Show IH: the most useful thing I did this week was write zero new code
Last week I wrote about learning to sit in the silence after shipping instead of rebuilding. This week I got to test whether I actually meant it.
The urge showed up right on schedule: "just add one more feature, then promote." I've killed 4 products that way. So this time I made myself do the opposite — build nothing, and instead check whether the thing I already shipped actually works end to end.
Two things I found:
-
My bottleneck was never traffic. Strangers do scan my tool — I even found scans in my database from URLs I've never seen. The leak wasn't "nobody comes." It was "people come, scan, and leave with no way to stay in touch."
-
A quiet bug was making it worse: when the paid checkout was live, the "leave your email" box on the report didn't render at all. So anyone not ready to pay just... vanished. I fixed it so the report now offers three doors: pay, leave your email, or come back later.
Today I ran the whole funnel myself — scan a page, get the report, confirm both the checkout and the email box show up. They do now.
The unglamorous lesson: "staying" didn't look like shipping. It looked like checking that the door I already built actually opens.
Still building AbleScan — one-click accessibility scans for websites: https://ablescan.app
For those further along: what's the leak you almost missed because you were busy building the next thing?
the missing email box is a good example of why the funnel should include an “interested but not ready” path. i would treat the three doors as different commitments: payment, permission to follow up, and a return visit. then compare which one predicts a later paid report instead of treating every captured email as equal. walking through the funnel as a new user is probably worth repeating after every checkout or report change.
You landed on the exact thing I've been circling — "repeatable without turning into a pitch" is the real test, not whether I can do it once.
And the record-keeping suggestion is sharp, because it forces the honest version of the question: if I can't write down what someone's actual problem was before I showed up, then I wasn't having a conversation — I was just waiting for a slot to pitch into.
So I'm starting to log exactly that: the original problem, what they'd already tried, what they did with the scanner, and what they asked for next. Five half-laps of quitting means I have zero data on the second half. This is the first time I'll actually have a paper trail of whether the motion repeats or was just one lucky loop.
Thanks for pushing on the repeatability part specifically — that's the bit I'd have let stay vague.
Finding that the email box vanished while checkout was live is the kind of leak that only shows up when you walk the funnel yourself. Since you're pricing in euros per report, are you handling EU VAT and invoicing inside your checkout today, or is that still manual? I've been comparing checkout and merchant-of-record setups for one-person products and can share what I've found.
Good eye — that one only showed itself on foot. On the euro side: I'm on Polar, which acts as merchant of record, so they handle EU VAT and invoicing rather than me doing it by hand. That was honestly a big reason I went with them over rolling my own Stripe flow — I didn't want to become a part-time tax-compliance desk as a solo. Curious what you've been finding, though — always up to compare notes on the MoR options.
This is such a good example of why “we need more traffic” is often a dangerous diagnosis.
The interesting leak wasn't getting people to the product. It was what happened after they arrived but weren't ready to buy.
That middle group is easy to overlook because they're not a conversion failure in the obvious sense. They didn't buy, but they also weren't necessarily saying no. They just disappeared.
Giving someone a way to leave their email, come back later, or buy creates different levels of commitment instead of forcing a binary “pay or leave” decision.
I think founders would find a lot of these leaks if they stopped looking at the funnel as a straight line and started asking:
“What can a user do at each point when they're interested but not ready for the next step?”
That's often where the quiet money is.
Good eye — that one only showed itself on foot. On the euro side: I'm on Polar, which acts as merchant of record, so they handle EU VAT and invoicing rather than me doing it by hand. That was honestly a big reason I went with them over rolling my own Stripe flow — I didn't want to become a part-time tax-compliance desk as a solo. Curious what you've been finding, though — always up to compare notes on the MoR options.