Advertising as Marketplace Revenue

Turn this article into takeaways for your work.

Each assistant summarizes the article only for you and suggests best practices for your work.

A marketplace that has built a lot of traffic and a lot of sellers has something advertisers want: people who are ready to buy, and sellers who want to be seen by them. Selling placement on that traffic is a different business from charging a fee on each sale. It's priced by attention, not by transaction, and it can sit on top of the commission without replacing it.

This article is the operator's view. Why do marketplaces sell advertising to their own sellers? How does it change the economics? What does it cost in buyer experience and seller goodwill? And when is a marketplace ready for it? If you're a seller trying to run campaigns, that's a different topic, covered in Amazon advertising. Here we stay on the platform side of the table.

What marketplace advertising is

Marketplace advertising is any paid placement the platform sells to participants on its own surfaces, or sometimes off them. The common forms:

  • Sponsored listings. A seller pays to have a listing appear higher or more often in search results and category pages. It looks like an ordinary result with a label.
  • Display and banner placements. Brand-level slots on home pages, category pages, or checkout flows, usually sold to larger brands.
  • Retail media. The industry term for ads sold on a retailer or commerce platform, targeted using the platform's own purchase and browsing data.
  • Off-site ads. The platform buys traffic on external channels (search engines, social networks) for its sellers' listings and charges the seller when a sale results.

Amazon's 10-K describes its version plainly: it provides advertising services to sellers, vendors, publishers, authors, and others, through programs such as sponsored ads, display, and video advertising, and recognizes revenue as ads are delivered based on clicks or impressions. Shopee's paid ads, per the Sea 20-F, work as a keyword auction: sellers bid for keywords that match their listing, and the listing shows higher in search rankings when users search for those keywords. Sellers prepay, and income is recognized on clicks.

That's the core mechanism. A seller wants more visibility, the platform controls visibility, and the two agree a price.

Why operators add it

Three reasons show up again and again.

It monetizes attention the commission doesn't capture. A commission is earned only when a sale closes. But a buyer who searches, browses, and leaves still generated value for sellers who wanted to be seen. Ads charge for that attention. A listing that never converts still earns the platform money when it's clicked.

It tends to carry less direct cost than fulfillment-heavy revenue. Ads are served from software the platform already runs. There's no shipping, no payment processing, no support ticket per click. We can't say what any company's ad margin is, because most don't disclose it. But the cost structure is why operators treat ads as a high-margin line, and why DoorDash describes advertising as a value-added service offered through its Marketplaces to help merchants and consumer packaged goods companies increase consumer engagement and drive incremental revenue.

It lets the platform lower other fees. Instacart's 10-K says its ability to sustain or increase profitability depends in part on advertising revenue, and that failure to grow it could affect its ability to strategically lower fees and invest in larger marketing campaigns. Ad revenue is a cushion that lets the commission or delivery fee stay competitive.

What the filings show

Each company defines its ad line differently, so the figures are not comparable. Read the definition column before the number.

Company Figure Fiscal year How the company defines it
Amazon $68,635 million "advertising services" net sales, up from $56,214 million 2025 Sales of advertising services to sellers, vendors, publishers, authors and others: sponsored ads, display, video
Instacart (Maplebear) $1,065 million "advertising and other" revenue, "substantially all of which relates to advertising revenue" 2025 Reported next to transaction revenue of $2,677 million
Etsy Services revenue $876.3 million, 30.4% of total revenue; on-site ads are the primary component 2025 Optional seller services, mainly on-site advertising and shipping labels. Offsite Ads fees sit in marketplace revenue instead
DoorDash No separate dollar figure; ads recognized within Marketplaces revenue 2025 Revenue growth outpaced Marketplace GOV partly on "increasing contribution from advertising revenue"
Shopee (Sea) No separate ad figure in the 20-F; paid ads named as the first of three ways Shopee is monetized 2025 Keyword-bid paid ads, recognized on clicks

Sources: Amazon 10-K, Instacart 10-K, Etsy 10-K, DoorDash 10-K, Sea 20-F, all for fiscal 2025.

A few computed comparisons, using only the filings' own numbers:

  • Amazon: advertising was $68,635 million of $716,924 million in consolidated net sales, about 9.6%. It grew about 22% from 2024 ($68,635 / $56,214). Third-party seller services, the closest thing to a commission line, was $172,162 million, so advertising was roughly 40% of that figure. These are our calculations from the 10-K's net sales table, not Amazon's statements, and the two lines are defined differently.
  • Instacart: $1,065 million of $3,742 million total revenue ($2,677 million transaction plus $1,065 million advertising and other), about 28%. Again computed by us.
  • Etsy: services revenue rose $88.7 million in 2025, and the filing attributes a $62.2 million increase in advertising revenue, largely driven by a higher average price per click on Etsy Ads, to that growth. In a year when consolidated GMS fell 5.3% (a figure that includes the Reverb sale), ads carried much of the revenue growth.

On Uber, the figure to know is that it's early. Uber's 10-K says it officially launched its advertising division in October 2022 and describes the offering as marketplace-centric advertising connecting merchants and brands with its platform network. We tried to confirm the run-rate Uber has cited on earnings calls from a readable primary page and couldn't, so we're not quoting it.

How ads change the take rate

If take rate is revenue divided by gross merchandise volume, ad revenue goes into the numerator and nothing into the denominator. A marketplace can leave its commission unchanged and still see its reported take rate climb as ads grow. The mechanics are covered in marketplace take rate, including how definitions differ by company.

Etsy shows why you must check the definition. Its 10-K defines revenue take rate as revenue divided by GMS, and reports 24.2% for 2025 against 22.3% in 2024. That figure includes the ad lines. Transaction fees are only part of what's in it. So a rising take rate there doesn't mean sellers are paying a higher commission. It can mean more sellers are buying ads, or paying more per click.

Etsy also splits ads across two lines. On-site ads count as services revenue. Offsite Ads, where Etsy places ads on third-party platforms and charges the seller only if a sale results, are charged as a transaction fee. The 10-K says sellers pay 12% or 15% of the value of a sale based on the seller's volume of sales when the sale comes from an Etsy-placed ad on those platforms, and the cost shows up in marketing expense. Check Etsy's help page for the full terms, such as thresholds and caps, before relying on specifics.

Ads are one of several value-added services a platform can sell. For the others, see marketplace value-added services.

The auction and relevance mechanics, at concept level

Most marketplace ad systems share a few pieces.

  1. Inventory. The slots: positions in search results, category pages, product pages, emails.
  2. An auction or price rule. Sellers bid on keywords or products, and the platform picks who appears. Shopee's description, bidding on keywords that match a listing, is the textbook case.
  3. A relevance layer. Pure highest-bid-wins shows irrelevant listings and teaches buyers to ignore the ads. Most platforms blend bid with a predicted chance the buyer will click or buy.
  4. Pricing by action. Pay per click, per impression, or per sale. Amazon recognizes revenue on clicks or impressions. Etsy's Offsite Ads charge only when a sale happens.
  5. Measurement. Sellers need to see what they got. Without attribution, spend stops.

Etsy's 10-K says improvements to Etsy Ads came from machine learning-based ranking optimizations, more efficient seller budget pacing, and more seamless integration of onsite ads within organic search results, which it says increased ad relevancy and efficiency. Note the direction: the platform made ads blend into organic results better. That's good for ad performance. It's also where the trade-offs start.

The trade-offs

Buyer experience. Every sponsored slot displaces an organic result. If buyers can't tell paid from earned, or the paid result is worse, trust erodes. Relevance filters help, but they cost some short-term ad revenue.

Ads as a de facto seller tax. When ads blend into organic search, a seller who doesn't pay can lose visibility they used to get free. Sellers then spend to hold their position rather than to grow. The platform's revenue rises, and the seller's margin falls. A marketplace that depends on sellers staying profitable has to watch this closely. See marketplace supply quality for why seller health feeds buyer experience.

Pay-to-play pressure on small sellers. Large sellers and brands can outspend small ones. If paid placement dominates the first page, a long tail of small sellers sees less traffic, and the variety that attracted buyers may thin out.

Conflict with organic ranking. The team that ranks results for buyers and the team that sells slots to sellers have different goals. Without clear rules on which wins, ads creep upward and organic quality slips. Instacart's 10-K lists this kind of tension among its risks: it says its advertising model is still being refined, and that growth rates have fluctuated, particularly when GTV growth accelerates or decelerates.

Regulatory and legal scrutiny. Amazon's 10-K discloses litigation alleging price fixing, monopolization, and consumer protection claims, including from state attorneys general and the Federal Trade Commission. The filing doesn't tie those to advertising specifically, so we won't claim it does. But a platform that controls both ranking and paid placement should expect questions about self-preferencing and disclosure.

Dependence on traffic. Ad revenue follows volume. Instacart's filing notes that its advertising and other revenue is seasonal, typically high in the fourth quarter. A downturn in buyer traffic hits ads and commission together.

When a marketplace is ready

Advertising is a second business. Rules of thumb, which are our reasoning rather than data from any filing:

  1. Sufficient buyer traffic. Ads need enough impressions and clicks that a seller can measure a return. Thin traffic means noisy results and disappointed advertisers.
  2. Enough sellers competing for the same buyers. An auction needs bidders. If a category has three sellers, there's nothing to bid for. This is the density idea again: depth within a category or area.
  3. Search that already works. If organic search is poor, ads amplify the poor experience. Fix relevance first.
  4. A way to measure. You need clicks, orders, and attribution before you charge for them.
  5. Sellers who are earning money. If sellers are barely profitable, ads will feel like a tax and push them out. Revisit your buyer acquisition engine too: ads that lean on paid acquisition to fill your own traffic can create circular spend.
  6. Rules written before launch. How many ad slots per page, how they're labeled, whether sellers can opt out, how complaints are handled.

Early marketplaces usually do better putting these months into liquidity, trust and repeat behavior. Ads are a way to monetize a working marketplace, not to create one. For how ads fit alongside commissions, subscriptions and listing fees, see marketplace revenue models.

Key Facts: Marketplace Advertising

About the author

Brian Tr

Brian Tr

Co-Founder & COO

Brian Tr is Co-Founder and COO of Rework, with 12+ years in B2B go-to-market and operations. Brian scaled Rework from 0 to 10,000+ B2B customers across CRM and productivity tools. Brian writes for founders and owner-CEOs: startup fundamentals, founder-led and family businesses, partnerships, and how SaaS, marketplace, AI and EdTech companies grow.