We read the same five pricing pages again the following week. Where a page gained, lost or renamed a control, this page names the control — rather than counting the edits, because on one of these pages the count is the part that misleads.
Five pricing pages, fetched and counted by code with no model in it — no buyer, no decider, nobody reading the page — read, and then read the same way again the following week. The pages we already publish a comparison about are here as the control, and they are the reason the rest of this page is worth reading: Linear’s pricing page did not move, Notion AI’s did not either, nor did Airtable’s. Not a control gained, lost or renamed on any of them, both times, by the same code. Elsewhere in the group, somebody shipped. Retool stopped serving a “Hire a developer” control that its pricing page carried at the earlier reading. On monday.com’s pricing page a “Notetaker” control appeared and an “AI workflows” control appeared beside it, and the words on links were rewritten while their addresses stayed exactly the same: monday.com renamed “monday agents” to “Agent builder”, “monday sidekick” became “AI assistant”, and “monday vibe” became “Vibe coding”.
Nobody was in the loop. No ghost user walked these pages, no model decided anything, and not one sentence or number here was written by one — which is why this is a reading and not a run, and why there is nothing on it about where a person got stuck.
- The instrument
- Each page was fetched several times on each of the two mornings and watched until its own list of controls stopped changing, then counted — the same way, by the same code, on both. Nothing was signed up for and nothing on this page is a sentence anybody or anything wrote about these products. A control is named as gained, lost or renamed only where every fetch on both mornings agrees about it: a control a page hands to some fetches and not others cannot be told apart from an edit by one reading of it, so no sentence here names one.
- When
- The pair of readings is the unit rather than the date: a pricing page is edited without notice, so a reading is a fact about the moment it was taken and nothing else. What is on this page is a statement about the distance between these readings and nothing wider — it is not an edit frequency, and nothing here says how often these companies touch these pages, or that the edits between them were made in one sitting. The table below is the later of them, and the record each of these sentences is checked against is the pair rather than that table.
- Pages read
- monday.com/pricingairtable.com/pricingretool.com/pricinglinear.app/pricingwww.notion.com/pricing
Counted on the pricing pages themselves.
We opened each pricing page, waited until its own list of controls stopped changing, and counted the controls that send a reader somewhere else in a new tab — recording, for each one, whether it goes to material of that company's own on a separate site or to somebody else's. These are a fact about these pages on 17 October 2026 and nothing else, and anyone can open the same pages and count them again.
| Pricing page | Controls on the page | Open a new tab | …to its own material on a separate site | …to somebody else's site |
|---|---|---|---|---|
| monday.commonday.com/pricing | 169 | 32 | 8community.monday.com, ir.monday.com, support.monday.com | 13 |
| Airtableairtable.com/pricing | 103 | 18 | 5community.airtable.com, support.airtable.com | 8 |
| Retoolretool.com/pricing | 231 | 10 | 4docs.retool.com | 2 |
| Linearlinear.app/pricing | 64 | 6 | 0 | 4 |
| Notion AIwww.notion.com/pricing | 73 | 0 | 0 | 0 |
What we read and are not publishing.
- A count of the edits
- There is no tally of edits on this page and that is deliberate, because the obvious tally lies. A renamed control departs and arrives in the same breath, so it nets to nothing in a subtraction of one control count from another, while a reader looking at the page plainly sees an edit. Every edit is therefore named above instead, and no number anywhere on this page is a difference between readings. The count in the table below is the later reading’s own and nothing is subtracted from it.
- The controls on Retool’s pricing page our own fetches disagree about
- Its page served a “Contact sales”, a “Book a demo” and a “Request trial” control to some of our fetches and withheld them from others, at both readings. A page that does not hand the same controls to every fetch of it can give us a difference with nothing behind it but our own sampling — so nothing here says anything about those controls, in either direction. That is not a claim that they are as they were; it is a claim we are not entitled to make either way, and the page this project was publicly corrected over turned on exactly one of them.
- The controls in monday.com’s navigation our records cannot pair
- An “AI platform”, a “Resources” and a “Solutions” control each went from a link to a button at an address that stayed the same. Our records hold both shapes, one at each reading, and cannot establish that the button is the same control as the link rather than a different one — so each of those labels sits in a departure and an arrival at once, and a sentence saying monday.com took one of them down, or that it put one up, would be false as a reader reads it. The absence is written here rather than left silent, because an omission reads as nothing having happened.
- A price reading
- This page publishes no reading of what anything costs. It is about which controls a page gained, lost or renamed, and the price readings of these same pages are published, with their records, on the pages that compare them.
- Anything a person did
- Nobody walked these pages. There is no buyer here, no model decided anything, and not one sentence on this page was written by one — so there is nothing here about where somebody got stuck, and these numbers are not evidence about what a reader would do.
What this page is for is the pair of halves, not either on its own. Some of these companies edited their pricing page in the gap between our readings of it and the others left theirs alone, and the second group is what makes the first a fact about them rather than about our instrument — same code, same group, both readings. The edits themselves are small and they are not a verdict on anybody: a link taken down, features added to a list, names rewritten over links at addresses that stayed put. What a reader can take from this is narrower and, we think, the useful part — that the thing we can do here is say which control on a named company’s page arrived, departed or was renamed, and show the readings it was established from, without anyone at that company telling us anything.
← Everything we have published, including the runs with a buyer in them