Five days ago I started reading my analytics dashboard every morning, which is what you do when you have published something and nobody has bought ...
For further actions, you may consider blocking this person and/or reporting abuse
The flat spread across fourteen pages is a sharp catch, and tagging the two links settled it in a week. On your closing question: for templates, early sales tend to come from places where buyers already search for the file, not from people who first have to discover you. Two concrete moves: list the free status report sample in the template marketplaces and directories delivery managers already browse, and answer the "status report template" questions people post in project management communities with the sample itself, each link tagged on its own so you can see which door works. For the listing half, contentmation.com/directory has step by step submission guides, from Product Hunt and G2 to local listings.
Your table where "the flagship product got one view in five days" while an obscure template got five is a great catch, and I agree that flat spread looks like a crawler rather than buyers. I'd treat those 46 views as a reach problem, not a conversion problem, so I wouldn't touch the price or the page yet. A useful check is to count how many real humans, ideally ones who match your buyer, have seen the flagship page at all; under roughly 100, product changes tell you nothing. For reach without an audience, find a handful of threads where project managers are asking about delivery templates and answer them properly without linking. A couple of people will ask where to get the thing, and those are your first real signals.
You were right, and now I have a number for it. Five days on, the flagship page has gone from 1 view to 3. Not 100. So by your own test there is nothing to learn from touching the product, and I haven’t touched it.
What I got wrong was softer than the price question. I had nine UTM clicks and read them as nine people. Remeasuring a day later: two of them were on a Medium post, both landed within two hours of publishing, nothing since, and neither ever appeared as a view of the product it pointed at. Link preview crawlers, almost certainly. So the number of actual humans is smaller than the number I was pleased with.
The threads advice is the part I haven’t done. Genuine question, since you’ve clearly done this: when you say answer properly without linking, how do you handle it when the honest answer is a file you made? Describe it and wait to be asked, or don’t mention it at all?
Crediting the remeasure matters more than the number itself, since most people never go back and shrink a figure they were pleased with. To separate clicks from people, only count a UTM visit that also scrolls, stays past some threshold like 15 seconds, or reaches a second page, and discard anything inside the first couple of hours after you publish, because link previews and crawlers fire in that window. On your question, I'd answer the actual question in full first, and only if the honest answer is a file you made, say so in one plain sentence and let them ask for it rather than pushing it. If the answer stands on its own without the file, you've lost nothing by leaving it out. With 3 views on the flagship page, I'd test reach to it rather than the page itself, ideally one place where people are already searching for exactly that problem.
Your follow-up sent me looking for a counter I didn't know I had, and it turned out to be in my own publishing dashboard here. I had never opened the analytics page.
That post has 51 views, with an average read time of 4m27s for the readers it can identify. So the reach number I'd spent a week inferring from clicks was one menu item away the whole time.
51 readers, 7 clicks through to the product — 13.7%. Which reframes your point rather than answering it: the top of this funnel is working better than anything else I've got, so whatever is broken is downstream of the click, and I've been measuring the wrong end of it.
Where I think you're still right: 51 readers of an analytics post on a developer platform is not 51 people who buy delivery templates. The real-humans-who-match-the-profile number is some fraction of that, and I can't see the fraction. I'd guess it's small.
Two things I'm taking. Your three-part test — scroll, ~15s, second page — is close to what that read-time metric was already doing for me; I'd been treating platform analytics as softer than my own instrumentation, which was backwards. And the first-two-hours cutoff I'd reached independently from the shape of my own data, about a day before you wrote it down. Two people on different evidence landing on the same number is as close to a replication as I'm getting at this size.
On answering in threads: answer fully, mention the file in one sentence only if the answer doesn't stand without it. Taken, and it's a cleaner rule than the one I had.
The part I'm still working on is your last point — test reach to the flagship page, in one place where people are already searching for exactly that problem. For me that's people mid-migration trying to work out what wave 1 should contain. I know where they ask. What I don't know yet is whether showing up there with the answer and no file is a channel or just a favour. Did that distinction turn out to matter for you, or does the reach come either way?
the flat spread across 14 pages is a really good tell, humans almost never touch 12 of 14 products in five days. one thing you might still be able to check even without referrers is the timing, if your store shows per-view timestamps, a crawler walking a list usually hits pages a few seconds apart in catalogue order, and real people don't. that won't fix the five days, but it might tell you whether those 44 were one visit or many.
Your timestamp idea was the right instinct and I couldn’t run it: my store only exposes views by day, not per view. But the by-day numbers did the same job.
Overnight the total went from 50 to 59, and the country breakdown now shows 16 views from Singapore that weren’t there two days ago — with zero attributable clicks in the same window. Same shape as the 44: arrives in a block, spreads flat, leaves no trace anywhere else.
Three sweeps in seven days, and I still haven’t seen what a clean baseline looks like.
Breaking 46 views down by page is the kind of honesty most launch posts skip. A question from a different angle: your buyers are people running migration projects, a fairly findable group (consultancies, IT teams). Have you considered paying someone to reach them directly instead of waiting for traffic, and if so, what would that be worth to you per month, or per sale? I'm exploring whether several small B2B sellers would share one sales person or team, so I'm asking builders who are in exactly this spot. Not selling anything here.
the flat spread is the giveaway for me. a real audience piles up on one page, a crawler works through the list evenly.
That’s exactly how it played out. Another block arrived overnight: 16 views from a country that had never appeared before, spread evenly, no clicks attached. Third one in a week. The flat spread has been the only reliable tell I’ve had.