
Last week I posted about the founder whose competitors' homepages were getting cited by AI while she was invisible. The best pushback in the comments was that I was mixing two kinds of question: the broad one, best budgeting app, and the narrow one, budgeting app that works without linking a bank.
So I built the split into the index. Every question we ask an engine is labeled by type, and we can now show recommendation share separately for broad category questions and narrow capability questions. First category we looked at, one accounting product ,holds 95% share on the broad question and 40% on the narrow one. The blended number hides that a person in that company should be rewriting a page, not buying placements.
Here is the offer. Drop your product category in the comments and I will run it both ways and reply with who wins the broad question, who wins the narrow one, and where you sit in each. No signup, I will just post the result in the thread.
If the split looks useful and you want it tracked weekly, I will set you up free for a while in exchange for blunt feedback. Looking for what is confusing, what is wrong, what you would never pay for, whats useful. I have 26 users and more measurement than judgment about what matters to anyone who is not me. If you end up liking it and say so somewhere public that is appreciated, but it is not part of the deal, and I would rather have the criticism.
Thread is quiet so I will seed it with one run nobody asked for, the category from the original post. Consumer budgeting apps, both shapes, 28 runs across 4 engines. Small sample, so read it as structure rather than rates. The broad question, best budgeting app, is carried by financial media: Forbes 9 citations, CNBC 6, NerdWallet 5. The narrow one, budgeting without linking a bank, flips to tiny product homepages: pocketclear.app 7, moneypeas.app 6, getfinny.app 4, and reddit at three times the community share it gets on the broad question. Different questions, different gatekeepers, measured. Name yours and I will run it the same way, n stated every time.