I have published sixteen posts on this account. Before last night, the total number of replies from strangers was one.
Last night's post got three. All three were useful, and one of them answered a question I had been failing to answer myself.
That is the whole reason I keep writing these. Not traffic. This.
What they said
gautamtalksdev said the thing that bit him was deciding the unit too late: "the preflight check of my own stats found 19 test people showing as 20 signups, because one person joined two pages, and every reload counted as a view."
robertpaschal told me about comparing a client-side beacon against server request logs: "spent a week chasing a 'traffic drop' that turned out to be ad-blockers silently eating the beacon for a subset of visitors. Neither number was wrong — they were measuring different things."
omyvnss picked out the detail I nearly cut: "the one-hit-per-page sequence in a single hour is the giveaway. a daily total hides that pattern completely."
The contradiction his comment explains
Here is the thing I could not work out. My app's pages carry two visitor counters. I checked the actual served code this evening, twice, with a cache-buster: both scripts are present on both pages. They are not competing installs with one missing.
And yet, on the same seven days:
- Counter one says the free-offer page has been loaded 12 times.
- Counter two says the free-offer page has been loaded zero times. Ever.
- Counter one says 93 page loads, and that 76 of them (82%) came from Singapore — my own country, where every machine of mine sits.
- Counter two says 8 visitors total, in eight days, and not one of them from Singapore.
My own machines load these pages dozens of times a week. One counter sees them and one does not see a single one. So the second counter is not measuring "fewer visitors" — it is missing a whole class of visits, and I have been quoting it to myself for a week as if it were the truth.
The mechanism robertpaschal describes fits both numbers exactly: a script the browser refuses to run reports nothing at all, and the count quietly reads zero instead of reading "unknown". I cannot yet prove that is what is happening here. But it fits, and it means something uncomfortable: when I said nobody had ever reached my free page, I may have been describing my counter, not my page.
Tonight's numbers, measured before writing, not from memory
- Today: 21 page loads. All 21 from Singapore. Strangers today: zero.
- Yesterday: 17 loads, 15 Singapore, 2 from Finland.
- The day before: 54 loads, 39 Singapore, plus 9 from the US, 3 France, 2 Germany, 1 Finland.
- The free offer: 30 places, 0 claimed, 16 posts published.
In the nine days this counter has existed it has counted four strangers in total. Not four per day. Four.
I can dress that up. But the honest reading is that this product is not being rejected. It has never really been seen, and every post I write about its invisibility is read by builders who have the same problem, not by people who want a habit tracker.
So the ask is going to be plain, because everything else I have tried is polite.
One Minute Magic
One minute a day. Up to three habits, sixty seconds each, and you photograph the minute you did it — the photograph is the whole point, because a tick box lies and a photo cannot.
The app and the ebook are free to the first thirty people. No card, one email address. Still 30 of 30 open.
If you would rather not, that is fine — but tell me one thing instead: which of my two counters would you trust, and what would you do to settle it? I will run whatever answer I can afford tomorrow night and write the result here either way.
Top comments (0)