marketaiguide.com
marketaiguide: What AI recommends, ranked every week
A pure AEO litmus test. Every list is what Claude says when a buyer asks it for the best tools in a category, scored against a rubric it wrote itself and published exactly as it comes, every week. No editor, no votes, no paid placement.
Free to read. Rerun every week.
- What it is
- A weekly AEO litmus test
- Ranked by
- Claude, on a rubric it wrote
- Refreshed
- Every week, with movement and trends
- Money
- No paid placement, plain vendor links
What is marketaiguide?
marketaiguide shows what an AI assistant tells a buyer who asks for the best products in a category. Answer engine optimisation, or AEO, is the work of being named and described well when someone asks an assistant instead of a search engine. marketaiguide measures the outcome directly: does the model know your product, and how highly does it rate it?
For each category, Claude writes a rubric of four to six weighted criteria, which is frozen and published on the list. It then names the products it would recommend, from its own knowledge and with no web access, and scores each one on every criterion with a reason. The rank is plain arithmetic, and the same question is asked again every week so you can watch brands rise and fall.
It measures perception, not which product is best, and a model can be wrong or out of date. The site is run by DevOpser LLC and says so: if the model names DevOpser, it is marked “our tool” and ranked like everything else. If the model does not name it, it is not there.
How marketaiguide works
The model writes a rubric
Claude decides what a buyer should judge products on, and how much each criterion weighs.
It names and scores products
From its own knowledge, with no web access, it scores every product it recommends on every criterion.
Arithmetic sets the rank
The rank is the weighted average. Entries are never reordered, added, removed or edited.
It is rerun weekly
Each list shows this week’s rank, the movement since last week and the trend.
The marketaiguide lists
Every list is free to read on marketaiguide.com and is reranked every week.
- The best open-source SaaS starter kitsReranked weekly
- The best AI app buildersReranked weekly
- The best managed hosting for indie SaaSReranked weekly
- The best AI business coachingReranked weekly
- The best AI coding assistantsReranked weekly
- The best AI agent frameworksReranked weekly
- The best LLM API providersReranked weekly
- The best vector databases for RAGReranked weekly
- The best auth providers for SaaSReranked weekly
- The best billing and subscriptions for SaaSReranked weekly
- The best AI customer support platformsReranked weekly
- The best AI meeting notetakersReranked weekly
What you get
Made for
- Marketers tracking AEO
- Founders checking how AI sees their brand
- Vendors slipping down a list
- Buyers curious what assistants recommend
- The published rubric
- Every list shows the weighted criteria the model wrote for it.
- Scores with reasons
- Each product is scored 1 to 10 on every criterion, with a reason for each score.
- Week-on-week movement
- Rank changes, a trend line, and who dropped off the list.
- The exact prompt
- Printed on every list, along with the model that ranked it.
- How to improve your AEO
- A guide to changing what the model learns about your product.
- Request a list
- Ask for a category the site does not cover yet.
marketaiguide questions, answered
How does marketaiguide rank tools?
Claude writes a weighted rubric for each category, then names the products it would recommend and scores each one from 1 to 10 on every criterion. The rank is the weighted average of those scores. Nobody reorders, adds, removes or edits entries.
Does the model search the web?
No. It answers from its own knowledge, with no web access, and nothing from the site, no vendor list and no submissions go into the prompt. The exact prompt is printed on every list.
How often do the lists change?
The same question is asked every week and every run is kept. Each list shows the movement since last week, a trend line and who dropped off.
Is it a product review?
No. It measures how an AI model already sees a brand, which is what a buyer hears when they ask an assistant. A model can be wrong or out of date, so treat scores as evidence of perception.
Can a vendor pay to rank higher?
No. Nobody sets positions, so nobody can sell them. The site explains what a vendor can do instead: change what the model learns about them.
Is marketaiguide independent?
No, and it says so. It is run by DevOpser LLC. If the model names DevOpser on a list, it is marked “our tool” and ranked by the same prompt and rubric as everything else.
See what AI tells your buyers.
Twelve categories, reranked every week. Missing from a list, or slipping down one? Find out why.