What Is Gross Merchandise Value? GMV, Explained for Product Managers

Gross merchandise value is the total money that changed hands across a marketplace in a period, counted before the platform’s own fees, refunds, and cancellations are removed. No accounting standard defines it, and no auditor signs off on it. Which means two marketplaces can report the figure for the same quarter, using the same three words, and be counting materially different things. The version that reaches a board deck almost never carries the footnote explaining which.

That is not a rhetorical worry. It is visible in filings.

What Gross Merchandise Value Actually Measures

Some platforms fold shipping, handling, and gift-wrap into the headline. Others strip them out and report goods only. Both are defensible. But a PM benchmarking a shipping-heavy category against a competitor that counts shipping is comparing two numbers that were never the same number, and the gap is not rounding error.

The elasticity is not confined to what gets included. It extends to which parts of the company are still inside the figure, and Etsy’s 2025 reporting demonstrates both at once, in public, with the arithmetic printed.

The reasonable objection here is that this is a finance problem, not a product problem. It stopped being one in 2020. The SEC’s interpretive release on key performance indicators in management’s discussion and analysis, Release 33-10751, issued 30 January 2020 and effective 25 February 2020, set out what the Commission expects to accompany a disclosed metric: a clear definition of it and how it is calculated, a statement of why it is useful to investors, a statement of how management actually uses it to run the business, and, if the calculation method changes materially between periods, an explanation of the difference.

Read that list as a product manager rather than a filer: three of the four items are things most internal dashboards do not carry. Explainers written before 2020 describe gross merchandise value as though there were one canonical formula. There is a canonical disclosure obligation instead, and it is the thing that now governs how the number gets presented.

Which raises the harder question: if gross merchandise value is this elastic in its definition, is the metric worth reporting at all?

Etsy’s 2025 Filing Shows Two Right Answers for One Quarter

It is worth reporting. It stops being informative at an identifiable moment, and the moment has a signature.

Etsy’s fourth quarter of 2025 shows the signature with unusual clarity. Consolidated GMS was $3,592.6 million, up 2.4% year over year excluding Reverb from the prior-year period. On an as-reported basis, including Reverb’s fourth-quarter 2024 contribution of $228.3 million, the same figure was down 3.8%. Same quarter, same company, same three-word metric. One number rose 2.4%, the other fell 3.8%, and both are correct.

It was not a one-off. In the third quarter of 2025 the split ran wider: up 0.9% excluding Reverb, down 6.5% as reported once Reverb’s $213.7 million of prior-year volume was left in. A team that quotes “GMS growth” for either quarter without saying which basis it used has communicated nothing.

Where the Composition Changes Underneath the Line

Consolidated gross merchandise sales for the full year came to $11,916.9 million. Of that, the Etsy marketplace contributed $10,460.7 million, or 87.8%, and Depop contributed $1,074.9 million, or 9.0%. Those two named marketplaces account for 96.8% of the total. The missing 3.2% belonged to Reverb, the musical-instrument marketplace Etsy sold on 2 June 2025: still inside the annual figure, no longer inside the company.

Then the harder part, and it has since resolved. On 15 February 2026 Etsy agreed to sell Depop to eBay; that sale closed on 30 July 2026 for approximately $1.4 billion in cash, reflecting the original $1.2 billion purchase price plus roughly $200 million of post-signing adjustments and interest. So the 9.0% of full-year 2025 gross merchandise value that Depop contributed is now, as of this writing, entirely outside the company that reported it. Add Reverb’s 3.2% and 12.2% of that $11.92 billion headline describes businesses Etsy no longer owns at all. Charting that total forward without reading the composition charts a business that changed shape underneath the line twice in fourteen months, and has already finished changing it a second time since this figure was first reported.

The second signature is the direction of travel. Full-year consolidated GMS fell 5.3% against 2024, the Etsy marketplace itself declined about 4% to $10.5 billion, and consolidated revenue rose 2.7% to about $2.89 billion. Volume down, revenue up. The mechanism sits in the take rate, which Etsy defines as consolidated revenue divided by consolidated GMS: 23.3% in the first quarter of 2025, a quarterly high at the time, then 24.9% in the third and 24.5% in the fourth. A team treating volume as the health metric read that year as a decline; the income statement read it as a better year. The reverse case, volume climbing while take rate erodes, is more common, but both are the same failure of letting one number stand in for the economics of the whole.

A Worked Example: Two Conventions, One Quarter

The same trap is available at any scale, and needs no divestiture to spring.

Take a 30-person marketplace selling commercial kitchen equipment to independent restaurants, charging a 9% commission, with shipping billed separately because a combi oven does not go in an envelope.

In a quarter it completes 2,600 orders at an average order value of $1,240. Goods alone: $3,224,000. Shipping averages $75 an order, adding $195,000. Then 190 orders cancel before dispatch, removing $235,600 of goods and $14,250 of shipping.

Run the two conventions. On the exclusive definition, goods only and net of cancellations, the quarter’s figure is $3,224,000 minus $235,600, or $2,988,400. On the inclusive definition, goods plus shipping and net of cancellations, it is $3,419,000 minus $249,850, or $3,169,150. Same quarter, same orders, same cancellations. A spread of $180,750, or 6.0% of the lower figure, produced entirely by a definitional choice.

The commission is charged on goods and shipping together, net of cancellations, so revenue is 9% of $3,169,150, or $285,224. Divide that by the exclusive figure and the take rate reads 9.54%. Divide it by the inclusive figure and it reads 9.00%.

The decision this forces is not cosmetic. A pricing group looking at 9.54% reads half a point of headroom above the advertised 9% and may spend it on a supply-side subsidy. That headroom does not exist; it is an artifact of dividing revenue by a base that excludes part of what the revenue was charged on. The defensible call is to publish the exclusive figure as the headline, disclose the fee base beside it, and compute take rate only against the base the fee was levied on. That reconciliation is a design decision about who pays what, the same territory as pricing a two-sided marketplace. Get the base wrong and every downstream ratio inherits the error.

Is Gross Merchandise Value a Vanity Metric? The Criticism Aims at the Wrong Target

The standard criticism is that the figure is a vanity metric that flatters founders. The criticism is aimed at the wrong target.

A number is not vanity because it is large. It is vanity when nothing decides differently as a result of it. Volume is the correct input for questions about market position, category depth, liquidity, and whether a marketplace has enough transactional gravity for a new fee to be worth building. Etsy’s 2025 disclosures make the case for segmentation better than any argument could: while the consolidated line fell, Depop’s GMS grew 36.3%, Etsy marketplace active buyers declined 3% to 86.5 million, habitual buyers fell 9%, and reactivated buyers rose 4%. Four segments moving in three directions inside one aggregate, each a product finding with a roadmap attached, and each visible only because the volume figure gets cut apart rather than dismissed.

What deserves the criticism is the unaccompanied figure. Reported alone it answers no question a team can act on. Reported with take rate, cancellation rate, and a revenue line beside it, it locates where the business is growing and where it is only moving. The problem was never the metric. It was the three missing columns. The figure earns its place as a scale input and forfeits it the moment it travels alone. This is the same distinction that separates gross expansion from net revenue retention, where a gross figure and a net figure can point in opposite directions in the same period.

Why Analytics and Finance Will Never Agree on the Number

Here is the constraint most teams meet late. Product analytics tools count events; finance counts settlements. The two disagree by construction, and no dashboard resolves it.

An event fires when an order is placed. A refund three weeks later is a separate event, usually landing in a different period, frequently in a different table, and often with no field tying it to the original order. Cancellations before fulfilment may not fire an event at all if the cancellation happens in an ops tool rather than the storefront. Currency is worse: a multi-currency marketplace has to pick a conversion date, and transaction date versus settlement date moves the annual figure on nothing but exchange-rate timing. That is why Etsy publishes a currency-neutral version alongside the reported one, and why the two differ by a percentage point or more in a given quarter.

Treat the figure the way a surveyor treats an unregistered boundary line: usable for orientation, worthless as a basis for building. A public company with an audit committee still publishes one quarter’s growth on three bases. A Series A marketplace running the calculation in a spreadsheet has no stronger claim and fewer controls. Whether the underlying orders are even real is a different discipline again: collusive or fraudulent transactions inflate the number perfectly, which is one reason marketplace trust and safety work shows up in metric integrity, not only user experience.

Five Things to Do Before Gross Merchandise Value Leaves Your Team

The failure that damages a PM’s credibility is not misdefining gross merchandise value. It is presenting the number to someone who owns the income statement without pre-empting the questions they will ask.

So, before it leaves the team: first, write the definition in one sentence, naming whether shipping, taxes, and cancelled orders are in or out. Second, state the settlement basis, booking date or completion date, because one is a leading indicator and the other lagging, and they are not interchangeable in a forecast. Third, put the take rate beside it, computed on the fee base rather than the headline. Fourth, name what changed since it was last presented, including methodology changes and any entity that entered or left the perimeter, and restate the prior period if the method moved. Fifth, segment it by category, cohort, or geography, because an unsegmented total hides exactly the concentration that makes it actionable.

Those five map almost item for item onto what the SEC expects of a disclosed metric, which is not a coincidence. A PM who does them has handed over a reconciliation rather than a headline, and a reconciliation survives contact with a CFO. It also survives the review itself, which is the difference between a quarterly metrics review that changes the roadmap and one that applauds whichever line went up. The discipline separates a marketplace product manager from a B2B SaaS PM more reliably than any org chart does, because a SaaS PM rarely has to defend the denominator. Friction sitting underneath the volume line usually reads as an activation problem rather than a demand one.

The elasticity is getting more consequential, not less, as marketplaces bolt on advertising, logistics, payments, and lending, each with its own base. Etsy’s revenue growth in a down-volume year came substantially from ads, a line with no natural denominator in transaction volume at all. Every added revenue line widens the distance between what a platform transacts and what it earns, so the reconciliation a team writes this quarter is the one it will defend against a much more complicated income statement in two years.

References

• Etsy, Inc. — “Etsy, Inc. Reports Fourth Quarter and Full Year 2025 Results,” 19 February 2026, fetched 3 August 2026 — https://investors.etsy.com/news-events/press-releases/detail/218/etsy-inc-reports-fourth-quarter-and-full-year-2025-results
• U.S. Securities and Exchange Commission — Etsy, Inc. Form 8-K, Exhibit 99.1, Q3 2025 results, fetched 3 August 2026 — https://www.sec.gov/Archives/edgar/data/1370637/000137063725000098/exhibit991q32025.htm
• U.S. Securities and Exchange Commission — “Commission Guidance on Management’s Discussion and Analysis,” Release Nos. 33-10751; 34-88094, issued 30 January 2020, effective 25 February 2020, fetched 3 August 2026 — https://www.sec.gov/rules-regulations/2020/01/commission-guidance-managements-discussion-analysis-financial-condition-results-operations
• Etsy, Inc. — “Etsy Completes Sale of Depop,” press release, 30 July 2026, on the completed $1.4 billion cash transaction, fetched 9 August 2026 — https://investors.etsy.com/news-events/press-releases/detail/224/etsy-completes-sale-of-depop