Every November and December, a specific pattern repeats across the Indian marketplace ecosystem. Founders who have been building for twelve to eighteen months start preparing seed conversations for the January-to-March fundraising window. The decks come out, the metrics tabs get built, the GMV charts get polished. And then, in the first serious investor meeting, the conversation goes somewhere the founder was not prepared for, because the questions are not about the metrics the founder brought to the room.
The gap is not about preparation. It is about which numbers the founder treated as the headline, and which ones they treated as a footnote. For most pre-seed and seed marketplace rounds in India, the headline numbers founders bring (gross merchandise value, take-rate, user growth) are the ones investors already assume they will see, and spend the least time on. The numbers investors actually evaluate the business against are different, more specific, and genuinely harder to produce honestly.
This piece names those three numbers. It is written for the founder preparing for seed conversations now, who wants to walk into the room with the right data instead of the data they already have, and who would rather hear the critique in writing than in a term sheet negotiation.
Why marketplace unit economics are their own category
Before the three numbers, one framing move. The reason founders arrive at seed conversations with the wrong headline metrics is not that they do not understand their business. It is that the templates they have been given were built for SaaS or D2C, and marketplace unit economics do not actually fit either template.
A SaaS business sells a product to one party, prices it on a recurring basis, and the unit economics question is largely about customer acquisition cost against lifetime value against gross margin. A D2C business sells a product to a consumer, prices it on a transaction basis, and the question is largely about contribution per order against repeat purchase rate. A marketplace is neither. It facilitates a transaction between two parties, takes a slice, and the economics question depends on whether both sides of the market are being served well enough that the transactions keep happening.
The investor evaluating a seed marketplace is not asking "is this product good." They are asking three different questions, in sequence, and the three numbers below map directly to each of them.
Number 1 · Contribution margin per transaction (not take-rate, not GMV)
The first question is: when the marketplace facilitates a transaction, does the economics actually work? Not at some future scale with assumptions, but today, honestly, on the transactions you actually facilitated this quarter.
Founders bring take-rate to this question, because take-rate is easy to compute and sounds like an answer. It is not. A take-rate of 12 percent on a transaction tells the investor nothing about whether the business keeps money after it has paid payment processing, fraud losses, customer support, dispute resolution, platform infrastructure, and the subsidies the marketplace is paying to make the transaction happen at all. For most seed-stage marketplaces, the honest contribution margin per transaction, after all of these, is a fraction of the take-rate, and often negative.
The number investors want is contribution margin per transaction, defined specifically: the take-rate the marketplace actually captures, minus every variable cost directly attributable to that transaction. Payment gateway fees. Transaction-level customer support cost. Dispute and refund provisions. Fraud losses. Any discount, cashback or fulfilment subsidy that was offered to make the specific transaction close. In some marketplaces, platform infrastructure cost allocated per transaction, if it is a meaningful fraction.
The honest version of this number tells the investor whether the business works at the transaction level. Most founders, in their first attempt, compute it incorrectly in the direction of their own optimism. Three places to be especially careful:
- Customer acquisition cost is not a variable cost per transaction. It is a cohort investment. Do not include it in contribution margin, and do not let an investor pressure you into treating it as a transaction cost; that is a different conversation, below.
- Subsidies have to be counted at their actual value, not at the value you hope they are. A 10 percent discount offered to close a transaction costs 10 percent of GMV, not 10 percent of take-rate. This single mistake has made more marketplace unit economics look better than they are than any other.
- Variable support cost per transaction is often invisible until you measure it. If every transaction generates one customer service touch at a true cost of two hundred rupees, and your take-rate on a two-thousand-rupee transaction is 240 rupees, your contribution margin is 20 rupees, not 240.
The investor who sees an honest contribution margin per transaction, even a small or negative one, walks away respecting the founder. The investor who catches a dishonestly inflated version walks away not respecting the founder, and the news travels.
Number 2 · Repeat GMV share (the compound growth signal)
The second question investors ask, after they have decided the transaction-level economics are honest, is: when a user transacts once, do they come back? This is the question that separates a marketplace with compounding growth from a marketplace with a leaky-bucket acquisition problem, and it is the question that most founders answer with the wrong metric.
The metric that gets presented is total user growth or active users. Both are flow metrics, both can be bought with marketing spend, and neither tells the investor what they actually want to know. The metric that answers the question is repeat GMV share: the percentage of this period's GMV generated by users who also transacted in a prior defined period.
A simple version: of all the GMV your marketplace generated in October, what percentage came from users who had already transacted at least once in the three months before October? If that number is going up over time, you have compound growth and the acquisition cost is a durable investment. If that number is flat or going down, every new user you acquire is replacing a user you lost, and the marketplace is a leaky bucket.
For a seed-stage marketplace, the honest version of this number typically starts low (because the user base is small and new) and rises over the first twelve to twenty-four months as cohorts mature. The direction matters more than the absolute level at seed. An investor who sees a cohort retention curve where each successive cohort is retaining better than the one before it does not need the absolute number to be high; they need to see that the system is learning.
Three practical disciplines on this number:
- Define the "prior period" explicitly before computing it, and keep the definition constant across every report. If you quietly change from "transacted in prior six months" to "transacted in prior three months" to make the current period look better, a serious investor will see it, and will discount everything else in your deck accordingly.
- Report cohort retention curves, not just the headline percentage. The curve shape, how retention decays over months one to twelve for a cohort, is the real signal. The single headline number collapses the shape.
- Separate transactional repeat from engagement repeat. A user who logs in without transacting is engagement, not GMV. Only transactions count for this metric, and the investor who matters will press on this distinction.
Number 3 · Liquidity (the two-sided matching evidence)
The third question is the one that specifically applies to marketplaces and not to any other kind of business, and it is the one founders are least prepared for because it is the one no SaaS or D2C template has taught them to compute. The question is: of the supply you have listed on your marketplace, what percentage actually transacted in a defined recent window?
This is liquidity. For a product marketplace, it is the percentage of SKUs listed that sold at least one unit in the last thirty days. For a services marketplace, it is the percentage of listed providers who closed at least one booking in the last thirty days. For a B2B sourcing marketplace, it is the percentage of listed vendors who received at least one qualified enquiry that converted to a quote. The specific definition varies by category; the structural question is identical.
Liquidity is the single best test for whether the business is actually a marketplace or just a listings site. A marketplace with high liquidity is one where both sides are being matched; the supply is being discovered, chosen, and transacted. A marketplace with low liquidity has a supply-side problem (the inventory does not match demand), a demand-side problem (the buyers cannot find what they are looking for), or a search-and-matching problem (the two exist but the platform does not surface them to each other).
For a seed-stage marketplace, a liquidity below 10 percent is a red flag, between 10 and 25 percent is normal for pre-product-market-fit, and above 30 percent begins to look like a working marketplace. The investor who looks at this number is asking two things at once: whether the business has crossed the threshold of being a marketplace rather than a directory, and whether the founder knows which side of the market is their bottleneck.
Two framing moves that work in an investor conversation on this number:
- Report liquidity separately by category or vertical if your marketplace has distinct categories. A blended liquidity number can hide a mix where one category is working and three are not. A founder who shows the by-category version and names the bottleneck categories explicitly comes across as operating the business, not just reporting on it.
- Pair the liquidity number with the specific action you are taking on it. If liquidity is 14 percent and you have a specific hypothesis about why, with a specific intervention in flight, that is the conversation an investor wants to have. If liquidity is 14 percent and you have no narrative on it, the conversation stops.
How to present these three numbers in an investor conversation
The three numbers above are the headline. GMV, take-rate and user count are the context. The order matters.
Lead with contribution margin per transaction, honestly computed, and name the methodology before an investor asks. Follow with the cohort retention curve and the trajectory of repeat GMV share across cohorts. Close with liquidity, by category, with the bottleneck named. If you walk a seed investor through this sequence in the first fifteen minutes of a meeting, the rest of the conversation shifts from "convince me this is a real business" to "let's talk about scale and the next milestones." That shift is the whole point.
Three common mistakes to avoid:
- Leading with GMV. GMV is interesting, but it is also easily inflated through subsidies, fake transactions, or counting unmonetised volume. Investors assume this, and start every GMV conversation with suspicion. Starting there puts you on the back foot.
- Reporting any of the three numbers without a methodology note. "Contribution margin is 8 percent" is useless. "Contribution margin, defined as take-rate minus payment processing, fraud, support and subsidies, is 8 percent on an average transaction value of 3,400 rupees, computed on October 2026 transactions" is a conversation.
- Omitting the bad news. If one of the three numbers is weak, name it, quantify it, and explain what you are doing about it. Investors have heard a dishonest version of every marketplace pitch, and the first signal that this one is different is that the founder volunteered the weakness before being asked.
Talk to MagicWorks about your platform
MagicWorks advises platform and marketplace founders on strategy, audits and vendor evaluation. Founder-led, with no execution conflict.
Explore Platform & Marketplace Consultation, or book a discovery call.
Related reading on MagicWorks
- Pre-Launch or Post-Launch? The One Question That Decides If a Platform Roadmap Audit Will Help You · the upstream framework. The founder who can answer the audit question honestly tends to have the three numbers above in better shape too.
- The Platform Founder's Dilemma: When to Bring in Outside Counsel · for founders weighing whether to bring in advisory help during the fundraising cycle, including unit economics stress-testing.
- B2B Sourcing Marketplaces and India's New Data Protection Rules · a compliance-side read that increasingly comes up in investor diligence, especially for marketplaces processing significant personal data.
Swapnil Ughade is the Founder and Managing Director of MagicWorks IT Solutions Pvt. Ltd., an AI-first digital marketing agency based in Pune. Founded in 2009 and now in its seventeenth year, MagicWorks runs its Platform and Marketplace Consultation practice on a strict no-execution discipline: the firm advises, audits, and evaluates vendors for platform businesses across edtech, B2B sourcing, wellness and D2C, but does not build the platforms it advises on. Swapnil leads the practice personally and has 17+ years of experience across digital marketing, web development and AI consultation.




