We've helped marketplace founders pick or repair pricing on a Sharetribe vacation rental platform, a custom Rails B2B equipment marketplace, and two service-side platforms in the home services and freelance creative spaces. The pricing decision is one of the three or four choices that shapes everything else — what your LTV looks like, whether your sellers stay, whether you can afford paid acquisition, whether buyers feel ripped off at checkout.
And almost everyone gets it wrong the first time. Usually because they looked at Airbnb or Etsy or Fiverr and assumed the take rate that works at that scale will work at theirs. It won't.
What is the best marketplace pricing model? (the short answer)
For most two-sided marketplaces, a percentage commission (take rate) of 10-20% on every completed transaction is the right default. It aligns incentives — you only earn when sellers earn — and it scales naturally with GMV. Subscription pricing wins only when transactions happen off-platform or seller LTV is predictable enough to underwrite a fixed monthly fee. Hybrids dominate at scale.
Watch first: how marketplace take rates actually work
Before the model-by-model breakdown, this is a16z's Connie Chan walking through marketplace take rate strategy — defensibility, payments, and why some marketplaces sustain 25%+ take rates while others can't hold 5%. Worth 15 minutes if you're choosing a pricing model from scratch.
The five canonical models
1. Commission (take rate) — the default for a reason
You take a percentage of every completed transaction. Sellers list for free, buyers pay nothing extra (or a small service fee), and the platform earns when money moves.
Why it dominates: incentives align cleanly. The platform invests in growth, conversion, trust, dispute resolution — every improvement increases GMV, which increases revenue. Sellers don't pay upfront, so liquidity is easier to bootstrap.
The trap: take rate is visible and resented. Sellers calculate it constantly. A 25% take rate sounds fine in a board deck and feels predatory to a hairdresser keeping $75 of every $100 booking. Anchor your rate against the next-best alternative your sellers have — including no platform at all.
Real benchmarks we've seen and validated against published data:
| Marketplace | Take rate | Why it holds |
|---|---|---|
| Etsy | ~6.5% + $0.20 listing | Buyer demand sellers can't replicate elsewhere |
| eBay | ~10-13% | Auction format + audience scale |
| Airbnb | ~14-16% (host + guest split) | Search distribution + trust infrastructure |
| Uber | ~25-28% | Demand routing + insurance + payments |
| Fiverr | ~20% from sellers + 5.5% buyer fee | Project escrow + dispute infrastructure |
| Sharetribe median (across ~5,000 marketplaces) | ~9-10% | Wide variance by vertical |
The pattern: take rate scales with the operational value the platform contributes. Bare-bones listing platforms can't sustain Uber-grade rates. Managed marketplaces with insurance, dispute resolution, payments, and demand routing can.
2. Subscription — predictable revenue, narrow fit
Sellers (or buyers, occasionally) pay a recurring monthly fee. Transactions are free or capped. Think LinkedIn Premium, Shopify, some B2B equipment marketplaces.
When subscription wins:
- Transactions happen off-platform. If buyers and sellers will inevitably exchange contact info and finalize outside your platform (real estate leads, equipment quotes, B2B procurement), you can't enforce a take rate. Subscription captures value at access.
- Seller LTV is high and predictable. A real estate agent will pay $400/month for lead access because one closed deal pays for years.
- Sellers want predictable costs. Variable take rate creates anxiety for budget-bound businesses. Subscription is a known line item.
When subscription fails: when sellers can't predict whether they'll get enough activity in a given month to justify the fee. Then they churn after their first slow month, and you've trained the market that your platform is expensive without being valuable.
The pattern we recommend: start with commission to prove value, then graduate high-volume sellers to a subscription tier with reduced or zero commission. Etsy Plus, Amazon Professional Seller, and Upwork's plan structure all work this way.
3. Listing fees — niche, mostly additive
Sellers pay per item listed (Etsy's $0.20, eBay's variable insertion fee). Pure listing-fee models are rare; the model usually layers on top of commission to filter out low-quality listings.
The hidden value of a listing fee isn't revenue — it's signal. A 50-cent listing fee kills 95% of joke listings and bot spam without scaring off real sellers. We've recommended this to a Sharetribe B2B platform where buyers were drowning in noise; introducing a $1 listing fee cut spam by an order of magnitude and barely moved real-seller activation.
Don't use listing fees as your primary revenue source unless your category is so high-volume per seller (millions of listings, like classifieds) that fractions of cents add up.
4. Lead fees — the right model for "marketplaces" that aren't
Buyer submits a request, platform charges sellers per lead (verified contact). Common in home services, legal, and B2B procurement — Thumbtack, Bark, parts of HomeAdvisor.
The economics: lead conversion rates in the 5-15% range mean a seller paying $20 per lead and closing 1 in 8 effectively pays $160 customer acquisition cost. That works only if the average job value justifies it. For service marketplaces with high-ticket transactions (home renovation, B2B equipment) it works. For low-ticket repeating services (haircuts, dog walking) it doesn't.
Lead fee marketplaces also struggle with the refund death spiral: sellers demand refunds for bad leads (wrong number, tire kickers, jobs already booked). Every refund makes the unit economics worse and erodes seller trust. Build the refund policy and quality controls before launch, not after the first complaint cycle.
5. Featured listings / promoted placements — the addition, not the foundation
Sellers pay extra for visibility — top of search, homepage feature, "promoted" badge. Etsy Ads, eBay Promoted Listings, Airbnb's algorithmic boost-via-Plus.
Featured listings are a great secondary revenue stream once you have liquidity. They're a terrible primary stream. Pre-liquidity, paying to be featured on an empty marketplace is paying for nothing. Add this in year two, after take rate has proven the platform.
The decision framework: pick the model that matches your unit economics
We use a four-question filter when advising founders. Answer these honestly:
| Question | If yes → | If no → |
|---|---|---|
| Can you enforce transactions happen on-platform? | Commission viable | Subscription or lead fee |
| Is seller volume predictable month-to-month? | Subscription viable | Stick with commission |
| Average transaction value over $200? | Lead fee viable | Commission or listing fee |
| Sellers have realistic non-platform alternative? | Cap take rate at 10-15% | Take rate can go to 20-30% |
Most marketplaces should land on commission as the primary model with optional listing fee for signal control. Move to hybrid (commission + subscription tier) once you have power sellers willing to pay for unlimited transactions.
The buyer-side fee question
Should you charge the buyer, the seller, or both? The honest answer: depends on who has more pricing power.
Airbnb famously splits — guests pay a service fee on top, hosts pay a smaller percentage out of payout. This works because guests have already committed by the time they see the fee at checkout. Most marketplaces underestimate how much they can charge the buyer side when transparency is high and value is clear.
The opposite mistake: a marketplace we worked with charged sellers a 22% take rate and zero from buyers. Sellers built the take rate into their listed prices, making the marketplace look 22% more expensive than the seller's own website. Buyers cross-shopped and the platform lost. Rebalancing to 14% seller + 6% buyer fee fixed the optics without changing total economics.
When to change pricing (and how, without losing sellers)
Almost every marketplace raises take rate over time. Airbnb did it. Etsy did it. Uber did it (and ate the PR cost). The mistake is changing pricing without giving sellers value to point at.
Three rules from production:
- Grandfather your top 10% of sellers for at least 12 months. They're your liquidity and your most likely public defectors.
- Pair the price increase with a new capability — better promotion tools, lower payment fees, faster payouts, dispute insurance. Sellers tolerate a price increase that comes with new value far better than a pure margin grab.
- Announce six weeks out, in writing, with the math. Surprise pricing kills trust harder than the actual rate.
If your marketplace runs on Stripe Connect, factor payout schedule and refund mechanics into the pricing change too — those decisions are intertwined and we covered them in detail in our Stripe Connect production mistakes guide.
What this looked like on a real client
On Tutti Vacation, a Sharetribe-powered vacation rental marketplace, we worked through exactly this decision before launch. The initial instinct was a flat 15% take rate matching Airbnb's host-side fee. The wrong move, because Tutti's value proposition was a curated, non-algorithmic experience rather than Airbnb-scale demand routing. The final structure: 10% from hosts + a small booking fee on guests, plus a $5 listing fee per property to keep the inventory curated. That kept the take rate defensible against host comparisons to Airbnb while preserving margin through the booking fee.
If you want the broader picture of what marketplace ops looks like end-to-end — KPIs, cold-start, cost — start with our writeup on the marketplace KPIs that actually predict growth and the itemized cost breakdown for building a marketplace in 2026. The build cost and the pricing model are tightly coupled — a $30k MVP can't sustain a 5% take rate while subsidizing acquisition.
For the build-vs-buy decision that comes before pricing, our Sharetribe vs custom marketplace framework is the right starting point. Pricing flexibility is one of the bigger trade-offs between the two approaches.
External references worth reading
- a16z — 13 metrics for marketplace companies — the canonical metric framework. Take rate and GMV are two of them; the rest are how you justify the take rate.
- Sharetribe — how to set pricing in your marketplace — research-backed framework with the seven factors that drive ideal take rate, drawn from data across ~5,000 marketplaces.
- Tidemark — marketplace take rates — deep operator chapter on how managed vs unmanaged marketplaces sustain different take rates.
FAQ: Marketplace pricing models
What is a typical marketplace commission rate?
For product marketplaces, take rates cluster in the 5-15% range. For service marketplaces, 15-30% is common. The Sharetribe dataset of ~5,000 marketplaces shows a median around 9-10%, but the right rate depends on the operational value your platform contributes — payments, insurance, dispute resolution, demand routing.
Commission vs subscription — which is better for a new marketplace?
Commission. New marketplaces lack the seller LTV data needed to price a subscription correctly, and sellers won't pay a fixed fee until they've earned trust. Start with commission, prove value, then graduate top sellers to a subscription tier with reduced commission.
Should I charge the buyer, the seller, or both?
Both, in most cases. Buyer-side service fees are common (Airbnb, StubHub, Ticketmaster) because buyers have already committed by checkout. Splitting fees across both sides reduces seller resentment and prevents sellers from inflating listed prices to recover the take rate.
How do I raise take rate without losing sellers?
Grandfather your top 10% of sellers for 12 months, pair the price increase with a new capability (faster payouts, better promotion, dispute insurance), and announce six weeks in advance with the math written out. Surprise pricing destroys trust faster than the rate itself.
What pricing model works for B2B marketplaces?
Hybrid: subscription for buyer access (predictable B2B budgets), plus a smaller commission or transaction fee on closed deals. Pure commission rarely works in B2B because transactions often complete off-platform via direct contracts. Lead fees can work when average deal size exceeds $1,000.
How we can help
At TechVinta, we ship marketplaces on Sharetribe and custom Rails — and pricing strategy is usually the first conversation, not the last. We've helped founders pick, validate, and migrate between models without losing the sellers they spent months acquiring. Most engagements start with a pricing review tied to the platform build.
Need a second pair of eyes on your pricing model before launch, or considering a migration? Talk to our marketplace team or get a free estimate — we'll review your unit economics and propose a pricing structure within 48 hours.
