A marketplace can grow gross transaction volume 40% in a quarter and still miss its revenue target.
Volume and revenue aren’t the same number, and a lot of marketplace models quietly treat them like they are. GMV is what flows through the platform, and the company only keeps a slice of it. That gap is where most marketplace forecasting goes wrong.
GMV is not revenue, and the model needs to say so explicitly
The take rate, the percentage of GMV a marketplace keeps as revenue, sounds like a simple multiplier until you look at what moves it. A premium tier for sellers can hold GMV flat while pushing revenue up. A shift toward lower-margin categories can do the opposite without any change in transaction volume at all. A model that forecasts GMV and applies one flat take rate assumption on top will miss both of these every time. The take rate needs its own forecast line, broken out by category if the business has more than one, because it rarely holds steady as a marketplace matures.
Supply and demand grow on different timelines, so model them separately
Most marketplace founders know they’re running two businesses at once, one for supply and one for demand. Fewer build separate acquisition models for each side. Onboarding a new seller often means verification and a listing review, which makes supply slower and pricier to acquire than demand, something a marketing campaign can sometimes win in days. A single blended CAC number hides which side is the real constraint on growth. Build supply acquisition and demand acquisition as two separate models that feed into the same GMV forecast.
Incentives and subsidies belong in cost of revenue, not marketing
Driver bonuses and first-order discounts used to bootstrap liquidity in a new market are real costs. They tend to get buried inside a marketing budget, where they distort every other metric sitting next to them. These costs behave more like a cost of revenue than a growth expense, because they scale with transaction volume instead of a campaign calendar. A marketplace still subsidizing both sides to build liquidity in a new city needs a model showing what revenue and margin look like once those subsidies phase out, not just while they’re running. Boards ask this question earlier than founders expect, and a model without an answer reads as a company that hasn’t thought about its own path to profitability.
Liquidity and matching rate drive conversion more than either side’s growth alone
A marketplace with plenty of supply and plenty of demand can still convert poorly if the two don’t line up in the same city or the same price range. Matching rate and time-to-fill are leading indicators, and they show up in the model well before they show up as a declining conversion rate on the P&L. A home services marketplace with twice the providers it needs in one city and half of what it needs in another isn’t healthy just because the totals look fine on a dashboard. Segmenting supply and demand by geography or category, even roughly, surfaces this kind of imbalance before it turns into churn.
Concentration risk shows up on both sides, not just in customer revenue
SaaS companies watch for revenue concentration in a handful of large accounts. Marketplaces have to watch it on both sides. A handful of sellers driving most of a marketplace’s GMV is a real risk. So is a small group of high-frequency buyers propping up transaction volume that would otherwise look a lot smaller. Either one is worth a line in the model and a conversation with the board before an investor finds it first during diligence.
Marketplace models take more structure than a single-sided SaaS model because there are two of nearly everything to track. A marketplace that skips that split usually gets surprised in the same way: GMV keeps climbing while one side quietly falls behind the other, and nobody notices until a board member asks why growth slowed in a city that looked fine on the dashboard.