On November 6 I published a tutorial about a bank statement PDF-to-CSV converter I built. On November 7 I published 12 more posts in the same series, plus a D&D one-shot preview and a debt payoff post. Sixteen posts in about two days.
This post is the receipt. Every number below comes from the dev.to /stats pages for my own account, read on November 7 at about 19:41 CST, time range Infinity unless noted. I am not rounding anything up.
The sixteen posts
Readers / reactions / comments / bookmarks:
- Presale post, Nov 5: 0 / 0 / 0 / 0
- Part 1, the tutorial, Nov 6: 20 / 0 / 0 / 0
- Part 2, GUI and installer: 0 / 0 / 0 / 0
- Part 3, the OCR path: 0 / 0 / 0 / 0
- Part 4, why pdfplumber: 0 / 0 / 0 / 0
- Part 5, what the CSV looks like: 0 / 0 / 0 / 0
- Part 6, what it has not proven yet: 20 / 0 / 0 / 0
- Part 7, installing tesseract on Windows: 20 / 0 / 0 / 0
- Part 8, three edge cases: 0 / 0 / 1 / 0
- Part 9, a real bank statement found a bug: 2 / 1 / 0 / 0
- Part 10, the version check: 20 / 0 / 0 / 0
- Part 11, the first release: 0 / 0 / 0 / 0
- Part 12, using the GUI: 0 / 0 / 0 / 0
- Part 13, Show HN got flagged: 0 / 0 / 0 / 0
- D&D one-shot preview: 0 / 0 / 0 / 0
- Debt payoff post: 0 / 0 / 0 / 0
Totals for the sixteen: 82 readers, 1 reaction, 1 comment, 0 bookmarks.
Two footnotes on the method. Parts 5, 11, 12, 13, the D&D preview and the debt post were read on the default Week range, not Infinity — but all six were published on November 7, so Week covers their whole lifetime, and on the other ten posts Week and Infinity gave identical numbers. The dashboard shows all sixteen as under 25 views, which matches.
The oddities I will not smooth over
Four posts show exactly 20 readers each — parts 1, 6, 7 and 10 — and all four list their traffic source as Direct / Unknown, 20 out of 20. I do not know what that is. It could be crawlers, or a counting quirk. I am not going to claim 80 real readers.
Part 8 shows 0 readers and 1 comment. A comment exists, so somebody was there; the reader count says 0. The stats are noisy in both directions.
The one reaction in the whole set is a like on part 9, from Nils Harker, on November 7. Average read time on that post was 38 seconds. That is the only visible sign that a specific person read anything in this series.
For context, the account totals across all 97 articles are 1,398 readers, 16 reactions from 10 distinct users, 63 comments, 0 bookmarks, 20 new followers. Traffic sources: dev.to 57, google.com 51, tsecurity.de 10, Direct / Unknown 1,280. Almost all of that predates this series.
Show HN, same story
The Show HN post went up November 7 at 16:47 CST and was flagged within the hour. At 19:45 it still had 1 point and 0 comments, and had slid to page 3 of /newest, position 73. It was position 24 on page 1 at 17:42 and position 48 on page 2 at 18:42. I have not voted on it, commented on it, or appealed the flag.
What is actually selling
Nothing. The converter presale is still at 0 orders and 0 inquiries. The D&D trial PDF has 1 download of the Chinese version and 0 of the English version. The debt payoff spreadsheet has 0 request emails. Those are the numbers the posts were supposed to move, and they have not moved.
What I take from it
Volume is not distribution. Thirteen posts in a day produced 82 counted readers, one reaction, one comment, and no orders. The two things that got any readers at all were the tutorial itself and the posts about what the tool cannot do yet — parts 6 and 7, and part 10. The how-to posts in the middle got nothing.
I am keeping the two seven-day tests running as planned: the D&D trial needs 20 downloads, the debt spreadsheet needs 10 request emails. If neither moves, the conclusion is not "write more posts." It is that this channel, for these products, is not the lever.
The converter itself is here: https://github.com/alapha888/bank-statement-presale
The presale page is here: https://afa-kzz.onrender.com/bank-statement-presale/
Top comments (0)