Four numbers describe your position in a category: how often the engines name you, how much of the category's total naming is yours, where in the answer you land when you are named, and how the sentence reads when you are. Citino prints all four with the sample they came from and the interval around them, scoped to the segments you actually sell into.
Warehouse Management Software · US. 54 category prompts, 1,785 answers in the last 30 days, times in ET.
Ninth of 17 by visibility — and third of six by position when named. Those two facts point at different problems.
Every competitor renders in the same neutral grey at the same weight, whether they are first or sixteenth. Only your row carries the accent. A table that colour-codes six brands is asserting that six things are urgent, and a reader told everything is urgent stops reading colour as information. You are the only entity on the screen whose position you can do anything about, so you are the only one the eye is pointed at.
The same rule holds in exports and in white-label agency reports. The accent follows whoever the report is for.
Position and sentiment are conditional on being named. A brand named twice and praised twice scores 100 on sentiment and is still invisible.
IllustrativeTallyfold is up 1.7 points over thirty days with a band of ±3.0. The honest description of that is: we cannot tell yet. Citino prints it grey, labels it within noise, and sends nothing. The rule is symmetric — −0.8 is not a decline either, and you will not be paged for one.
This costs us the most satisfying feature in the category. Green arrows and weekly "you're up!" emails are generated by treating sampling error as performance. Act on them and you will rewrite a page that was never losing, measure a rebound that was always coming, and conclude the rewrite worked. Two of the six brands in the table above did move this month. Four did not, and are shown not to have.
Crateline +2.9 and Palletworks −3.4 clear their bands. Those two alert. The rest do not.
Tallyfold appears in one answer in ten, and when it appears it averages 2.6 in the naming order — ahead of Palletworks at 3.1, Shipmatic at 3.6 and Bindley Ops at 4.2, all of which are named more often. Buyers read answers from the top, so fourth in a list of four is a different commercial outcome from first, and a tool that counts mentions records both as one mention.
That combination has one reading: the engines rate you when they know about you, and mostly they do not. It is a coverage problem, not a reputation problem, and the two take opposite work. Chasing sentiment while visibility is 10.3% is polishing a sentence most buyers never see.
A category-wide visibility number averages over questions you do not compete for. Tallyfold sells to third-party logistics operators; "best WMS for a single-site manufacturer" belongs in the category but not in the comparison, and leaving it in drags the headline toward a market you do not want.
Every headline on this page is scoped by that selector. Narrowing the segment narrows the sample, so the bands widen — and they widen visibly, rather than the number simply looking sharper because it is based on less.
The 3PL segment carries the category's top prompt by value: "best warehouse management software for a 3PL". → Attribution
It will not tell you what any particular buyer saw. Answers vary with phrasing, history and the hour, and we sample rather than census — which is exactly why every number here carries an interval instead of a decimal place it has not earned.
It does not include Copilot or Google AI Overviews. Neither offers an official answer API, and a scraped number would be worse than no number, because it would look the same as the real ones.
It will not tell you that a fix worked. That takes a baseline and a second read, which is a different screen.
Pick a category and the table, the bands and the sample sizes are the first thing on the screen.