How we test · The criteria behind every score

One scale. Five criteria. Shown in the open.

A score you can't inspect is just a vibe with a decimal point. Here's exactly what our numbers weigh, how the scale reads, and what a score does — and doesn't — claim.

The rubric

What a score weighs

SCORING RUBRIC — five criteria, weighted to total 100%.
Criterion Weight The question we're answering
Does the job 30% Does it handle the work its class exists for — without bogging, stalling or excuses?
Build quality 25% Will it survive a decade of drops, dust and being loaned to your brother-in-law?
Ergonomics 20% Weight, balance, grip, controls. A tool that's unpleasant to hold doesn't get used.
Value 15% Is the price fair for what you get — counting batteries, blades and the platform around it?
Living with it 10% Runtime, noise, storage, blade/bit changes — the everyday friction nobody puts on the box.

Weights are fixed across a category so scores compare fairly within a table.

The scale

How to read a score

The process

Four steps, in order

  1. Define the class, pull the real specs

    We draw the class boundaries first, then take the numbers — weight, voltage, capacities, what ships in the box — straight from manufacturer documentation, not from memory or a retail listing.

  2. Synthesize the hands-on record

    We don't bench-test units ourselves. We read the people who do — established expert reviewers with real test benches — and cross-check them against long-term owner feedback, where the dead batteries and wobbly chucks actually show up.

  3. Score inside the class

    Everything runs through the fixed rubric above, same weights for every tool in a table. A 12V drill is scored against what a 12V drill should be — not against a job-site hammer drill.

  4. Revisit as the market moves

    Models get replaced and street prices drift. When a price shift breaks a value score or a successor redraws the class, the verdict gets re-run — not left to age.

What a score is

Our informed judgment from three inputs — manufacturer specs, published expert testing, and long-term owner consensus — applied through the rubric above and checked against the class. It's an opinion with a method, not a lab measurement, and we won't dress it up as one.

What we don't do

No sponsored reviews. No pay-for-placement. No scores softened to keep a brand happy. And no pretending we bench-tested a unit we didn't — when a finding comes from someone else's testing, we say so. If any of that changes, this page will say so in the first sentence. It won't.