Skip to content

We buy nothing on trust. Every recommendation starts from what goes wrong.

How We Analyze

This page describes how an article on this site is put together, so you can judge how much weight to give it.

What we do

Each analysis starts from the failure side rather than the feature side. Before looking at any specific product, we establish how the category fails: what goes wrong, what wears out, what catches buyers by surprise, and why.

We do not buy or test products. The inputs for an article are manufacturer specifications, manuals and warranty terms; published standards where a category has them (UL, ASTM, ANSI, FMVSS and similar); regulator data and recall notices (CPSC, NHTSA, FDA, EPA, FTC); and published studies. Every source we rely on for a figure is linked in that article’s Sources list. The evidence standards page sets out what counts as a source and what does not.

What the Fear Index is

A 0–100 editorial estimate of how much can go wrong when a typical buyer shops a category: safety risk, wasted money, counterfeits and scams, hidden costs, and compatibility traps. It’s a judgment we make from the evidence above, not a lab measurement and not a customer rating.

  • 0–33 — Low risk. The category is forgiving. Most reasonable options work out.
  • 34–66 — Think twice. There is a real trap here, and it catches a meaningful share of buyers.
  • 67–100 — High risk. The default choice is frequently the wrong one. Read before buying.

How we score it

Each Fear Index is built from five factors, scored 0–4 and multiplied by 5, for a total out of 100 rounded to the nearest 5:

  • Severity — how bad the dominant failure is: a safety issue, wasted money, or lost time.
  • Likelihood — how often it happens, backed by cited evidence: recall counts, warranty terms, standards, or published studies.
  • Detectability — whether a buyer could have spotted the problem before purchase, from the listing alone.
  • Recoverability — how easy it is to undo the mistake: return window, repair options, warranty coverage.
  • Market noise — how often listings in the category misstate or bury the spec that actually matters.

This is the method we’re applying going forward. We’re re-checking existing Fear Index scores against it; a score that hasn’t been re-checked under this rubric yet should be read as provisional.

Shortlist tables also show a Risk estimate next to each product type. That’s the same editorial estimate applied to a narrower category — lower is safer. It is explicitly not a customer rating, a star rating, or anything supplied by Amazon or any retailer; it’s our judgment, and it has no bearing on which links appear.

What we will not claim

We do not run laboratory tests, and we do not pretend to. Where a figure comes from a manufacturer, we attribute it to the manufacturer. Where it comes from a standards body or a published study, we cite it. Where it’s our judgment, we say so in those words.

We don’t show current Amazon prices on this site. Prices and availability change too often for a static page to keep up, and only Amazon-approved tools can display them accurately — so pricing is left to Amazon’s own page.

We do not accept products from brands on terms that involve any editorial condition, and we do not accept payment for placement of any kind.

Updating

Articles are revisited when a product in them is discontinued or revised, when new evidence changes a recommendation, or when a reader sends us something that contradicts what we wrote. Every article shows the date it was published. A note under the byline saying an article was updated, and what changed, appears only when that actually happened.