Traction Metrics That Matter to Seed Investors

Seed investors are grading on a different curve than they were in 2021, and most founders haven't gotten the memo. CRV's data shows median Series A revenue in 2025 ran roughly 75 percent higher than it did in 2021. The quiet bar for "ready to raise" has climbed right along with it. Capital efficiency and a visible path to profitability have taken over as the main filter investors use, pushing raw growth rate down the list, at least for companies outside the AI mega-deal category, which CRV's 2026 KPI guide treats as its own animal with its own math around inference costs, gross margin, and usage depth.

Founders, meanwhile, are often still showing up with the 2021 scoreboard: signups, total users, a nice spike from launch week. Those numbers feel like proof of life. To an investor reading cohort data for a living, they're closer to a pulse check than a diagnosis. CRV's seed guide, published March 1, 2026, puts it bluntly: seed rounds now look like what Series A used to look like, and by the time a founder is pitching seed, the question of whether the problem is worth solving is already assumed to be settled. What investors want instead is proof that customers use the product, pay for it, and stick around.

That's the gap this piece is built to close. The metrics that actually move a seed check are measuring something founders aren't used to measuring: not activity, but durability.

Retention as the primary signal investors use to distinguish real traction from a launch spike

If there's one number that separates a fundable company from a well-marketed one, it's retention, and it isn't close. Retention is the hardest metric to fake, because no amount of ad spend or clever landing page copy makes someone open the app in week eight if the product isn't solving something real for them. Signups can be bought. Retention has to be earned, one return visit at a time.

CRV's KPI guide names the specific shape investors look for: the "smile curve." Usage dips after the initial signup rush, as it always does, but then it stabilizes and holds flat. That flattening is the signal that the product has found people who actually need it. A company with a small user base showing flat retention at week eight will beat a company with a much bigger signup number and a cliff at week two, every time, because the first company has evidence of a reason to exist and the second has evidence of a marketing budget.

This is why investors ask to see cohorts plotted out over weeks and months rather than a single tidy retention percentage on a slide. A single number can hide a lot of sins. A cohort chart can't. It shows, visually, whether each new batch of users behaves like the last one, gets better, or gets worse, and that pattern over time tells a much richer story than any one-time snapshot.

Gross Revenue Retention (GRR) plays a supporting role here for companies that do have revenue. GRR strips out expansion revenue and caps at 100 percent, measuring pure churn prevention, the floor beneath the more commonly quoted net revenue retention number. CRV reports median GRR for private B2B SaaS companies is in the high-80s to low-90s percent range, with the stronger performers up in the mid-to-high 90s. Higher churn is tolerated at seed. What isn't tolerated is a churn line moving in the wrong direction.

Founders without revenue yet sometimes assume this entire section doesn't apply to them, but it applies just with a different data source. CRV specifies weekly active users and feature adoption rates as the engagement proxies investors use in place of NRR for B2B products that haven't started charging. The question investors are actually asking, whether someone has revenue yet or not, stays exactly the same: do customers depend on this, or did they just try it once?

Revenue quality signals investors weigh before they look at total ARR

Once there's revenue on the table, the instinct is to lead with the biggest number available. Investors read past that number almost immediately and start asking how it was built. A smaller ARR figure built on diversified, multi-year contracts is more fundable than a bigger ARR figure propped up by one customer on a month-to-month deal, because the second company's entire business can walk out the door with a single cancellation email.

CRV's KPI guide lays out what "revenue quality" actually means in practice: how concentrated the customer base is, whether contracts run annual or multi-year versus month-to-month, and how much of the revenue comes from renewals versus brand-new logos. Annual and multi-year contracts reduce volatility. Month-to-month arrangements don't, no matter how much revenue they generate this quarter.

Customer concentration deserves particular attention, because two companies can post the exact same ARR and carry wildly different risk. A business with one account representing a large chunk of total revenue is one bad renewal conversation away from a very different-looking year. A business with that same ARR spread across a diversified base doesn't have that single point of failure. The headline numbers look identical. The underlying risk does not.

None of this is about chasing a specific ARR threshold to unlock a meeting. Retention tells an investor the product sticks. Revenue quality tells them the business built around that retention won't crack under its own structure.

Burn Multiple and CAC payback as the capital-efficiency tests that now filter out most seed pitches

Burn Multiple, a framework popularized by one well-known investor, answers a simple question: how much cash does a company burn to generate each new dollar of annual recurring revenue? CAC payback asks a related question: how many months does it take for a new customer to pay back what it cost to acquire them? Together, these two numbers tell an investor how efficiently a founder turns capital into durable revenue, and they now carry as much weight in a seed pitch as the growth rate itself. A strong topline with a bad Burn Multiple doesn't get waved through anymore. It gets a follow-up question.

Early-stage companies naturally run higher burn multiples, since fixed costs get spread across a smaller revenue base, and investors expect that at seed. What they're watching for is the trendline. A burn multiple that's improving as revenue grows tells a very different story than one that's quietly getting worse, even if both companies show the exact same number on the day of the pitch.

The objection that comes up constantly is some version of: "We can't know lifetime value yet, we haven't been around long enough." That's fair, and investors aren't asking for a number that doesn't exist yet. What they're asking for is directional evidence: a short CAC payback period, NRR that's climbing instead of flattening, churn that's trending down. An investor can combine those three into a credible LTV thesis without needing three years of data they can't possibly have. The case being made at seed is probabilistic. Nobody expects a founder to produce a number that requires time travel.

What Pre-Revenue and Early-Stage Founders Can Offer

Not having ARR doesn't exempt a founder from the traction standard; the evidence looks different, while the underlying question stays the same: is there real demand out there that isn't manufactured by friends and family doing a favor?

CRV's seed guide is explicit on this point, almost to the point of being blunt about it. Customers who are friends, former colleagues, or family members don't prove the value proposition works for the actual addressable market, because they were always going to say yes. Organic referral activity matters so much to investors precisely because it's the one thing a paid acquisition channel can't buy. Nobody refers a product to a friend because an ad told them to.

Without revenue, waitlist conversion rates and Letters of Intent become the stand-ins. CRV notes these matter because they show that target customers have enough conviction to actually take an action, not just click a button out of curiosity. A strong set of LOIs from real target buyers is worth more to investors than a sprawling email list that ballooned overnight because a video happened to go viral. One of those lists represents people who said "I will buy this." The other represents people who said "that was a fun fifteen seconds."

Founder-written build updates sent to a tight list of ideal customers, and the open rates they generate, work the same way. A high open rate on those updates means the audience is paying attention to the problem being solved, not just a brand they clicked once and forgot.

CRV also names something less tangible: learning velocity. Investors want to see whether a founding team can spot its own gaps and close them fast enough to reach product-market fit before the runway runs out. This is the qualitative stand-in for every financial metric a pre-revenue company doesn't have yet, and it answers the same underlying question the retention and revenue sections do. Growth, even tiny growth from a small base, moving steadily upward carries more weight with investors than a bigger number that's gone flat. Direction is the argument, at every stage, with or without a revenue line to point to.

Metrics Founders Most Commonly Over-Report

Most founders are reaching for the numbers that feel like proof, without realizing those numbers read very differently from the investor's side of the table.

Total registered signups, reported without any retention context, tell investors almost nothing by themselves. CRV's KPI guide draws the comparison directly: a big signup number with a retention cliff by week two loses to a small user base with flat retention at week eight. Signups measure how good the funnel is at the top. Investors are trying to figure out what happens at the bottom.

A large email list or waitlist, shown without open rates, reply rates, or any sign of actual usage, tells investors the founder built an audience, not that anyone wants the product. An audience can be purchased with a good enough ad. Real demand can't be bought. That is why investors weight the two so differently.

Revenue concentrated in one or two logos raises a flag even when the total ARR number looks impressive. CRV points out that a company with a single customer representing 40 percent of ARR carries a fundamentally different renewal risk than a company with that same revenue spread across a diversified base. The size of the number doesn't change. The risk changes instead.

Gross revenue reported as a single lump figure, without a breakdown of new, expansion, and churned revenue, hides the one thing investors most want to see. A flat MRR number could mean a stable, loyal base, or it could mean a leaky bucket where new signings are just replacing customers who quietly left. Investors ask for the MRR composition broken into those three lines for a reason: the single number can't tell them which story they're looking at.

How to build toward investor-ready metrics from the day marketing starts, not the month before you pitch

Every metric covered so far, retention depth, NRR, Burn Multiple, diversified organic demand, comes out of how a company builds its marketing from the start. None of it can be assembled the week before a pitch from whatever data happens to be lying around.

ICP clarity has to come before channel selection, because spending on acquisition before understanding who actually sticks around, and why, drives CAC up and drags the Burn Multiple down with it, the two efficiency numbers getting the most scrutiny at seed right now. Plenty of seed-stage teams spend a full year running campaigns before they figure out which motion even works, and that year appears later as a messy, unconvincing metrics slide.

Founder-led outreach tends to be the right first motion because it generates the exact behavioral data that later becomes the retention and organic-demand story investors want: which messages get a response, which types of buyers convert, which customers go on to refer someone else.

Owned channels build on that foundation. An email list with real, measurable engagement and a content program that compounds organic traffic over time both produce evidence of demand that nobody paid for, which is precisely the kind of signal CRV flags as something paid channels can't manufacture.

None of this works without measurement infrastructure in place from day one. CAC, pipeline velocity, and MQL-to-customer conversion need to be tracked starting with the very first campaign, not reconstructed later from a lump-sum spend total on a P&L. Investors find a trendline over six or more months convincing precisely because it can't be faked after the fact. A founder standing in front of investors with that kind of system already running isn't scrambling to assemble a story. The numbers have been telling it the whole time.