How we rate
Our scoring method, in full, including what it cannot tell you.
Most recommendation sites ask one question: is this product popular? That question is easy to answer and nearly useless, because popularity tells you nothing about whether a product suits you.
We ask four questions at once. Is there real demand for this? Do owners actually like it? Is it priced fairly against what it does? And can we explain, in plain words, who it is for? A product has to do well on all four before it earns a page here.
The two numbers on this site
They measure different things and we keep them visually separate so they never get confused.
Our rating, out of 10. This is our editorial judgment of the product itself. It answers "is this good, and who is it good for?" It is the number at the top of every review.
Product Opportunity Score, out of 100. This is an internal number that decides whether a product gets covered at all and how prominently. It answers "should we be writing about this?" We publish it because a recommendation site that hides its selection criteria is asking you to trust a black box.
The eight factors
The Opportunity Score is the sum of eight weighted factors. The weights total 100.
| Factor | Weight | What it measures |
|---|---|---|
| Demand signals | 25 | How many people are actively trying to buy in this space |
| Customer rating | 15 | What owners say once the novelty has worn off |
| Review volume | 10 | Whether there is enough owner feedback to draw a conclusion |
| Price attractiveness | 10 | Value for money against direct alternatives |
| Commission potential | 10 | What we would earn if you buy |
| Search demand | 15 | Whether people are looking for guidance on this |
| Social potential | 10 | Whether the product is easy to explain in a short video |
| Problem-solving value | 5 | Whether it fixes a real, specific frustration |
Commission potential is on that list, and it is worth being direct about why. We are paid by commission, so pretending money plays no part would be dishonest. It carries ten points out of a hundred. Demand, customer satisfaction and search interest carry fifty-five between them. A product cannot buy its way onto this site with a high commission rate, but we are not going to claim the rate is invisible to us either.
The tiers
- Tier A, 80 and above. Hero products. These get a full review and a place on the homepage.
- Tier B, 65 to 79. Supporting products. These appear inside guides and comparisons.
- Tier C, 50 to 64. Experimental. We test the idea on social first before committing a page.
- Below 50. We do not cover it.
Tested or researched
Every review states which of these it is, at the top, before the verdict.
Hands-on means we used the product ourselves. Research-based means we did not, and the verdict is built from manufacturer specifications, published measurements from people who do run labs, and the pattern of what owners report over time.
Research-based reviews are still real work and we stand behind them. They are not the same as having used the thing, and we are not going to blur that line. If a review does not say hands-on, assume we have not held it.
What this method cannot tell you
Scoring is not a substitute for knowing your own situation. The method has real limits and you should know them.
- It cannot tell you whether a product suits your ears, your hands, or your desk.
- It reflects a moment in time. A great product with a firmware problem next month still reads well here.
- It is weighted by our judgment. Someone sensible could weight these factors differently and rank things differently.
- It says nothing about long-term reliability, which nobody can honestly assess at launch.
Prices
Every price on this site is a snapshot with the date it was taken, shown next to it. We do not display live prices, because a static page cannot honestly claim to have one. Amazon shows the current price when you click through, and that is the number that counts.
Corrections
If something here is wrong, we want to know and we will fix it and say that we did. Tell us what we got wrong.