The month I rewrote onboarding at $10K MRR: a founder diary (2026)

The month I crossed $10K MRR should have felt like a landing. On paper it was a good month: revenue up a little, churn flat, a steady drip of new trials for the small B2B SaaS I run alone. But the growth line had gone soft for three months in a row, and I could not explain why. I was getting the same number of signups I had always gotten. I just was not turning enough of them into anything.

Quick answer (2026): At $10K MRR my growth stalled, not because signups dried up but because only about 22% of new signups ever reached the product's core outcome. I rewrote onboarding around a single activation milestone reachable in the first ten minutes: fewer signup fields, no settings wall, a real starter template instead of an empty screen, and one plain first-run email. Over eight weeks activation climbed to about 41% and MRR started moving again, from $10K toward roughly $11,800. This is a first-person composite founder diary; the figures are self-reported and rounded, but the pattern is the point.

A note before the numbers: the founder here is a composite, stitched from my own logs and from conversations with several other solo operators sitting between $8K and $15K MRR. The exact figures are self-reported and rounded. The mechanics are real even where the decimals are stylized.

The month the numbers stopped moving

I had spent a year believing my problem was the top of the funnel. So I did top-of-funnel things. I wrote posts, I answered questions in communities, I shipped a small free tool. Trials ticked up. Revenue did not follow.

The uncomfortable truth showed up when I finally lined the funnel end to end. Roughly 260 people started a trial that month. About 57 of them ever did the one thing the product exists to do. The rest poked at an empty dashboard, maybe clicked two buttons, and left. My Stripe dashboard was not the leak. The first ten minutes of the product were.

That is a different problem than "I need more traffic," and it needs a different fix. More traffic just pours more people into the same leaking bucket. I had been scaling the leak.

What I actually found when I looked at the funnel

I picked one honest definition of activation and held to it: a new account reaches its first real outcome, the thing they signed up to get, not a tour completed or a checkbox ticked. By that definition my activation rate was about 22%.

Then I read around to see whether 22% was normal or bad. It is bad, but not unusually bad. In Amplitude's 2025 analysis of time to value, as many as 91% of new users drop off within 14 days, and more than 98% of users churn within two weeks if they never experience value. The window is brutal and it is short. Whatever the product is going to prove, it has to prove it fast, or the account is already gone and you are just waiting for the card to fail.

What made it click for me was seeing where my drop-off actually sat. It was not spread evenly. It was a cliff on the very first screen. People signed up, met a blank canvas and a settings page, and quietly closed the tab. Nobody was rage-quitting. They were just never starting. This is the front end of the same problem I wrote about when my churn quietly doubled at $20K MRR: retention is mostly decided in the first session, long before the renewal date shows up.

The rewrite: one job, done in the first ten minutes

I gave myself four weeks and one rule. Every screen a new user sees has to move them toward that single activation milestone, or it gets cut. Here is what that actually meant.

I cut the signup form to two fields. Email and password. Everything else I used to ask for on the way in, company size, role, use case, I either inferred later or asked once the person had already gotten a win. Every field you demand before value is a toll booth in front of a product nobody has decided to like yet.

I killed the settings wall. The old flow dropped people into configuration first: connect this, name that, set your preferences. I moved all of it behind sensible defaults. You can change everything later. You should not have to change anything to see the point.

I replaced the empty state with a real starter template. This was the biggest single lever. Instead of a blank dashboard that says "create your first thing," the account now opens with a pre-built example already populated with realistic sample data, one click away from becoming the user's own. An empty screen asks the user to imagine the value. A populated one shows it.

I wrote one first-run email, in plain language. Not a five-part drip. One message, sent the moment the account is created, with a single link straight back to the activation step and one sentence about why it matters. I deleted the other four emails in the old sequence. They were all asking for attention before I had earned any.

I put a three-step progress marker on the main screen. Not a nagging product tour with tooltips. Just a quiet checklist showing where you are and what "done" looks like, so the first outcome felt reachable instead of vague.

None of these are clever. That is sort of the point. I did not add anything. I mostly deleted things that stood between a new account and the reason they showed up.

What moved, and what did not

I shipped the rewrite over the last week of the month and watched for eight weeks, because a new onboarding flow needs a few weeks of real cohorts before you can trust the read.

Activation went from about 22% to about 41%. Not top-quartile, but nearly double. For context, Userpilot's 2025 product-metrics benchmark puts the average B2B SaaS activation rate at 37.5% across 62 companies, with the median close behind, so the rewrite took me from well below average to a little above it. The same report cites a striking figure: a 25% increase in user activation is associated with a 34% increase in MRR over a year. I did not see anything that clean, but the direction was unmistakable. Over the eight weeks my MRR moved off its plateau, from $10K to roughly $11,800, on the same trial volume as before. I did not find new users. I stopped losing the ones I already had.

Now the honest part, because a diary that only reports wins is a brochure.

Two of my changes I reverted. The three-step progress marker helped, but a first version of it auto-expanded into a full guided tour, and people skipped it exactly the way I skip tours. I cut it back to a passive checklist. And I got greedy and tried to also gate a second, "power user" feature behind its own mini-onboarding; that just reintroduced a wall, and activation on that path dipped until I removed it.

I also want to be clear about what activation did not fix. It did not save accounts that were a genuinely bad fit, the ones who signed up hoping the product was something it is not. Those still churned, and they should. A better first ten minutes gets the right people to value faster. It does not turn the wrong people into customers. That is the same lesson I learned the hard way when I finally killed my free plan at $16K MRR: removing friction is only a win when the people on the other side of it actually wanted what you built.

If you take one thing from this

Pick one activation milestone, the single moment a new account first gets the thing they came for, and write it down as a sentence. Then instrument it so you can see your real activation rate, not a proxy like "logged in twice." Then go through every screen between signup and that milestone and delete the ones that do not lead to it. Most of the growth I was chasing at the top of the funnel was sitting in the first ten minutes the whole time, waiting for me to stop getting in the user's way.