, your leverage in this negotiation depends on your opponent’s BATNA (Best Alternative to a Negotiated Agreement).
In other words, what your customers are willing to pay depends on how good or bad their alternative is to not using your service. If the alternatives are direct competitors and the category has winner-take-all dynamics, this could result in a pricing war (more on this below). If not a winner-take-all category, the pricing will most likely converge for all competitors, effectively settling on an implicitly agreed upon price (see the crowdfunding space above as an example — IndieGogo and Kickstarter). Regardless, the cheaper or more convenient the alternatives are, the lower the willingness-to-pay will be for the marketplace in question.
Take Opentable and Stubhub. Both operate in highly fragmented industries, and do not have unique inventory (note: Stubhub does offer some managed services such as payment handling and transfer of the ticket, but this is light compared to other true managed marketplaces, and is offset by Opentable’s software solution that is provided to restaurants). Yet, why is Opentable only able to command a ~2% take rate while Stubhub is able to charge 22%?
If you think about the BATNA, Stubhub’s business model clearly offers a much stronger value proposition than Opentable’s. If you are a restaurant, your BATNA to Opentable is to put a phone number on your website and tell your customers to call you. It adds some friction to gaining new customers and some additional work on the backend, but it gets the job done and is not a bad alternative. On the other hand, if you are a ticket scalper, your BATNA to Stubhub is to stand outside Madison Square Garden and hope that someone who walks by is interested in buying your ticket to the Knicks game. Tedious, time-consuming, and likely low chances of getting a successful sale of your ticket. Thus, ticket scalpers are willing to pay a high fee for the benefit of quickly getting connected to many potential buyers.
2) The category is highly fragmented
For marketplace businesses, the more fragmentation in the category, the better. This ties into the above, as higher fragmentation usually means that the BATNA is incredibly inconvenient. The customer’s alternative to the marketplace would mean sorting through hundreds (or, in some cases, thousands) of options on his or her own. No one wants that!
An illustration of this is the difference in the take rates for hotels as compared to flights booked through OTAs (online travel agency) such as Priceline, Expedia and Orbitz. For hotels, a typical take rate for an OTA is ~15%, with some boutiques paying more than double that. For airlines, a typical take rate for an OTA is ~3%. This makes a lot of sense — the airline industry is highly concentrated, and if you want to book a flight, even if Priceline didn’t exist, you could search 3–5 airline websites on your own and get the same results. For hotels, however, the landscape is very fragmented. As a consumer, on your own you may be able to find the major hotel chains in a new city you’re visiting, but finding all the boutiques would be a longshot. And thus, for hotels, the benefit of being included on an OTA that aggregates everything as a “one-stop-shop” resource for consumers presents a huge discovery benefit, which is reflected in the take rates.
As another example, the highest take rates by far are those in the photography category. The users and contributors of creative content is probably one of the most fragmented industries on both sides. In fact, Shutterstock states that it has 1.8 million users licensing content and 350,000+ contributors providing content. For both the demand- and supply-side, it would be incredibly difficult to search through the opposite side on their own without an aggregator.
Finally, it is important to note that network effects are much stronger in fragmented categories. Once a marketplace has enough supply and demand to get the flywheel going, it becomes very hard for any other alternative to compete, as it is hard to replicate the same density of supply / demand. This is made even stronger if the marketplace can obtain unique inventory that is not listed anywhere else.
3) The addition of value added services which creates stickiness with its users (aka managed marketplaces)
A managed marketplace is a marketplace that provides an additional value-added service beyond just connecting buyers and sellers. A value-added service could be anywhere from trust (e.g., the background checks on drivers that Uber / Lyft provides) to payment processing to holding and delivering inventory. Taking a look at the table above, most of the high take rate companies are managed marketplaces. Managed services usually create a superior customer experience, leading to greater word of mouth as well as greater stickiness. Managed marketplaces which also use the consignment model (where inventory is taken into possession) also have the additional advantage of obtaining exclusive inventory. This creates even more value for the buyers.
For example, take the RealReal and Poshmark. Both are trying to solve the same problem — many people have clothes and accessories sitting in their closet collecting dust which they would rather sell for cash. However, the two businesses have taken very different operational approaches. Poshmark is a true marketplace that does not have any managed services. Sellers upload images on their own, create their own listings, and ship the items on their own. For this, sellers pay 20%. The RealReal, on the other hand, takes a fully managed approach. They have a consignment model, where they take the product from the seller, authenticate it, photograph it, list it and then send it to the buyer after the purchase has been made. For the seller, this is truly an “out of sight, out of mind” experience. For the RealReal’s efforts, sellers pay 30–50% depending on the value of the item. This is ~2x what Poshmark charges on average, yet sellers are still flocking to the RealReal given the incredible convenience and trust its services provide.
Of course, more managed services means higher operating costs and it is always important to consider fully-loaded unit economics. The good news is that when you provide many managed services, it is usually easier to raise pricing over time which is typically very hard for any other type of marketplace to do.
Other considerations
There are a number of other considerations as well which will impact take rate, but the underlying key drivers for these considerations still tie into the top three factors listed above. For example, the likelihood of disintermediation will impact take rates — the easier it is to disintermediate, the less likely customers will be willing to pay for a marketplace service. However, likelihood of disintermediation usually can be determined by the level of fragmentation (if it is a highly concentrated market, once a connection is made, there is less need to have to go through the marketplace every time to work with the same provider) and the amount of value-added services (as long as the services provided are truly value-add!).
Do we always want high take rates?
Bill Gurley makes the point in his famous post that the highest take rates are not always the best. It’s true — there are certain situations where lower take rates are actually better, such as if it’s a winner-takes-all category. In these cases, you need to get as much supply and demand on your platform as fast as possible, meaning it may make sense to reduce friction by lowering take rates. For example, Opentable needs network effects to provide value to both sides of the market and operates in a business category where the product is very sticky and restaurants only use one provider for such a service. Thus, it was vital they captured as much of the market as quickly as they could to prevent a second player from gaining scale, incentivizing them to charge a relatively low take rate. However, you’ll often see that these marketplaces will slowly raise their take rates after they’ve won the category in an attempt to reach profitability in the long-run. At this point, the network effects are so strong that it is hard for the supplier or buyer to switch to another cheaper alternative.
It is also important to note that these cases are very rare — it is hard to start with a low take rate and then raise it without losing customers. Oftentimes, this only works if it is truly a winner-take-all situation and you’ve already taken all or if it is a highly sticky product (and even then, if the price differential is big enough, your customers may still switch to a competitor). Furthermore, if you are worried your take rate is too high and that a competitor can underprice you to win away business, you’re likely in a situation where pricing quickly races to the bottom, which is challenging for any young business to survive in.
Conclusion
Knowing what take rate a business can support is one of the many challenges to running a marketplace business and it can oftentimes feel quite opaque. This analysis is meant to guide these decisions through reflecting on where a company stands in regards to the above key factors. Also important to remember is that pricing a take rate is also impacted by what other competitors are charging and the goals of a business depending on the stage of life (e.g., to gain early adopters, to build to general adoption or to scale and sustain).
Should you have any thoughts on other additional factors that I may be missing, please send me a message or leave a comment! I’d love to hear your thoughts.
Note: Special thanks to Daniel Simon, Will Nichols and Vivian Graves for their thoughts and contributions to the making of this blog post.
