Every marketplace founder I've worked with has a "metrics dashboard." Most of them measure the wrong things. GMV charts climb because the founder seeded supply. Listing counts grow because they ran an acquisition campaign. None of it tells you whether the marketplace is actually working — whether buyers and sellers are finding each other, transacting, coming back.
This post is the ordered short-list. Five metrics, in priority order, with the thresholds we've seen separate marketplaces that survive from the ones that quietly fold.
The short answer
The KPIs that predict marketplace growth are liquidity (match rate of listings to transactions), 30-day repeat purchase rate, take rate net of payment fees, seller retention at 90 days, and LTV/CAC ratio per side. Vanity metrics like GMV, total users, and listing count are lagging — they show what happened, not what's about to.
Watch first: marketplace metrics that matter
Sharetribe's Marketplace Academy published a short orientation on the metrics that reliably tell you how your marketplace is doing. Worth watching before the deep dive below — it covers the conceptual framework around the specific numbers.
Why most marketplace dashboards are wrong
The metric most founders watch first is GMV. It's the wrong one. GMV is the output of a healthy marketplace, not the predictor. You can pump GMV with seller incentives, buyer discounts, and paid acquisition — and have an unhealthy marketplace underneath. We watched one services marketplace grow GMV 4x in six months while their repeat purchase rate dropped from 32% to 11%. The growth was a churn-funded mirage. Six months later they ran out of runway.
a16z's 13 marketplace metrics is the canonical reference, but it's a buffet — useful if you know what to pick. The five below are the prix-fixe menu: the smallest set that gives you the full picture.
KPI 1: Liquidity (the only metric that matters in year one)
Liquidity measures whether the two sides of your marketplace are actually finding each other. The strict definition: the percentage of listings (or buyer searches) that result in a transaction within a defined time window. The window depends on your category — rentals run 7-30 days, on-demand services run hours, ecommerce runs days.
Two common forms:
- Supply-side liquidity: percentage of active listings that get at least one booking/order in the period.
- Demand-side liquidity: percentage of buyer sessions that end in a transaction.
On our vacation rental marketplace deployment, supply-side liquidity sat at 18% in month 1 (post-launch) and climbed to 41% by month 6. The shift came from removing 30% of the worst-converting listings (they were dragging the average and confusing buyers). Sometimes fixing liquidity means removing supply, not adding it — counterintuitive but real.
The threshold: across categories, marketplaces that survive long-term hit at least 30% supply-side liquidity by month 12. Below 20% and you're in a discovery problem; below 10% and you have a fundamental product-market fit issue. The marketplace cold-start playbook covers how to push liquidity up specifically — it's almost always the first lever to pull.
KPI 2: Repeat purchase rate (the leading indicator GMV hides)
Repeat purchase rate measures whether buyers come back. The strict definition: percentage of buyers who completed at least 2 transactions within their first 30 days, or sometimes 90 days for less frequent categories. This is the closest thing to a true product-market-fit signal a marketplace gets.
Why this beats GMV: a buyer who transacts once is acquired traffic. A buyer who transacts twice is a real user of your product. The difference compounds — by month 6, repeat buyers drive 60-80% of GMV on healthy marketplaces, even though they're only 20-30% of users.
Thresholds we've seen separate winners from churners:
| Category | 30-day repeat rate (healthy) | 90-day repeat rate (healthy) |
|---|---|---|
| On-demand services (cleaning, ride, food) | 35-50% | 60-75% |
| Rentals (vacation, gear) | 5-12% | 15-30% |
| Physical goods (used/handmade) | 15-25% | 30-45% |
| Skilled services (consulting, design) | 10-20% | 25-40% |
Below these floors, the cost of acquisition will eat the lifetime value. Above them, your CAC payback compresses and growth becomes self-sustaining.
KPI 3: Take rate net of payment fees
Take rate is the percentage of GMV your platform keeps as revenue. Most metric guides stop there. The honest measurement is take rate net of payment fees — what's actually left for the business after Stripe, FX, cross-border surcharges, dispute costs, and Connect platform fees.
Gross take rates of 10-15% are normal on marketplaces. Net take rates of 6-10% are typical after fees. On a global marketplace with 30%+ cross-border volume, the gap widens to 200-400 basis points. We documented the full breakdown in the Stripe Connect production gotchas guide — the short version is that cross-border alone eats 50-100 bps that most founders never model.
For a working calculator with your specific country mix and account type, the marketplace fee calculator models all of this. Plug in your numbers and you'll see the net take immediately — usually 30-40% lower than the gross take you've been quoting investors.
The number to watch: if net take rate trends down month over month while gross GMV grows, you're scaling a less-profitable mix (more cross-border, more dispute-prone categories, more international sellers with higher Stripe fees). Fix the mix before the volume buries you.
KPI 4: Seller retention at 90 days
Buyer churn gets all the attention. Seller churn is more dangerous. A churned buyer is a single transaction lost. A churned seller takes their entire listing inventory and trust signals with them — and if they leave a bad review of your platform on the way out, they damage acquisition.
The metric: percentage of sellers who completed at least one transaction in days 1-30 who also complete at least one in days 61-90. Anything below 40% means sellers are trying you and leaving. Anything above 70% means you have a sticky supply side, which is the foundation of marketplace defensibility.
What kills seller retention, in order of frequency:
- Payout friction or delays. If sellers wait 14 days for their first payout and it's $40, they don't come back. The Stripe Connect mistakes guide walks through fixing the worst payout-hold patterns.
- Onboarding KYC drop-off. Sellers who never finish verification never sell. Common in India/Brazil/Mexico where KYC requires more documents.
- Low demand-side liquidity. Sellers leave when their listings don't sell. This loops back to KPI 1 — fix liquidity, retention follows.
- Fee opacity. Sellers who can't predict what they'll earn after platform + Stripe fees stop trusting the platform.
KPI 5: LTV/CAC ratio per side
The standard SaaS metric, with a marketplace twist: you have two sides, each with its own LTV and CAC, and both have to clear 3:1 for the marketplace to be sustainable. A 4:1 buyer ratio with a 1:1 seller ratio means you're acquiring expensive sellers who don't pay you back — your marketplace is one side healthy and one side bleeding.
Real numbers from marketplaces we've worked with (2026 dollars):
- Vacation rental marketplace: Buyer CAC ~$25, buyer LTV (gross profit) ~$95 → 3.8x. Seller CAC ~$180, seller LTV ~$640 → 3.6x. Both clear 3x — healthy.
- Services marketplace at scale: Buyer CAC ~$15, buyer LTV ~$80 → 5.3x. Seller CAC ~$420 (paid acquisition + onboarding ops), seller LTV ~$1,100 → 2.6x. Buyer side strong, seller side marginal — they're growing demand faster than supply can monetize.
The trap most marketplaces fall into: only tracking the cheap side. Buyer-side metrics are easier to measure and usually better, so founders proudly quote those. Seller-side economics are where the unit-economics truth lives — and where the platform actually makes money.
What to track weekly (and what to ignore)
Once you have the five above, ignore everything else for the first 12 months. The metrics most founders fixate on — total users, listing count, app downloads, social followers, press mentions — are vanity. They go up while the business falls apart.
The honest weekly dashboard for an early-stage marketplace:
| Metric | Frequency | What to do if it drops |
|---|---|---|
| Supply-side liquidity | Weekly | Audit listings; remove low-converters; improve discovery |
| 30-day repeat rate | Weekly (rolling) | Investigate first-transaction experience; talk to one-time buyers |
| Net take rate | Monthly | Check fee mix shift; review dispute trend |
| Seller retention (D90) | Monthly | Payout audit; KYC drop-off check; demand-side liquidity |
| LTV/CAC per side | Quarterly | Reduce paid acquisition on the weaker side; double down on referrals |
For founders deeper into marketplace measurement specifically, the Sharetribe Marketplace Academy reference covers the 26-metric superset; Lenny Rachitsky's most important marketplace metrics post is the practitioner's complement.
Tooling: what to build vs buy
You can run the five KPIs above on a spreadsheet for the first $50k/month GMV. Past that, instrument them properly. Build the dashboards on top of your transaction data — every Sharetribe or custom Rails marketplace has the raw events you need (transaction.transitioned, listing.created, user.signed_up). Pipe them to a warehouse (Postgres or BigQuery), build the five views as SQL, refresh nightly. That's it. Don't pay for a marketplace-analytics SaaS at this stage — they're priced for $1M+ GMV marketplaces and don't customize for your specific liquidity definition.
For the cost of building this in-house vs alternatives, the true cost to build a marketplace breakdown covers ongoing analytics as a line item. And if you're still deciding the platform foundation (Sharetribe vs Custom), the decision wizard walks through the trade-offs.
FAQ: Marketplace KPIs
What's the single most important marketplace metric?
Liquidity — the percentage of listings (or buyer searches) that result in transactions within a category-appropriate time window. Below 20% supply-side liquidity, no marketplace survives long-term. Above 30%, almost everything else becomes solvable.
How is take rate different from net take rate?
Take rate is the percentage of GMV your platform charges as a fee. Net take rate subtracts payment processing, cross-border surcharges, FX conversion, and dispute costs — what's actually left for the business. Net is typically 30-40% lower than gross.
What's a good LTV/CAC ratio for a marketplace?
3:1 or higher for both sides. A healthy buyer ratio with a marginal seller ratio (or vice versa) is unsustainable — both sides have to pay back. Marketplaces fail when founders only track the cheap side.
How often should I review marketplace KPIs?
Liquidity and repeat rate weekly. Take rate and seller retention monthly. LTV/CAC quarterly. More often than that is noise; less often than that and you'll miss directional shifts until they're catastrophic.
Why is GMV a misleading metric?
GMV measures activity, not health. You can pump GMV with seller incentives, buyer discounts, or paid acquisition while underlying retention and liquidity collapse. GMV is a lagging output — by the time it drops, the underlying metrics have been broken for months.
Stuck on marketplace measurement? Let's review your metrics stack.
We've built marketplace analytics dashboards on top of Sharetribe Flex and custom Rails marketplaces — including instrumenting the five KPIs above end-to-end. If you're early-stage and trying to figure out what to track, or scaling and trying to instrument liquidity properly, our marketplace engineering team does free 30-minute architecture reviews on measurement infrastructure too — not just code.
Get a free estimate — we'll review your requirements and propose a plan within 48 hours.
