400K impressions. 4 organic users. 0 paying customers.
Someone built a real budgeting PWA, not a toy.
Demo mode, Stripe, free trial, CSV import/export, savings goals, paycheck planner, habit streaks, the whole "this is a real product" stack.
Then they posted around. Replied under big Twitter threads. Got hundreds of thousands of impressions.
And the numbers were still:
4 organic users.
0 paying customers.
That is the kind of result that messes with your head, because it looks like distribution happened.
But I don't think all attention counts the same.
A reply under a viral thread can put a product in front of people who are mildly interested in the topic. That is not the same as being found by someone who just failed at budgeting for the third time this month and is actively looking for a less painful system.
From far away, those look like the same audience.
They probably are not.
The founder compared the product to YNAB on price. About 53% cheaper. I get why. It is concrete.
But people who pay for YNAB probably are not mainly thinking, "I wish this were cheaper."
They may be thinking, "I trust this enough to run my money through it."
The sharper clue was buried somewhere else: this was for people who tried YNAB or ledgers and quit because the maintenance felt too heavy.
That is a much more specific person than "people who want a budgeting app."
So I would not read 400K impressions -> 4 users as just "the landing page is bad" or "the product needs more features."
Maybe the audience was too cold.
Maybe the promise was too broad.
Maybe the traffic came from people willing to glance, not people already trying to fix the exact behavior.
The painful part is that those all look the same from the founder side:
silence.
For people who have had a big attention number and almost no users, when it finally started working, what changed first: the copy, the person you were aiming at, or the place where they found you?
400k impressions → 4 users usually means the channel is broadcasting to the wrong audience, not that the product is broken.
what's your motion right now — any 1:1 in communities where your ICP posts their problem daily, or mostly content/SEO waiting for inbound?
Right now, more 1:1 than SEO.
Content is useful, but mostly as a slow asset layer. The active work is finding places where the problem is already visible.
The tricky part is that people rarely name the problem cleanly. They describe the symptom:
lots of attention, no users
free users who never pay
client work that will not standardize
a workflow that quietly becomes support
Those can look like different problems, but often the question underneath is the same: did this reach someone with a real buying moment, or just someone who thought the idea was interesting?
So I am less interested in "more distribution" by itself, and more interested in whether the surface contains people already dealing with the failure pattern.
this resonates — "symptoms not category" is exactly why searching by product name on Reddit fails.
the failure-pattern surfaces idea is smart. curious what you're building and where you're doing the 1:1 right now — Reddit, IH, LinkedIn, something else?
i've been testing a small workflow that scores threads by whether the post sounds like a real buying moment vs "interesting idea" venting. happy to run a free sample if you share a one-liner + 3–5 communities where your ICP might post those symptoms — only if you're actually replying there this week.
That buying-moment vs interesting-discussion split is exactly the hard part.
The edge case I'd want to understand: how does your workflow treat a thread that sounds painful, gets smart replies, but shows no sign the poster is actually in a position to act on it?
great edge case
how i handle it: score isn't just "painful?" but "painful + plausibly in motion."
signals that bump score down even when the post hurts:
signals that keep score up:
so a painful thread can land 4/10 with reason "sympathetic vent, unlikely to act this week" vs 8/10 "founder actively hunting, wrong channel not wrong person."
that's the split you're describing — buying moment vs interesting discussion.
if you want to see it on your ICP: one-liner on what you're building + 3–5 subs where those failure-pattern symptoms show up. i'll run a small sample and you can tell me if the low scores feel right.
This is useful. "Plausibly in motion" is a clean way to frame it. The workaround signal maps to the same thing I'd weight highest.
Once someone is already stitching together spreadsheets, tools, or manual steps, it stops being just pain and starts showing movement.
I'll skip the sample for now.
fair call — sample only makes sense if you're replying this week. the workaround signal is usually the fastest filter in practice.
if you start doing community 1:1 once you're past the "symptoms not category" stage, happy to compare notes on what scored high vs noise.
That makes sense. I'd probably keep it at the pattern level too.
One thing I'm still not sure about: does the age of the workaround matter? A two-year workaround might be habit, but one from last week might be a signal. I haven't tested that yet.
Going to run a small batch before I pretend to have a clean read on it.
age of the workaround is a real filter — i'd weight "started last week / this month" higher than "we've always done it this way."
habit workarounds often mean they've adapted and won't pay. fresh ones usually mean the pain just crossed a threshold.
curious what your small batch shows — if last-week workarounds convert better, that's a clean scoring rule.
I'd push back slightly on the last part.
Age is a useful filter, but I would not make it the score by itself. A fresh workaround can mean the pain just crossed a threshold. It can also mean someone is only experimenting and has not committed to the pain yet.
The thing I would want to know is what changed right before the workaround appeared.
If the small batch shows anything useful, I would probably read age together with that trigger, not separately.