Pattern · Strategy & competition
The Marketplace Take Rate
Also known as Commission marketplace, Transaction fee model, Brokerage. Anchored to No single originator. The choice between operating a marketplace and acting as a reseller was formalised by Andrei Hagiu and Julian Wright., 2015.
Where this is contested
Brokerage is ancient and commission has been charged on other people's transactions for as long as there have been merchants. What is modern, and what makes this a distinct pattern rather than a version of the multi-sided platform, is the deliberate choice between taking a percentage of a transaction you do not own and buying the goods to resell them. Hagiu and Wright set out the conditions determining which is correct, and that decision is what this entry is about. The pattern is dated to their analysis rather than to the practice.
Taking a percentage of transactions between two parties you never own the goods for, which produces beautiful margins and no inventory, and leaves the whole business resting on whether the two parties keep coming back through you rather than going round you.
Plate · The shape
How it works
A marketplace does not buy or sell. It matches, it holds the payment, it enforces the rules, and it keeps a percentage. Because it never owns inventory, gross margin approaches the take rate itself and the balance sheet stays light, which is why the model is so admired and so often attempted.
The decision that defines it is control. Hagiu and Wright's analysis turns on who holds the information needed to make the transaction good. Where the seller knows things about the product or the customer that the intermediary cannot know, letting the seller keep control and taking a commission produces better outcomes. Where the intermediary knows more, or where consistency across sellers matters more than any individual seller's judgement, owning the goods and reselling them is better. Most businesses that fail at this chose the model by aspiration rather than by that test.
The take rate itself is the whole commercial question. Too low and there is no business. Too high and both sides start calculating what it would cost to deal directly, which they can do at any time, because the marketplace has by construction introduced them to each other.
The parts of the model
Supply
The sellers, providers or hosts. Usually the harder side to build and the side with the most to gain from leaving once a relationship is established.
Signals
Supply quality varies and the marketplace is judged on the worst of it · The largest suppliers ask for a lower rate and have the leverage to get it · Suppliers list on competing marketplaces as a matter of routine
Demand
The buyers. Attracted by choice, price and the confidence that a transaction with a stranger will be all right.
Signals
Buyers arrive for the selection and stay for the safety · Repeat buyers begin to develop preferences for specific suppliers · Acquisition cost per buyer is the largest marketing line in the business
The transaction
The event the fee attaches to, and therefore the thing that must happen on the platform rather than beside it. Payment handling, escrow and dispute resolution exist to make sure it does.
Signals
Payment flows through the marketplace rather than around it · There is a real service attached to transacting on-platform, not merely a rule against leaving · The proportion of introductions that convert on-platform is measured
The rate
The percentage taken, which is the entire revenue model and the number both sides think about most. Rarely uniform in a mature marketplace, and usually tiered by volume, category or service level.
Signals
The rate is defensible in a sentence to the side that pays it · Large suppliers are on different terms and everybody knows · Nobody has raised it in two years, or somebody raises it every year, and both are tells
How you know you are in it
You do not fill a pattern in. You recognise yourself in it, or you do not. Read these as a list about your own business rather than as a definition.
- Your revenue is a percentage of somebody else's revenue, and you have no control over the price they charge.
- Your largest suppliers account for a disproportionate share of transactions and have begun asking about their rate.
- You can see repeat pairings in the data: the same buyer and the same seller, transaction after transaction, and you know some of those relationships have already moved off the platform.
- Growth conversations default to raising the rate, because it is the only lever that does not require more customers.
- You have a category that looks liquid in the totals and is dead in most individual postcodes or specialisms.
- Somebody has proposed buying stock to guarantee availability, which is the point at which you are deciding whether you are still a marketplace.
The numbers that decide it
- Take rate against the cost of the services justifying it. Payment handling, insurance, dispute resolution and trust apparatus all cost money, and a rate charged for introduction alone is competed down faster than a rate charged for a service.
- Leakage: the proportion of matched pairs whose subsequent transactions happen off-platform. Almost nobody measures this and it is the most important number in the model. Estimate it from repeat behaviour if you cannot see it directly.
- Liquidity by segment, meaning the probability that a given listing transacts within a given period, measured within each category and geography rather than across the whole marketplace.
- Supplier concentration. Where a small number of suppliers carry most of the volume, the take rate is not yours to set, and the negotiation is coming whether or not it has started.
- Contribution per transaction after payment costs, support and the cost of resolving the small proportion that go wrong, which is invariably higher than the average suggests because disputes are expensive and clustered.
When this shape works
- Sellers hold information the marketplace cannot: local knowledge, bespoke judgement, or product detail that varies per item. That asymmetry is what makes commission the right structure rather than resale.
- Transactions are frequent enough to build habit, but not so relational that each pair settles into a long private arrangement.
- The marketplace supplies something genuinely valuable at the moment of transacting: payment protection, verified reviews, insurance, dispute resolution, guaranteed standards.
- Fragmentation on both sides, so no single supplier or buyer accumulates the leverage to renegotiate the rate.
And when it doesn't
- The relationship is naturally long-term and high-value, in which case the second transaction has every reason to happen off-platform and the marketplace is paid once for a relationship that lasts years.
- Consistency matters more than choice, which is the condition under which Hagiu and Wright's analysis says to buy the goods and resell them instead.
- One side is concentrated. A marketplace whose top ten suppliers carry half the volume does not set its own take rate, whatever the terms say.
- There is no service attached to transacting on-platform, so the fee is a toll on an introduction and is treated as one.
How this pattern dies
Leakage
The defining failure. The best pairings, the ones that transact repeatedly and profitably, are precisely the ones with the most to gain from dealing directly. The marketplace ends up carrying the cost of matching everybody and earning from only the low-value, one-off end of its own network.
Take-rate compression
Competition arrives, suppliers multi-home, the largest ones negotiate, and the rate falls a point at a time. The business responds by adding services to justify the rate, which raises costs, which requires the rate the market will no longer bear.
Liquidity that exists only in the totals
The marketplace reports healthy aggregate activity and is dead in every individual segment, because liquidity is local. Buyers in a category with three listings do not experience a large marketplace; they experience three listings. Growth spent on breadth rather than depth makes this worse.
Drifting into resale
To fix availability or quality, the marketplace starts buying stock, guaranteeing delivery or employing the suppliers. Each step is sensible and the cumulative effect is a business with inventory, working capital and operational risk that is still valued and managed as though it were asset-light.
Competing with your own suppliers
The marketplace uses its transaction data to launch its own competing offer in the best categories. It is immediately profitable, it is the fastest way to lose supplier trust, and it is now attracting regulatory attention in several jurisdictions.
In the wild
Etsy
Take rate on handmade and vintage goods, with a long public record of rate rises and the seller reaction to each. Useful because the compression and resentment dynamics have played out in view rather than behind closed doors.
Checkatrade and its peers
UK, and an honest illustration of the leakage problem. The introduction between a householder and a tradesman is genuinely valuable once. The second job goes direct, which is why these businesses charge suppliers a subscription rather than a per-job commission, having discovered that the transaction is not the thing they can hold onto.
The livestock auction
Owner-managed, UK, and centuries old. Commission on the hammer price, no ownership of the animals, and a business built entirely on both sides trusting the ring. It has survived because the auction supplies price discovery and settlement, not merely an introduction.
The independent insurance broker
Owner-managed, UK. Commission on premiums placed with insurers, with the broker holding the client relationship and the advice. Notably, this version has survived direct-to-consumer competition wherever the broker's judgement is genuinely worth the commission and been hollowed out wherever it was not.
The owner-managed version
Two things are worth knowing before an owner-managed business goes near this.
The first is that percentage revenue on somebody else's price is a weaker position than it looks. You cannot raise your price without raising theirs, you carry the reputational risk for work you did not do, and your income moves with a market you do not control. Brokers and agents have made good livings this way for centuries, but they have done it by being genuinely necessary rather than by being in the middle.
The second is leakage, and I would put it at the top of the page. If you introduce a customer to a supplier and they get on well, the second job goes direct. That is not dishonesty; it is two sensible people saving money. Any small marketplace whose only product is the introduction will be paid once per relationship and will spend the rest of its life finding new relationships, which is an expensive way to run a business.
What that points to is a design question rather than a policing question. What can you supply at the moment of the transaction that neither party wants to do without? Payment security, a guarantee, vetting, insurance, dispute resolution, a standard they can point to. In a small UK market the honest answer is often a subscription rather than a commission, which is exactly the conclusion the trade directories reached after years of trying to hold onto transactions they could not see.
And if you find yourself buying stock to fix availability, stop and say out loud that you are becoming a reseller. It may well be the right move. It is a completely different business, with working capital and inventory risk attached, and it should be decided rather than arrived at.
Changing out of this shape
Choose between marketplace and reseller on the information test rather than on the balance sheet: where the seller knows things you cannot, take a commission and leave them in control; where consistency matters more than individual judgement, buy the goods. Then build the on-platform service before setting the rate, since a rate charged for introduction alone has no defence. Measure leakage from the first quarter, using repeat-pair behaviour as the proxy if direct observation is impossible, and treat a rising leakage rate as the signal to change what you supply rather than to tighten the rules. Where leakage proves structural, as it does in most high-value relational categories, the usual and correct escape is to move the supplier side to a subscription and stop pretending the transaction can be held.
Where to go from here
Frameworks that work inside this pattern
- Porter's Five Forces
A structural analysis of the five competitive forces that determine an industry's long-run profitability, and therefore where and how a business can defend or improve its position.
- Unit Economics
Tests whether a business makes money on each unit sold and each customer acquired, tracing revenue through variable costs to contribution and comparing the cost of acquiring a customer with the lifetime value they return. If the unit loses money, scale multiplies the loss.
- Business Model Canvas
A one-page visual template that describes how an organisation creates, delivers and captures value across nine building blocks, from customer segments through to cost structure, so an entire business model can be seen, questioned and redesigned in a single view.
- Value-Based Pricing
A pricing discipline that anchors price to the economic value an offer creates for a defined segment, with cost setting only the floor, and that tests willingness to pay before launch using tools such as Van Westendorp's Price Sensitivity Meter and Good-Better-Best tiering.
- Value Chain Analysis
A disaggregation of the firm into nine strategically relevant activities, five primary and four support, to locate where cost is incurred and differentiation is created, on the premise that advantage lives in activities, not in the firm as a whole.
- Stakeholder Mapping (Power-Interest Grid)
A two-by-two grid that positions each stakeholder by their power over your work and their interest in it, then sets an engagement strategy per quadrant: manage closely, keep satisfied, keep informed, or simply monitor.
Playbooks that work the problem
- Entering a New Market
Four frameworks in the order I would actually use them: read the macro weather, judge whether the industry can pay you, name the risk you are really taking, then make a choice whose reasoning you can still defend a year later.
- Pricing with Confidence
A sequence for putting a number on your work without flinching: find the job customers actually hire you for, price the value rather than the hours, check the volume the price must sustain, then imagine the new price list failed and find out why before it does.
Neighbouring patterns
- The Multi-Sided Platform
A business that makes money by bringing two groups together who each need the other, where the decisive question is not what to charge but which side to charge, because the side you subsidise is the side that makes the whole thing work.
- The Long Tail
Selling small quantities of a very large number of things, on the argument that when holding and shipping stock costs almost nothing, the combined demand for everything unpopular can exceed the demand for the few things that are popular.
- Subscription and Recurring Revenue
Charging a regular fee for continuing access rather than a one-off price for a transaction, which converts selling from an event into a relationship and converts profit from something you earn once into something you have to keep deserving.
What the critics say
The choice between marketplace and reseller is usually made on aspiration rather than analysis. Hagiu and Wright show it should turn on where the relevant information sits: give control to whichever party holds the knowledge that makes the transaction good. Firms instead choose the marketplace because it is asset-light and admired, then spend years adding the operational control they gave away at the outset.
Hagiu, A. and Wright, J. (2015) 'Marketplace or Reseller?', Management Science, 61(1), pp. 184-203.
The asset-light story understates what a working marketplace actually costs. Trust, verification, dispute resolution and fraud prevention are the services the take rate is really paying for, and they are labour-intensive and scale with volume rather than being absorbed by it. Marketplaces that modelled themselves as software businesses have consistently discovered they were operations businesses.
Evans, D. S. and Schmalensee, R. (2016) Matchmakers: The New Economics of Multisided Platforms. Boston, MA: Harvard Business Review Press.
The take rate is not a free variable. Where the supply side is concentrated or multi-homes cheaply, the rate is set by the market rather than by the platform, and the network-effect argument offers no protection: a supplier who lists in three places is comparing three rates. Several marketplaces have raised rates into a competitive market and discovered their moat was narrower than the story required.
Rysman, M. (2009) 'The Economics of Two-Sided Markets', Journal of Economic Perspectives, 23(3), pp. 125-143.
Sources and further reading
- Hagiu, A. and Wright, J. (2015) 'Marketplace or Reseller?', Management Science, 61(1), pp. 184-203.
- Evans, D. S. and Schmalensee, R. (2016) Matchmakers: The New Economics of Multisided Platforms. Boston, MA: Harvard Business Review Press.
- Rochet, J.-C. and Tirole, J. (2006) 'Two-Sided Markets: A Progress Report', RAND Journal of Economics, 37(3), pp. 645-667.