This is the last article in the series, and it lands on the dimension that matters most by the time a company is raising a serious round. It is also the one founders most consistently fake themselves out on, because the word traction has been stretched to cover everything from signed contracts to people who clicked a link, and those are not the same, and any investor knows it. That is why this piece starts with the distinction that matters most.
This article is one in a series. The series discusses the method we use to judge whether a business plan is fundable. We have developed a multipart weighted scorecard, based partly on research into best practices and partly on my experience in the startup and venture space: five startups, two exits, startup investing, and founding a startup accelerator. There will be one article for each scoring category, plus one on red flags. That context matters because the next sections focus on one category at a time.
There’s a selfish reason for doing it in the open: we are building the methodology into a managed AI agent that evaluates plans against real investor decisions, and writing each part out is how we find where it’s wrong. If you’re raising funding for a startup, you get the rubric we would use to grade you. If you think we have weighted something badly, tell us. That’s the most useful note we can get, and it is why we want the critique now.
Traction and evidence of demand carry the heaviest weight in our scorecard, tied with the team, and unlike the team, its weight climbs steeply with stage. At pre-seed there is barely any traction to judge, and weighting it heavily would be a category error, the same mistake we warn about throughout this series. By Series A the calculus has inverted, and traction becomes the proof the reader expects to see; “great team, early days” has stopped being enough. So more than any other dimension, what counts here depends on how much money you are asking for, and that shift drives the rest of this discussion.
The proof hierarchy
This is the most developed part of our rubric, and the part I am most settled on, so I’ll lay out the ladder. Every form of so-called traction sits somewhere on it, and the higher the rung, the more it counts, because the higher rungs are harder to fake and closer to the only thing that ultimately matters: someone choosing to pay. With that in mind, here is the proof hierarchy.
At the bottom is stated interest: survey responses, “I would definitely use this,” a pile of emails saying the idea sounds great. This is the weakest evidence there is, and founders lean on it hardest because it is the easiest to gather and the most flattering. People are generous with enthusiasm that costs them nothing. Stated interest is close to worthless as proof of demand, and a deck built on it reads as a deck with nothing better. My advice, when asked, is: don’t bother asking.
Above that is revealed interest that comes at a cost. A waitlist signup with a real email and a real reason. Low cost and highly recommended. Time spent in a prototype. A letter of intent. Better, because the person did more than nod, but still short of the real thing, since none of it is money changing hands yet.
Higher still is usage with retention: people using the product repeatedly and coming back without being dragged back. Retention is the single most honest early signal there is, because it is almost impossible to manufacture and it answers the question that interest never can: not “would you want this” but “do you actually keep using it.” A small number of users with strong retention beats a large number with heavy churn every time, and it isn’t close.
At the top is revenue, and revenue that recurs and grows: customers paying, staying, and ideally paying more over time. This is the proof every other rung is a proxy for, and once it exists in any quantity the lower rungs stop mattering much. The founder who can show a retention curve and a revenue line doesn't need much else.
The single most useful thing a founder can do with this ladder is be honest about which rung they are actually on, and not dress a lower rung as a higher one. Calling a waitlist “traction” or survey enthusiasm “validated demand” doesn’t fool an experienced reader. It just trips the dishonesty gate and makes them distrust any real numbers.
Quality beats quantity, and the shape beats the number
A common mistake is to lead with the biggest number available regardless of what it measures. Fifty thousand signups sounds better than forty paying customers, so the deck leads with fifty thousand. To a careful reader, forty paying customers is the stronger story by a wide margin, because forty people parted with money and fifty thousand parted with an email address.
What I look at past the headline number is its shape. Is usage growing, flat, or quietly declining behind a cumulative total that can only go up. Cumulative signups always rise, which is exactly why founders show them; they hide the trend that matters. The honest charts show the rate, the retention, the cohorts, and they let me see whether the thing is accelerating or stalling. A smaller number with a healthy shape is far more fundable than a large one engineered to obscure a bad one.
What good looks like by stage
To make the stage point concrete, because it is the crux of the whole dimension. At pre-seed, evidence of demand might honestly be a handful of design partners using a rough prototype and giving real feedback, plus a clear experienced read on the customer’s pain. That is appropriate, and holding it to a revenue standard would be the error. At seed, I’d expect early usage with some retention signal, and ideally the first revenue, however small. By Series A, the conversation is about revenue, growth rate, retention, and the early shape of acquisition economics, and an absence of those at that ask is itself the answer. The standard rises with the round. The mistake cuts both ways: penalizing a pre-seed founder for no revenue, and excusing a Series A founder who still has none.
Current view, subject to change
I'll break the series pattern here, because this is the dimension I'd be slowest to revise. The proof hierarchy has held up across every plan I've read and every company I've watched succeed or fail. Interest, then revealed commitment, then retention, then revenue. That ranking is as close to a law as this messy field offers.
The part still genuinely flexible is the weighting curve by stage. I assert that traction’s weight should climb steeply from pre-seed to Series A, and the exact slope of that climb is a hypothesis, not a measurement. This is the central thing we are testing by running real plans through the agent and comparing its verdicts to what investors actually decided. If the data shows traction predicting outcomes even at the earliest stages, where I currently discount it, then I am under-weighting it early, and the curve is wrong. That would change my mind, and it would change it with numbers, which is the right way to be argued out of a position.
And that is the through-line of this whole series. A business plan is not a story you tell well. It is a set of claims you can or can’t back, scored on the evidence and not the idea. We have written ten of these now, one per piece of the method, in public and on purpose, because writing each part out is how we found where it was thin. If you’ve read this far and you think we’ve weighted something wrong, that note is the most useful thing you can send us, and it is how the series should end.