What Is Marketplace Liquidity?
Turn this article into takeaways for your work.
Each assistant summarizes the article only for you and suggests best practices for your work.
Every marketplace sells one product above all others: the chance of a successful match. A buyer arrives with a need, a seller arrives with something to offer, and the marketplace is worth exactly as much as its ability to bring them together quickly. That ability has a name. It's liquidity.
The word is borrowed from finance, where a liquid asset is one you can sell fast without hurting its price. Marketplaces use it in a looser but related way. This article defines liquidity, splits it into its buyer-side and supply-side halves, lists the metrics teams use to measure it, and explains how it relates to network effects and to density. It also covers what liquidity can't tell you, because a single healthy-looking number can hide a broken market.
What marketplace liquidity means
The most quoted definition comes from James Currier of NFX: liquidity is the probability of selling something you list, or of finding something you're looking for. That wording is relayed in a Point Nine essay by Julia Morrongiello from October 2019, which also describes it as the efficiency with which a marketplace matches buyers and sellers on its platform. Andreessen Horowitz's marketplace glossary, written by Li Jin and published in February 2020, puts it as the ease with which buyers and sellers can find the right counterpart.
Notice what both definitions share. Liquidity is a probability, not a count. It isn't how many listings you have or how many people visited last month. It's the odds that a specific attempt, one search or one listing, ends in a transaction. That's why a marketplace with fewer users can be more liquid than a bigger one.
A working definition for this article: marketplace liquidity is the share of buyer and seller attempts that end in a completed transaction within a time window the participants consider reasonable. The time window matters. A used-car listing that sells in six months and a ride request that's matched in six minutes may both be "successful," but they live in different markets with different expectations.
Two sides of the same idea
Because a marketplace has two sides, liquidity has two faces. People often say "the marketplace is liquid" when they really mean one side is.
Buyer-side liquidity asks: when a buyer comes looking, do they find something and buy it? Point Nine's essay measures this with search-to-fill, the likelihood that a search or request leads to a transaction. Examples it gives include the percentage of search sessions that end in a purchase on a shopping marketplace, or the share of job postings that end in a hire on a freelance platform.
Supply-side liquidity asks: when a seller lists something, does it get used? The same essay captures this with the utilization rate, the proportion of supply that's actually used. Think of stock sold per month on a retail marketplace, rooms booked per night on a lodging one, or the share of time a driver spends with a passenger.
The two don't move together automatically. Here's an illustrative example, with invented numbers, to show how they can split:
| Scenario (illustrative) | Buyer-side | Supply-side | What it means |
|---|---|---|---|
| Too little supply | Searches often return nothing useful | Almost every listing sells | Buyers leave disappointed, sellers are happy |
| Too much supply | Searches almost always find a match | Most listings never sell | Buyers are happy, sellers quietly churn |
| Balanced | Most searches convert | Most listings get used | Both sides keep coming back |
A marketplace in the second row looks fine from the buyer's side and may be quietly bleeding sellers. That's why a one-sided metric is risky, and why the supply retention question can't be answered from buyer data alone.
How liquidity is measured
The a16z glossary says liquidity can be measured through metrics like fill rate (also called match rate or utilization rate), market depth, and time to find a match. It doesn't give a threshold for any of them, and neither does this article. The right level depends on the category.
| Metric | What it measures | Side it describes |
|---|---|---|
| Search-to-fill | Share of buyer searches or requests that end in a transaction | Buyer |
| Sell-through rate | Share of listings that sell in a given period | Supply |
| Utilization rate | Share of supply capacity that's actually used (nights booked, hours driven) | Supply |
| Time to fill (or time to match) | Time from request or listing to a completed match | Both |
| Market depth | How many relevant options sit behind a typical search | Buyer |
| Buyer-to-supplier ratio | How many buyers one supplier can serve | Balance |
A few notes on using them.
Pick the metric that fits the business model. Point Nine's essay observes that time to fill matters most where the marketplace picks the match for the buyer, as ride-hailing does. Where buyers browse and choose, as on a classifieds site, search-to-fill and sell-through tell you more.
Cut the numbers finely. The same essay stresses measuring by category, geography, and time interval. A blended number is an average of strong and weak pockets, and the weak pockets are where users leave. If you've read our piece on search and matching, you'll know that a failed search is often a ranking problem, not a supply problem, which is another reason to look at searches individually.
Define the denominator carefully. Search-to-fill depends entirely on what counts as a search. If bots, repeat queries, and window-shoppers are all in the denominator, the metric will look worse than the real buyer experience. If you filter to only the sessions that showed purchase intent, it'll look better. Write down the definition and keep it stable, or the trend line will mean nothing.
Watch the time window. A sell-through rate of 60% in a month and 60% in a year are different businesses. Always state the window next to the number.
Liquidity versus density
People use the two words as if they were synonyms. They aren't, though they're close cousins.
Density is about concentration: how many suppliers and buyers sit within reach of each other in a given place or category. It's a property of the market's structure. Liquidity is about outcomes: how often reaching actually ends in a transaction. It's a property of the results.
The relationship is causal, in one direction. Higher density tends to produce higher liquidity, because more nearby options mean more searches find something and more listings find a taker. But density doesn't guarantee liquidity. A dense market can still be illiquid if the supply is the wrong kind, if prices are far out of line with what buyers will pay, or if search shows the wrong results. Think of a city packed with sellers of items nobody in that city wants.
So the practical split is this. Density is something you build by choosing where to concentrate. Liquidity is what you check to find out whether it worked. When a team says "we need more liquidity," the useful follow-up is whether the problem is too little reach (a density issue) or too little conversion once reach exists (a matching, pricing, or quality issue).
Uber's annual report gives a feel for how public companies use the word. It says its platform dynamics grow network scale and liquidity, which further increases the value of the platform. Notice that scale and liquidity are listed as two separate things. Scale is size. Liquidity is how well that size converts into completed trips.
Liquidity and network effects
A network effect, in the broad sense, is a product becoming more valuable as its user base grows. The a16z glossary also defines the two-sided version: a marketplace's network becomes more valuable as the number of users on the other side increases. For the fuller treatment, see our reference on network effects.
Liquidity is the mechanism through which that value shows up. More sellers raise the odds a buyer finds a match. More buyers raise the odds a seller gets one. Each side's improved experience pulls in more of the other side, and the loop feeds itself. That's the logic behind the flywheel effect, and a marketplace's liquidity is the speed at which the wheel turns.
Two cautions keep this from sounding too tidy.
First, the loop runs in reverse. If liquidity drops, buyers who found nothing don't return, sellers who sold nothing leave, and the market thins out further. A marketplace can pass a tipping point in either direction.
Second, a network effect isn't proof of liquidity. Plenty of marketplaces have lots of users and weak conversion. The size of the network is the potential. Liquidity is the realized value. That distinction is why teams that obsess over user counts sometimes discover that their growth is made of people who never transact.
It also explains the cold-start problem. A new marketplace has no liquidity because it has few participants, and it has few participants because it has no liquidity. We cover the way out of that loop in the chicken-and-egg problem article.
What liquidity can't tell you
A liquid marketplace is a necessary condition for a healthy one, not a sufficient one. A few things liquidity alone will miss.
- Quality of the match. A transaction that happens isn't the same as a good transaction. A marketplace can hit a strong fill rate by connecting buyers to mediocre suppliers, and pay for it later in refunds and lost repeat business. See our reference on repeat usage.
- Leakage. If buyers and sellers meet on the platform and then complete the deal elsewhere, the transaction never shows up in your fill rate. Your real liquidity is higher than your measured liquidity, while your revenue lags.
- Multi-homing. A supplier who also lists on three competing platforms may show healthy utilization on yours while giving you only the leftovers. We cover that dynamic in multi-tenanting.
- Price. A market can clear only because suppliers cut prices to the bone, which looks like liquidity and is really a margin problem.
For where liquidity sits in the larger path from visit to purchase, a conversion funnel breaks that journey into stages, which helps locate where matches are being lost.
Practical rules of thumb
These follow from the definitions above. They're reasoning, not data.
- Measure both sides. Track a buyer-side metric and a supply-side metric together, and treat a gap between them as a signal.
- Segment before you average. Compute liquidity per category, per area, and per cohort before you report a company-wide figure.
- Fix the definition once. Write down what counts as a search, a listing, and a completed transaction, and don't change it mid-quarter.
- Diagnose before you spend. Low liquidity can mean too little reach, poor matching, wrong supply, or price misalignment. Subsidies only help with the first.
- Compare against yourself. There's no universal benchmark, so judge a market by its own trend and by how its best segments perform.
Key Facts: Marketplace Liquidity
- NFX's James Currier defines liquidity as the probability of selling something you list or of finding something you're looking for, as relayed in Point Nine's October 2019 essay.
- That essay measures buyer-side liquidity with search-to-fill and supply-side liquidity with the utilization rate, and recommends cutting both by category, geography, and time interval.
- The a16z marketplace glossary (Li Jin, February 2020) says liquidity can be measured through fill rate, market depth, and time to find a match.
- Uber's FY2025 10-K treats network scale and liquidity as distinct drivers of platform value.
- Density describes structure (how close supply and demand are). Liquidity describes outcomes (how often they actually transact).
- Neither source gives a universal "good" liquidity number. Targets have to come from your own category data.
Related Reading
