Fourteen thousand two hundred and six rows, in nine comma separated files, about what the feedback and roadmap tool category charges, publishes, writes and links to.
They were all built because something here needed them. A page cannot say what thirty two vendors charge without somebody reading thirty two pricing pages first, and once that reading is done the file is more useful in public than in a private directory.
Every one of them is free to copy, change, quote and use commercially, with no attribution and no link back.
What is in the nine?
| The file | Rows | What each row is |
|---|---|---|
| Pricing | 32 | One vendor, its entry price, the unit the bill is metered on, and what its free plan stops at |
| Capabilities | 32 | One vendor against nine questions, each cell naming the plan tier that gates the answer or recording that nobody publishes one |
| Page addresses | 9,432 | One address slug, how many vendors publish it, and which ones |
| Headings | 1,731 | One heading at least two vendors write, with how many pages and how many companies carry it |
| Page counts | 69 | One vendor's whole sitemap, counted twice two days apart, sorted into path clusters |
| Link graph | 2,706 | One outside domain that links to more than one vendor in this category |
| Domain authority | 70 | One vendor domain and where it sits on the web's link graph, across seven releases |
| Directories | 55 | One software directory, opened on a real listing, with the rel attribute on the link it gives |
| Template overlap | 79 | One pair of a vendor's own pages, and how much wording they share |
What does the licence let you do?
Everything, including the commercial things.
The dedication is CC0, which is as close to public domain as a licence gets. Copy a file. Put it in a spreadsheet and sell the spreadsheet. Publish a chart from it under your own name. Train something on it. Rewrite the conclusions to disagree with ours.
No attribution is required and no link back is required. If you would rather cite than copy, every file has a page of its own with a permanent address and the date the figures were read, and citing the page rather than the file means your reader can check whether it has moved since.
Why is a vendor publishing data about its own category?
Because the alternative was leaving it in a directory nobody can see, and because the figures do not flatter us.
The capability matrix has a column we fail on. The authority file ranks seventy domains in this category and puts this one sixty ninth of the seventy, with five referring domains against the category leader's three thousand and eighty nine. The directory file says that of fifty five software directories, seven pass ranking credit, which is a fact about how little a listing is worth, including ours.
A file that only made its author look good would not be worth downloading, and anybody who checked two rows of it would know.
What do all nine have in common?
Three things, and they are the reasons to trust any of them.
A date on every row. Not a date on the page, a date in the row. Pricing changes and a table without a read date is a claim about an unspecified past.
The address it was read at. Most rows carry the page they came from, so a row can be checked rather than believed.
Silence recorded as silence. Where a vendor publishes no answer, the cell says unpublished. It does not say no. An absence of evidence on a marketing site has never been proof that software cannot do a thing, and a file that quietly turns one into the other is worse than no file.
What none of these is
None of them is a scorecard, and none of them weights its columns against each other.
They also stop where the measurement stopped. A sitemap is what a vendor chose to list, so pages that exist and are not listed are not counted. The link graph is one crawler's view of the web and not the whole of it. A pricing page read on one day says nothing about the next.
Where a file has a limit worth knowing, that limit is written on the file's own page rather than left for you to find.
When do they go stale?
The pricing ones go first, and they can go within a week.
Prices and plan tiers move whenever a vendor edits a page, which is why the date sits on the row. The structural ones, being the addresses, the headings and the link graph, move much more slowly, because a company changes its page template far less often than it changes a number.
Each page says when it was read. Nothing here is on a promise to be current, and a file whose date has aged is still true about the day it names.