What Is a Managed Marketplace?

By Brian Tr · Updated Oct 6, 2026

A managed marketplace is a platform that connects buyers and suppliers and also takes responsibility for part of the transaction itself. It vets who can sell, sets or guides the price, controls quality, handles fulfillment or service delivery, and steps in when something goes wrong. In an open marketplace the platform provides a venue. In a managed one it provides an outcome.

The idea is easiest to see in a16z's own definition. Li Jin describes managed marketplaces as ones that intermediate parts of the service delivery, adding value by taking on functions like identifying high-quality providers, standardizing prices, and automating matching between demand and supply. Each of those functions is work the platform does so the buyer doesn't have to.

This article defines the model, places it on the spectrum between open marketplaces and full-stack companies, and covers what it does to take rate, margins, trust and scalability. It's a spoke in the fundamentals section of this collection, and the hub is the overview of what a marketplace business is.

What a platform takes on in a managed model

Open marketplaces mostly match and get out of the way. Managed marketplaces take on some or all of the following.

• Vetting supply. Not everyone who applies gets to sell. The platform checks identity, credentials, quality or authenticity before a supplier is visible.
• Setting or guiding price. The platform standardizes price, suggests it, or sets it outright, so buyers aren't comparing a hundred different quotes.
• Quality control. Inspection, review of each submission, photography, written descriptions, or standards that suppliers must meet.
• Fulfillment and service delivery. Logistics, shipping, scheduling or the service visit itself runs through the platform.
• Guarantees and dispute handling. Returns, buybacks, service guarantees and a human who resolves problems.

Ezra Galston's 2017 TechCrunch analysis lists the same work in a consumer context. He describes a managed marketplace as a value-added intermediary that provides a superior experience versus more traditional peer-to-peer marketplaces, and notes the platform performs authentication, quality assurance, inspections, logistics, customer service, photography and copywriting. Many of these costs are incurred before a sale happens.

The reason platforms take this on is trust and conversion. A buyer purchasing a luxury handbag, hiring a stranger into their home or committing to a home sale can't easily judge quality alone. If the platform absorbs that uncertainty, more buyers complete more purchases, and suppliers get a steady flow of demand without building a brand themselves.

The spectrum: open, managed and full-stack

Think of control as a dial rather than a switch. At one end the platform controls almost nothing about the transaction. At the other it owns everything, and it stops being a marketplace.

Chris Dixon's 2014 essay on full-stack startups marks the far end of that dial. He contrasts the old approach, which was to sell or license your technology to the existing companies in that industry, with the new one: build a complete, end-to-end product or service that bypasses existing companies. The advantage, he argues, is the ability to bypass incumbents, control the customer experience and capture more of the economic benefit. He names Tesla, Warby Parker, Uber, Nest and Netflix as examples.

A managed marketplace sits between the pure listing site and that end-to-end company. It controls the experience, but supply still comes from independent third parties.

The edges blur. A platform can be open for some categories and managed for others, or add managed features such as a payment guarantee or a certification program over time. The useful question isn't "is this a marketplace?" but "which parts of the transaction does the platform own?" For the open end of the dial, see what a marketplace business is and the marketplace business model overview.

Named examples

Real companies show different points on the dial.

TheRealReal. Its annual report describes an end-to-end model for luxury consignment. According to the company's 10-K for 2025, all consigned items are authenticated, written up, photographed, priced, sold and fulfilled on behalf of the individual consignor. That's close to the full list of managed functions. In the same filing, the company says consignors achieved an overall commission rate of approximately 62 percent in 2025, which implies the platform kept roughly the other 38 percent of sale proceeds (by subtraction, not a figure the filing labels). That's a much larger share than a typical listing fee.

Etsy. In contrast, Etsy's 10-K says that because it does not hold physical inventory, it has an asset-light operating model. Sellers hold stock and ship. It sits much nearer the open end of the spectrum.

Earlier managed experiments. The 2017 TechCrunch piece gives a snapshot of the model's first wave, with figures reported as of that date. It cites Opendoor, which bought homes and managed repairs, and TheRealReal, whose take rate it put at 30 percent, effectively triple peer-to-peer rates. It also cites Beepi, a used-car platform offering inspections, return windows and buyback guarantees at a fee of about 9 percent, compared with eBay Motors' 1.25 percent on a $10,000 car. The article reports that Beepi burned through nearly $300 million in the span of two years before closing. That's the cautionary side of the model, and the next section explains why.

What management does to the economics

Take rate

Taking on more of the transaction lets the platform charge more. The managed examples above retained far larger shares than typical listing fees. The higher rate isn't pure margin. It pays for inspection, logistics, support, photography and the guarantees. A high take rate on a business with high costs can still produce a thin result. Compare the two models on what's left after costs, not on the headline percentage. The two-sided markets article covers why platforms can price each side differently, and that logic carries over here.

Margins and capital

Open marketplaces are asset-light. Managed ones often aren't. Galston points out that these models add risk through activities like pre-purchasing and holding inventory, which front-loads cost before revenue arrives. The question to ask is whether the service cost per transaction falls as volume grows. Galston warns that it's ultimately the technology, not heavy services, that cultivates strong business models and margins over the long term. He uses Shutterstock as a case where content review costs rose, not fell, over 14 years of operation. For the underlying math, see unit economics and gross margin.

Trust and repeat behavior

Management is what lets a platform promise something specific: this item is real, this provider was checked, this job will be redone if it goes wrong. That promise reduces the buyer's perceived risk and can raise conversion. It also changes what the platform is accountable for. When an open marketplace's seller disappoints a buyer, the seller takes the blame. When a managed platform's vetted provider disappoints, the platform does.

Disintermediation

A platform that adds value after the match is harder to bypass than one that only makes introductions. If the platform provides guarantees, payment protection and support, buyer and supplier lose something by taking the relationship off-platform. Galston adds a caveat: future automation could also lower the barriers and defensibility of these companies, letting peer-to-peer players copy the experience with similar software. Operational depth can be a moat or a cost center, depending on whether competitors can copy it cheaply. See economic moat for how to judge that.

Scalability

Supply grows more slowly in managed models because every supplier has to pass the platform's checks and every transaction touches the platform's operations. Growth is gated by vetting and fulfillment capacity, not just by marketing. That's a real constraint, but it's also the source of the quality that buyers pay for.

Supplier classification considerations

When a platform sets prices, directs the work or controls how suppliers behave, regulators may ask whether the people doing the work are independent businesses or workers. This applies mainly to service marketplaces where individuals perform labor, not to platforms selling goods. The rules differ by jurisdiction and change, so treat this as a flag to check, not a legal conclusion.

The European Union's Directive 2024/2831 on platform work is one example. It sets a rebuttable legal presumption of employment between a digital labour platform and a person performing platform work when facts indicating control and direction are found under national law. A platform that wants to rebut the presumption must prove the relationship isn't employment. The deadline for member states to transpose the directive into national law is 2 December 2026. The features that make a model "managed," such as setting price and enforcing standards, are close to the factors regulators look at, so managed service marketplaces should get legal advice early in each market.

When a managed model tends to fit

Managed models tend to appear where the following are true.

Li Jin's piece frames this around regulated categories. Her eight considerations include the downside risk of unlicensed supply, the burden of licensing requirements, existing industry satisfaction, the opportunity to lower prices, market size, latent demand, underutilized assets and regulatory reform tailwinds. Her examples of the platform doing licensing-adjacent work include Outschool's background checks on teachers and Nana's academy for training appliance technicians.

Managed models fit poorly where goods are commodities, quality is obvious from a listing, or the transaction value is too low to cover the cost of human involvement. In those cases an open marketplace is usually cheaper to run. How the models differ for business and consumer buyers is covered in B2B vs B2C marketplaces, and the difference between holding inventory and matching others' is explored in marketplace vs ecommerce store.

Key Facts: Managed Marketplaces

• A managed marketplace takes on functions like identifying high-quality providers, standardizing prices, and automating matching (a16z, Li Jin).
• Dixon's full-stack model builds a complete, end-to-end product or service that bypasses existing companies (2014).
• TheRealReal's consignors achieved an overall commission rate of approximately 62 percent in 2025 (FY2025 10-K).
• In 2017, TechCrunch reported Beepi charged about 9 percent versus eBay Motors' 1.25 percent on a $10,000 car, and had burned nearly $300 million in two years.
• EU Directive 2024/2831 creates a rebuttable presumption of employment for platform workers under control and direction, with transposition due by 2 December 2026.

Related Reading

• What Is a Marketplace Business?
• B2B vs B2C Marketplaces
• Marketplace vs Ecommerce Store
• Marketplace Business Model
• Two-Sided Markets
• Unit Economics