A shortlist you can explain.

Why unknown is not zero

Zero is an entered judgment at the bottom of your scale. Unknown means you have not made a supported judgment. The worksheet substitutes zero only for the lower bound and five only for the upper bound. Neither endpoint is a prediction. Do not interpret the middle as an expected result.

What overlapping ranges mean

If A spans 2.50–3.75 and B is 3.25, either could lead once A’s missing score is resolved. Even non-overlapping preference ranges do not settle a missing mandatory requirement. Gather the evidence with the greatest decision consequence first.

A consistent scoring rubric

Define a score before looking at brands. For care-routine fit, for example, 0 could mean an unacceptable documented task and 5 could mean tasks the owner explicitly accepts. Intermediate values express preference, not hours or measured durability. Use the same definition for every candidate and explain disagreements in the notes.

Sensitivity is a conversation

The sensitivity table varies one weight from 0 through 5 while holding the other weights and scores fixed. It shows whether your conclusion depends heavily on one preference. It does not assign probabilities to preferences or automate a purchase decision.

What this tool does not calculate

No structural capacity, slip safety, surface temperature, fire approval, compatible fastener, lifecycle cost or replacement date is calculated. Follow the exact product instructions and responsible professional review for those matters. For ownership-cost assumptions, use Deck Lifecycle; for attachment evidence, use Fastener Compatibility.

Keep a useful decision record

Export candidate identity, the mandatory screen, each rating, source revision and open question. Reopen the decision if the supplied product, application or source revision changes. A later reviewer should be able to distinguish new evidence from a changed personal preference.