How the TrustScore works

Every product on this site is scored 1โ€“10 as a weighted average of four sub-scores:

Sub-scoreWeightWhat it measures
Quality & depth35%How substantive the product is versus its promises โ€” depth of the reading, program or report.
Value for money30%What you get for the price, including cheaper alternatives that exist.
Delivery & UX20%Speed, reliability and polish of the purchase-to-product experience.
Support & guarantee15%Refund policy reality, support responsiveness, company track record.

This page covers the scoring rubric. For what we check and how we label tested vs research-based reviews, read how we review.

Our editorial rules

1. Research first, and we label what we tested. Every review is built on the vendor's real checkout, terms and refund policy, plus public complaint records (BBB, Trustpilot, regulators). Where we bought or used a product ourselves, the review says so; where we did not, it is a research-based review. How we review.

2. The cons are mandatory. No review publishes without at least three genuine drawbacks. If we can't find three, we haven't looked hard enough.

3. Scores are opinions, dated. Every review carries its update date. Products change; so do scores.

4. Belief-based products get honest framing. Psychic readings, numerology and manifestation are not science, and we never pretend otherwise. We score the product execution โ€” delivery, depth, fairness โ€” not whether metaphysics is true.

5. Commissions don't move scores. We earn affiliate commissions on most products reviewed, including products we score 7.2. The ranking is ours; the disclosure is on every page. Full disclosure.

6. Supplements are scored as products, not medicine. We evaluate formulation transparency, vendor track record, refund reality and independent customer sentiment โ€” and we never imply treatment of any condition. Products making claims they cannot support lose points for it.

What the sub-scores mean in practice

The weighting is published above, but the weights are only useful once you can picture what they do. Quality and depth at 35% is the largest because it is the hardest to fake: does the product do the thing its sales page says, and is there substance behind it or just presentation? Value at 30% is a comparison against what else exists at the price, not a judgement about whether the price is affordable โ€” an expensive product can score well here if the alternatives are worse and cost the same.

Delivery and UX at 20% covers the purchase-to-product experience: how long it takes, whether it works, whether the checkout does anything surprising. Support and guarantee at 15% is the smallest weight but the most predictive of regret, because it captures whether the refund is real or theoretical. A product can score well on all three of the others and still be a bad buy if you cannot get your money back.

What the scale is not

A TrustScore is not a percentage and not a probability. Scores cluster in the 6 to 8 range because most products in these categories are competent but unremarkable, and a 7 is not a failure. What matters more than the number is the reason behind it, which is why every review sets out what cost the product points rather than only publishing the total.

Two limits are worth stating plainly. We cannot test how any product holds up over months of real use, so a score describes the product as it is now. And where a capability sits behind a sales call or an enterprise plan, we could not test it directly โ€” we say so rather than repeating the vendor's description of it.