John Walker

Hey everyone 👋 I built AI Got Ranked because I kept running into the same problem: every "best AI tools" list I found online was either a thinly-disguised affiliate page or a "top 10" that mysteriously matched whoever paid for placement. I wanted something that actually scored tools honestly. So every tool on the site gets rated across six weighted metrics — usefulness, quality, ease of use, value, reliability, and popularity — and there are zero paid placements. If a tool scores low, it stays low. No exceptions, no sponsorships buying a better spot. It's still early — currently indexing 2,000+ tools across categories like writing, coding, image gen, and more — and I'm building this solo, so there's a long roadmap ahead (comparison tools, personal "stacks," score-change tracking). Would genuinely love feedback from this community — what would make this actually useful to you? And if you've built an AI tool yourself, feel free to submit it; happy to take a look. Thanks for checking it out 🙏

"Put your business decisions on trial" is a great framing — most decision-support tools just give you a rosy forecast, but the actual value here is having something argue against your plan before reality does. Competing AI agents exposing weaknesses in your assumptions is a much more honest way to stress-test a strategy than the usual "here's your SWOT analysis" output. The leaderboard/run format in the screenshots is a nice touch too — turns what could feel like criticism into something more like a game. Would love to know how it decides when an agent's pushback is a genuine flaw vs. just devil's-advocate noise — that's the line that'll make or break trust in the output over time.

The "proof of work" feature (live camera shots before a task can be closed) is the standout here — for mobile/field crews this solves a real trust problem that most task managers just ignore, since a checked box tells you nothing about whether the job was actually done right. The self-assignment marketplace angle is smart too, since it shifts dispatch overhead onto the team itself rather than a manager manually assigning everything. Flat pricing up to 500 accounts is a genuinely rare and generous structure at this end of the market — most tools nickel-and-dime per seat well before that number. Would be curious how disputes get handled when a "proof" photo is submitted but still gets rejected by the requester.

The "spend most of your time chatting with the agent rather than managing it manually" framing is the right instinct — most CRMs fail because updating them feels like a chore layered on top of the actual sales work. Routing that through WhatsApp/Slack instead of a separate app login is smart, since it meets reps where they already are instead of asking for a new habit. Curious how it handles edge cases where the AI misreads a lead's intent from a messy conversation thread — that's usually where "AI-managed CRM" tools start to show cracks. Following this one.