Vettey
Try free

How Vettey Works

YouTube's algorithm optimises for clicks and watch time. Vettey optimises for something different: did people actually achieve their goal?

Step 1: We search YouTube for every video on a topic

For each goal (like “fix lower back pain” or “learn to budget”), we search YouTube with dozens of queries and collect hundreds of candidate videos. We look at everything — not just what's popular, but niche creators with small audiences who might have the best content.

Step 2: We filter for videos with real success evidence

We analyse the comments section of every video looking for time-delayed success reports — people who tried the advice and came back weeks or months later to say what happened. A comment like “came back to say this actually cured my sciatica after 3 weeks” is worth more than 10,000 “great video!” comments posted the same day.

Step 3: AI scores every video across 5 signals

Each video that passes our evidence filter gets scored on:

Step 4: Only the best survive

Most videos don't make the cut. For a typical goal, we review 100–200 videos and only 5–10 pass our quality threshold. These are the videos you see on Vettey — the ones where real people report real results.

What makes this different from YouTube search?

YouTube ranks by engagement — what gets clicks and watch time. Vettey ranks by real results — what actually helps people achieve their goal. A video with 50 million views but generic advice will rank below a 50,000-view video from a physiotherapist whose patients report specific improvements.

How we rate the research we cite

Every study we cite has a Study Rating out of 100 that you can open to see how it was scored. Health research and investing research are rated on different scales. Read how we rate research.

Who's behind the ratings

AI does the reading, published rules do the scoring, and a person owns the method. Every video and every study is read and scored by AI (Anthropic's Claude) against the rules published on these pages. For research, two independent AI passes record the facts of each study; a script turns those facts into points, so the same facts always give the same score; and where the passes disagree, we check the paper and record the deciding fact. In our checks so far, the two passes landed within 10 points of each other for 86–92% of studies, and in the pilot our evidence grades matched every published GRADE assessment we compared against. We spot-check by hand, fix what's wrong and log every change to the method.

The method is designed and run by Tim Hammacher, the founder: a former trader, data analyst and consultant at Quantium, the data science company, where he consulted for blue-chip Australian companies using big data. He has qualifications in actuarial science and economics, including statistics, mathematics and physics.

No paid placement, ever: payment or affiliate status never affects a rating.

We're actively expanding — adding new goals every week and improving our scoring methodology based on user feedback. Have a goal you'd like us to cover? Let us know.