Marketplace Metrics: The KPIs That Actually Matter (And the Ones That Lie to You)

Every marketplace founder I have worked with tracks metrics. Most of them are tracking the wrong ones.

Not because they are not smart — they are. But because the metrics that look good on a dashboard are often not the ones that tell you whether your marketplace is actually working. And in a two-sided market, bad metrics do not just give you a false sense of security. They send you in entirely the wrong direction.

After ten years and 30+ marketplace builds, here is how I think about measurement as a CPO — what to track, when to track it, and which numbers you should treat with serious skepticism.

The metrics that lie to you

Before we get to what matters, let us talk about what does not.

Total registered users

This is the vanity metric that kills more marketplaces than any other. You have 10,000 registered users. How many transacted in the last 30 days? How many listed in the last 30 days? For most marketplaces, the gap between registered users and active users is enormous — and it is the active number that actually tells you anything.

Gross Merchandise Volume (GMV) in isolation

GMV is a meaningful number, but only in context. High GMV with low repeat purchase rate means you are acquiring customers who try the platform once and leave. A smaller GMV number with a 60% repeat rate is a healthier business than a large GMV with 10% repeat.

App downloads or page views

Top-of-funnel activity before you have solved liquidity is noise. Traffic that bounces because there is nothing to buy or no buyers to be found is not progress — it is evidence that you are spending marketing budget too early.

The metrics framework I use with every client

I organise marketplace metrics into four layers. Each layer tells you something different, and you need all four to have a complete picture.

Layer 1: Liquidity

Liquidity is the foundation. Before anything else, you need to know whether buyers and sellers can reliably find each other and complete transactions. The key metrics here:

• Listing fill rate. Of searches conducted on your platform, what percentage return at least one relevant result? Anything below 70% means your supply is too thin for the demand you are trying to serve.
• Time to first transaction. How long after a new supplier lists do they complete their first transaction? A long time-to-first-transaction is a supply engagement problem that will erode supplier retention.
• Liquidity rate. The percentage of listed inventory that transacts within a given period. For a rental marketplace this might be monthly. For a goods marketplace it could be quarterly. Low liquidity rate = supply is not getting value from your platform.

Layer 2: Transaction quality

Transactions are not created equal. A transaction that ends in a dispute or a bad review is worse than no transaction at all — it damages trust on both sides of the market. Metrics here:

• Dispute rate. What percentage of transactions result in a dispute, chargeback, or complaint? Track this by category, by seller cohort, and over time. Rising dispute rate is one of the earliest warning signs of a trust problem.
• Review completion rate. What percentage of completed transactions result in a review? Low review completion is a signal that users do not feel invested in the platform — which predicts churn.
• Average review score. Track this separately for buyers and sellers. A platform where sellers rate buyers poorly is a supply retention problem waiting to happen.

Layer 3: Retention and engagement

The most important metric in any marketplace is repeat behaviour. A marketplace where both sides come back is a marketplace with network effects. One where they do not is a pipeline.

• Buyer repeat rate. What percentage of buyers who transacted in month one return to transact again in month three? I use a 90-day window. Below 20% is a serious problem. Above 40% is a good signal.
• Seller retention rate. What percentage of sellers who listed in month one are still active in month six? Seller churn is often invisible until it is too late — supply disappears quietly.
• Cross-side engagement. Are buyers becoming sellers, or sellers becoming buyers? This is a strong signal of community and network effect depth, and it is almost never tracked.

Layer 4: Unit economics

Once you have some traction, you need to understand whether the business is structurally sound. This is where most early-stage founders under-invest in measurement.

• Take rate. What percentage of GMV do you capture as revenue? Track this over time — pressure on take rate is an early signal of competitive tension or supplier dissatisfaction.
• CAC by side. What does it cost to acquire a buyer versus a seller? These are almost always different numbers, and the ratio tells you where your acquisition efficiency is weakest.
• LTV by cohort. Do not calculate LTV as an average. Segment by acquisition channel, signup month, and geography. Averages hide everything interesting.

The most underrated metric: time segmentation

Here is something most marketplace analytics setups miss entirely: the behaviour of a user in their first 30 days is completely different from their behaviour in month six. If you average these together, you get a number that describes no one accurately.

Segment every retention and engagement metric by account age. New users (0-30 days), developing users (30-90 days), and established users (90+ days) should be tracked separately. The actions a new user takes that predict long-term retention are often not the same actions an established user takes that predict churn.

This is what I mean when I say not all KPIs are one-dimensional. A falling overall engagement rate might be entirely explained by a large cohort of new users still in their onboarding window — and signal nothing at all about the health of your established user base.

How to set up your metrics stack

You do not need complex tooling to do this well at early stage. You need three things:

1. An events framework. Define the 8-12 actions that matter most in your marketplace — listing created, message sent, transaction initiated, transaction completed, review submitted — and make sure every one of these is being captured with a timestamp and a user ID.
2. Cohort-based reporting. Group users by the month they signed up or first transacted. Run your retention and engagement metrics against cohorts, not totals.
3. A weekly review habit. Metrics only matter if someone looks at them and asks why. Schedule a weekly 30-minute review of your key indicators. The question is always the same: what changed, and why?

The best metric is not the one that looks most impressive in a pitch deck. It is the one that tells you the truth about whether buyers and sellers are finding each other and coming back.

Most marketplace founders get their measurement strategy wrong because they inherit a SaaS metrics playbook and apply it to a two-sided market. The dynamics are fundamentally different. The sooner you build a measurement practice designed for how marketplaces actually work, the sooner you will make decisions that compound rather than cancel each other out.