Compare three to five rows from the same category. Choose five fields that matter, label each field known, unknown or mismatch, and add price only after the evidence table is complete. Remove rows with destination conflicts before scoring anything else.
Make every row answer the same question
A fair comparison needs a shared job. “Which hoodie has the clearest fit evidence?” is useful. “Which item is the best value?” is too broad because value changes with category, size, weight, documentation and your actual need.
Start with the shortlist workflow if you still have unrelated product types or more than five candidates. You only need to decide which similar rows deserve the next fifteen minutes.
Choose five fields that matter for the category
Use the same columns for every row. A practical default is destination match, photo usefulness, measurement detail, likely weight and price context. Swap one field when the category demands it: bag dimensions, shoe insole length or electronic specifications may matter more than a generic size field.
- Destination match: does the external page still describe the same product type and variant?
- Photo usefulness: do images show the angles and functional details that matter?
- Measurements: are values visible with a method you can understand?
- Weight: is the number item, packed, estimated or unknown?
- Price context: is the amount comparable with what is included in the other rows?
Use three statuses instead of a false precision score
A numerical score can hide why a row earned points. These three labels keep uncertainty visible. If you later use the seven-point checklist, retain the notes beside the total.
Build a small comparison table
| Field | Row A | Row B | Row C |
|---|---|---|---|
| Destination | Known: matching hoodie | Known: matching hoodie | Mismatch: page shows shirt |
| QC photos | Known: front, back, cuffs | Unknown: front only | — |
| Measurements | Known: chest and length | Known: size chart, method unclear | — |
| Weight | Unknown | Known: listed as item estimate | — |
| Price context | Middle of comparison | Lower, but fewer photos | Not compared |
Row C leaves immediately because its destination conflicts. Row A has stronger fit and photo evidence but unknown weight. Row B keeps a weight clue and lower price but needs better photos. The table does not invent a winner; it tells you the next useful question.
Add price after evidence, not before
When price appears first, every missing detail is unconsciously treated as a small inconvenience. Reverse the order. Complete the photo, measurement, destination and weight columns, then compare price among the rows that remain.
Shipping weight belongs beside price because a low item amount can lose context when packaging or volume is unclear. The shipping weight guide explains how to label estimates without inventing costs.
Give every row one of four outcomes
- Continue: evidence is specific enough to inspect the external details more closely.
- Ask one question: one important unknown could change the comparison.
- Hold: the row may be relevant, but better-documented options exist.
- Remove: destination mismatch or several missing essentials make further work inefficient.
Write the outcome and the reason in one sentence. “Ask one question: packed weight is the only important unknown” is actionable. “Maybe” is not.
Outcome: ask one question
Reason: photos and measurements are useful; packed weight is unknown
Do not infer: final cost, fit guarantee or seller reliability
Next check: current external product details
What a good comparison produces
You should finish with one or two rows that have fewer unresolved questions, plus a record of why the others were removed. That makes a future update easy: when a destination or measurement changes, you know which decision it affects.
Use the QC photo method when images are the weak column, or return to the seven-step workflow when the comparison set is still too broad.
Use the result carefully
Continue to Findsindex only after the rows share a category and the remaining unknowns are explicit. This guide cannot verify sellers, orders or outcomes.