Last updated: August 20, 2026. This update corrects five sourcing errors that had been live since May, repoints the Sean Ellis and Mark Suster citations to their primary sources after the Suster link went dead, and adds Carta's Q2 2026 data on how pre-seed capital is concentrating.
Pre-revenue traction is measurable. Seed investors expect proof from one of three places: usage data with a retention curve, paid pilots or letters of intent with named counterparties, or founder-led sales conversations that produce signed commitments. Without revenue, the bar shifts to signal density and quality. This guide explains the signals investors weight, the "traction" categories they ignore, and how founders without revenue make the case anyway.
The proof problem when there is no revenue
A pre-revenue founder is asking an investor to bet on claims the market has not yet priced. Revenue is the cleanest pricing signal in early-stage venture. When it is absent, every other claim sits under more pressure. Investors know this. They are doing first-principles bet-making on whatever proof exists.
Most pre-revenue founders bring the wrong proof. A 20-slide deck with five hypothetical customers, a five-year financial model, and a market-sizing slide built from a TAM number does not carry investor conviction in 2026. None of those artifacts answers the question the investor is asking: is this founder, building this thing, in this market, real?
The bar has also moved. Forum Ventures surveyed 150 active North American VCs and analyzed over 300 B2B SaaS pre-seed and seed deals closed between January and October 2024. Its finding on the seed bar is blunt: $300k to $500k ARR is now seen as the standard, against roughly $200k in 2023. That is a revenue benchmark, and a revenue benchmark is the one thing a pre-revenue founder cannot produce. Waveup asked 56 VCs the same year and found round expectations jumped one full stage: pre-seed now expects seed traction and seed plays by Series A rules. The gap has to be closed with other proof.
What investors actually count as traction
Traction is a set of signals, each of which tells the investor something different about whether the founder's claims about the market, the product, and themselves are real. At seed in 2026, the signals that move conviction usually fall into seven categories.
The pre-revenue traction signals table
Signal type | What investors look for | What does not count | Why it moves conviction |
|---|---|---|---|
Engaged active users | Daily or weekly active users with a stable or growing retention curve over 8+ weeks | Total signups, beta waitlists, Product Hunt vanity spikes | Retention is the hardest signal to fake. A flat or rising curve at 8+ weeks signals the product is solving a real problem. |
Paid pilots | A small number of named counterparties paying any amount for a defined scope of work | Free pilots, design partner promises, generic LOIs | Money changes hands only when buyer pain is real. Even a $2K pilot is more signal than a $200K LOI. |
Letters of intent (LOIs) | LOIs from named buyers, signed by a real decision-maker, with quantified scope and timing | Unsigned MOUs, "interested" emails, LinkedIn DM enthusiasm | LOIs only count when they specify what would convert to revenue and by when. Without specifics, investors discount them to zero. |
Organic acquisition | Users finding the product without paid ads, with a measurable CAC of ~$0 or qualitative referral evidence | Press mentions, viral one-week spikes, paid ad results | Organic acquisition with no funnel investment indicates word-of-mouth at the user level, which is the cheapest acquisition channel that scales. |
Depth metrics | Time-in-product, repeat usage frequency, qualitative "I'd be very disappointed if this went away" survey data | Total time, raw session count without retention context | Depth signals product-market fit before revenue scales. Sean Ellis's 40% rule was built for this stage. |
Founder-led sales | Direct evidence the founder is closing buyers personally, with a written record of conversations | "I have a strong network" claims without close attempts | The Suster line vs dot principle applies here. Closed conversations are dots; a 6-month conversation log is a line. |
Domain founder-market fit | Founder credentials, prior work, or specific lived experience that uniquely qualifies them for this problem | A polished LinkedIn profile, generic "I'm passionate about" claims | Founders cited as an important factor by 95% of VC firms in the Gompers et al. survey of 885 institutional VCs (Journal of Financial Economics 2020). |
The seven categories are not a checklist. No founder hits all seven at pre-revenue. The question investors are answering is: how many of these are present, and how strong is each one? Two strong signals beat seven weak ones every time. The strongest single signal at pre-revenue is usually retention plus founder-led sales conversation depth.
Vanity metrics investors ignore
Investors discount most vanity metrics on sight. The signals dismissed in meetings include total signups without retention, beta waitlist sizes (the inverse of conversion), LinkedIn followers, Twitter likes on launch posts, press coverage, "viral" Product Hunt launches without retention follow-up, and TAM-as-traction math ("we only need 0.1% of a $100B market"). The CB Insights post-mortem analysis of VC-backed failures identifies "no market need" as the most common root cause cited by failed founders, often after building on top of one of these vanity signals.
The team-as-traction shortcut
When usage and revenue are thin, the founder becomes the strongest proof asset. The investor is betting on the founder's ability to learn, iterate, and close customers fast enough to outrun cash burn. This is why the team slide carries disproportionate weight at seed. The HBR summary of the Gompers et al. survey documented that VCs source over 30% of deals through professional networks and another 20% through referrals from other investors, while only 10% arrive inbound from company management. The same survey clocks the average deal at 83 days to close, with the average firm spending 118 hours on diligence and calling 10 references. The founder's network feeds the funnel before any deck does.
The DocSend 2024 Funding Divide Report, based on deck-engagement telemetry across thousands of pitches, found that the Team slide was among the slides where investor attention grew at both pre-seed and seed rounds in 2024. As paper traction has thinned out, founder evaluation has thickened.
The founder credibility asymmetry is real and well-documented. Gompers, Kovner, Lerner and Scharfstein's study of performance persistence in entrepreneurship found that a founder who succeeded before has a 30% success rate on the next company, a founder who failed before 20%, and a first-time founder 18%, across several thousand venture-backed companies from 1986 to 2003 with success defined as going public or filing to. Repeat founders also negotiate better terms: more retained equity, more board control, less dilution per round. Serial founders with prior exits can close $10 million seeds in under a week.
For first-time founders, the asymmetry is structural and it is closeable. The way to close the gap is to surface specific founder-market fit that does not depend on a prior exit:
Direct buyer-side experience. A founder selling to compliance officers who spent 7 years as a compliance officer is harder to discount than a generalist.
Technical credibility. A research-backed AI infrastructure founder with peer-reviewed papers carries different weight than a wrapper-builder.
Repeat customer relationships. A founder selling into a buyer network they built at a prior role can close pilots in the first month of company existence.
Operational pattern recognition. A founder who has scaled the specific function they are now building software for usually understands the buyer's pain better than the buyer can articulate.
None of these requires a prior exit. All of them are demonstrable in a 15-minute conversation if the founder knows how to surface them. Most founders do not. The result is the investor walks away with a generic team impression and prices the round at the floor.
The PMF question pre-revenue founders must answer
Product-market fit is the most over-used phrase in early-stage fundraising. Investors mention it on every call. Founders claim it after one cohort with 60% retention. The reality is harsh.
The Sean Ellis 40% rule, still the cleanest founder-side proxy for PMF, requires that at least 40% of surveyed users say they would be "very disappointed" if they could no longer use the product. Ellis arrived at the threshold after benchmarking nearly a hundred startups with the same survey. Most companies never clear it, and the graduation data shows what that costs. Carta's numbers put seed-to-Series-A graduation at 25% to 30% within 24 months in a normal year like 2018, and at roughly 17% for the 2022 seed cohort over its first two years. The Series A bar in 2026 is a real product-market fit signal plus 12 to 18 months of clean monthly growth.
For pre-revenue founders, the implication is that PMF claims have to be supported by the right kind of evidence. The conversation that moves investors sounds like this: "here is the cohort data and survey methodology that suggests we are approaching PMF, here is what is missing, and here is the experiment we are running to close that gap in the next 60 days." Investors take that founder seriously. They roll their eyes at the founder who declares PMF on the basis of a 50-person beta with 30% retention.
The repetition trap that punishes pre-revenue founders most
Pre-revenue conviction does not transfer between investors without artifacts. Each fund starts the diligence process from zero. The founder explains the company from scratch, walks through the same 12 questions, defends the same model assumptions, and clarifies the same misreads of the deck. Twenty meetings later, the founder has answered the same questions 20 times. The signal has not compounded. The fatigue has. The Carta bridge-rounds Q2 2025 data shows 16.6% of total cash raised in Q2 2025 came via bridge rounds, up from 11.8% a year earlier. A meaningful share of seed-stage cohorts now run a second extension before reaching a priced Series A, which extends the repetition cycle further.
The meeting-level evidence in Fu and Taylor's NBER study of roughly 21,000 US venture deals used cell phone location data to detect in-person diligence meetings and found the measure sitting at zero in 95% of deals. Where diligence is detectable at all, competition compresses it. One standard deviation more abnormal deal volume in a startup's sector is associated with 15% to 34% fewer diligence hours, and a doubling of the number of other VCs meeting the startup each month is associated with 13% fewer. The deal flow is too high for any one investor to do deep work on most opportunities. Investors lean on signals from other investors, prior founders, and pattern matching. The Venture Capital journal signaling-theory review, a systematic literature review on how signals shape funding decisions, maps the full ecosystem of inherited signals investors lean on when first-principles diligence is too expensive.
This compounds the pre-revenue problem. A founder with thin signals walks into a 30-minute meeting where the investor is making a snap judgment partly on the live conversation and partly on signals from earlier investors. If the earlier signals were weak, the new investor inherits that weakness. If the founder cannot demonstrate a clearly compounding line of progress, the meeting closes flat. The structural fix is the same: reduce the information asymmetry the investor is working against by presenting proof artifacts that travel with the founder between funds.
The Mark Suster line-not-dots principle applies precisely here. As Suster wrote in his canonical investor essay Invest in Lines, Not Dots, an investor watching a founder over multiple meetings sees the slope of progress rather than a single data point. The pre-revenue founder who can show a line, even a steep one starting from a low base, beats the founder who shows a single high dot with no context. The artifact a founder needs is the trajectory, with the current state as one point on it.
The proof layer
What every pre-revenue founder needs is a way to make their traction case once, in their own voice, and have it travel without them. The signals are real. The conviction comes from how they are assembled. Most founders assemble them slide by slide across many separate live meetings, and the assembly starts again from zero with every new fund.
SeedForge produces a Living Profile for the founder, built from one 30-minute AI session that asks the founder the same questions an investor would ask in the first three meetings. The session is voice-based and conversational. The founder walks through what they are building, why they are building it now, what counts as traction so far, what does not yet exist, and what the next 90 days look like. The system structures the answers into a profile that includes traction breakdowns by category, founder-market-fit context, the specific PMF evidence the founder can defend, and the open questions the founder is actively running experiments on. The profile lives at a SeedForge Link the founder shares with any investor. Investors arriving at the link see the same proof picture every time, in the founder's voice, before the first call. What that is designed to buy the founder is a first call that opens one level deeper than a cold introduction, with the repeated groundwork already done in writing.
Pre-revenue traction proof checklist
Before sending the first investor email, a pre-revenue founder should be able to answer all of the following with specifics. Several will be a no at this stage. A specific no is a real answer: name the gap and name the experiment that closes it. The generic answer is what loses the room.
Which two traction categories are your strongest? Pick from the seven in the table above. Have a one-paragraph answer for each, with numbers and time frames.
What is your retention curve at 4, 8, and 12 weeks? If you cannot produce this, your usage data is not yet at the point where it carries weight.
What is your most recent paid pilot or signed LOI? Counterparty name, scope, dollar amount, signing date. Generic enthusiasm does not count.
What is your strongest piece of founder-market fit evidence? Not a LinkedIn profile. A specific story, a specific connection, a specific qualification that makes you the right person for this problem.
What is your Sean Ellis 40% rule survey result, with sample size? If you have not run this survey, run it before the raise.
Who are your three closest reference customers? People investors can call. Names and titles.
What is your 90-day proof plan? What experiments will you run between now and the close of this round to add signal to the picture?
What does your line look like, beyond today's dot? Mark Suster's frame. Investors want to see slope.
What is your founder-led sales record? Not "I have a network." A list of specific conversations with specific buyers and specific outcomes.
What is your "I would be very disappointed" rate, and what does it say about PMF? Numbers, sample size, and your own interpretation.
A founder who can defend all 10 points walks into a seed conversation with the right proof posture even at zero revenue. A founder who can defend fewer than 5, with no credible plan for the rest, will struggle to convert meetings into term sheets regardless of how polished the deck is.
What changed at pre-seed in 2026
The pre-seed market is concentrating, and that changes what a pre-revenue founder is up against. Carta's State of Pre-Seed for Q2 2026, published on 13 August 2026, reports that US startups on Carta raised $3.19 billion across more than 11,500 pre-seed instruments in the quarter, against $3.22 billion across 14,825 instruments in the same quarter a year earlier. Carta notes that the Q2 2026 instrument count is provisional and will be revised upward as more data lands, so the size of the gap is directional. The direction is the point: close to the same money, spread across materially fewer instruments.
The average instrument size makes the same point with a firmer number. Carta puts it at $276,000 in Q2 2026, a 27% rise year over year and the highest in more than four years. Carta also reports AI startups taking 49% of all pre-seed dollars in the first half of 2026, close to their 50% share for full-year 2025.
For a pre-revenue founder this matters more than any valuation benchmark. Bigger cheques spread across fewer companies means every individual decision carries more weight for the investor writing it, so the screening bar rises with the cheque size. That sits alongside the screening compression described in the next section rather than against it. AI lets an investor look at more companies; concentration means fewer of those looks end in a cheque. Both forces push the same way for a founder with no revenue, who is asking an investor to commit more capital against less priced evidence.
What pre-revenue founders can learn from the AI tooling shift
Investor diligence is evolving faster than founder fundraising materials are. The Affinity AI in VC research, based on a survey of nearly 300 private capital dealmakers, found that 85% of dealmakers now use AI to automate daily tasks, up from 76% the prior year. The screening step is being compressed. The conviction step is not. AI lets investors screen more companies faster. It does not change what they need to see to write a check. If anything, it raises the bar on what counts as "interesting enough to talk to" because the top-of-funnel is now wider.
For pre-revenue founders, this means two things. First, the screening step is no longer where rejection happens slowly. Investors decide within minutes whether to dig deeper. Second, structured proof artifacts an investor's AI can parse (specific numbers, names, time frames) move founders through screening faster than narrative materials. The SVB State of the Markets H1 2026 report frames the same shift: deal velocity has returned, and clean structured proof beats hidden numbers behind narrative.
The PitchBook-NVCA Venture Monitor reported full-year 2025 US VC fundraising at $66.1 billion, the lowest since 2018, with AI and machine learning capturing 65.6% of all VC deal value. The capital backdrop means seed investors are pricing for capital efficiency. Demonstrable unit economics and a defensible go-to-market plan are now baseline asks even at pre-seed.
How this compounds with the broader seed signal stack
Investors evaluating a pre-revenue startup weigh traction against team, market, product, and capital efficiency simultaneously. The SeedForge guide to what investors look for in a startup at seed walks through the five-signal framework: founder credibility, market choice, product wedge, traction signal, and cap-table cleanliness. A pre-revenue founder with a 30% repeat-founder track record and a deep buyer-side connection can raise on team alone. A first-time founder in a generic market with thin signal usually cannot.
Valuation is downstream of all of this. The SeedForge breakdown of seed stage startup valuation explains how investors anchor to comp data and adjust up or down based on the proof picture; pre-revenue founders who get priced usually land at the floor of the bracket. The SeedForge due diligence checklist for seed-stage startups lists the artifacts investors will ask for; pre-revenue founders who prepare those artifacts in advance close faster.
Frequently asked questions
What counts as traction when you have zero revenue?
Pre-revenue traction is real signal that the market is responding to the product. Engaged active users with a retention curve at 8 or more weeks, paid pilots from named counterparties (any dollar amount), signed LOIs with quantified scope and timing, organic acquisition with measurable referral evidence, and depth metrics like time-in-product or Sean Ellis 40% rule survey results all count. Vanity metrics like total signups, waitlist size, or press coverage do not.
How much traction do investors expect at pre-seed in 2026?
The bar has moved up. Forum Ventures, surveying 150 North American VCs and over 300 B2B SaaS pre-seed and seed deals closed between January and October 2024, found the seed standard has risen to $300k to $500k ARR from roughly $200k in 2023. At pre-seed, where revenue is usually zero, that pressure lands on everything else. In practice this means a working product with at least 8 to 12 weeks of usage data, a clear retention story, and one or two named buyer relationships or paid pilots. Pure idea-stage pre-seed rounds still exist but are increasingly rare outside of repeat-founder deals.
Can a pre-revenue startup raise a seed round?
Yes, but the bar is steeper than at pre-seed. Pre-revenue seed rounds usually require a combination of strong founder-market fit, a clear product wedge with usage evidence, and one or more paid pilots or signed LOIs. The Carta State of Private Markets Q3 2025 data shows the median seed pre-money valuation at $16 million, but pre-revenue founders typically price at the floor of the comp bracket unless other signals are exceptional.
What metrics matter most for pre-revenue B2B SaaS?
For pre-revenue B2B SaaS, the highest-signal metrics are: paid pilot count and dollar value, deep usage by named buyers, retention at 8 to 12 weeks among target-buyer accounts, sales conversation count and conversion to demo, and qualitative buyer testimonials with named titles. Investors will discount any metric without a named counterparty attached.
Do letters of intent count as traction?
LOIs count when they are signed by a real decision-maker, quantify the scope of what would convert to revenue, and specify a time frame for conversion. Generic "interested" emails or unsigned MOUs do not count. Investors look for LOIs that read like the first 80% of a contract, not a marketing one-pager.
How long should I gather traction before raising?
Long enough to show a line rather than a single dot. The Mark Suster principle applies. Most pre-revenue founders need 4 to 8 months of consistent product and customer activity before the traction picture supports a seed pitch. The exception is repeat founders with prior exits, who can raise on team signal alone in weeks. Everyone else benefits from a slower start and stronger first impression.