run-competitive-analysis

Anchor the analysis to one decision sentence, pick 3-5 competitors across direct, analogous, and status-quo rings, then define dimensions and tasks before touching any product. Execute identical tasks on each product, score with behavior-anchored marks, and end every matrix row with a design implication.

Holistic evaluation runs the same core tasks across products and ranks overall UX quality per dimension, suiting kickoffs or annual checkpoints. Focused evaluation decomposes one flow into steps and compares step-by-step, suiting an imminent specific design decision.

Include 3-5 competitors total: fewer becomes anecdote, more never finishes. At least one must come from a non-direct ring, such as an analogous product or the status-quo workaround like a phone call or spreadsheet.

Skip it when the question concerns the general market rather than specific competitors, when you need a deep heuristic review of one product, or when you already have teardown evidence and need positioning. Those cases belong to desk research, heuristic evaluation, or positioning skills.

Averaged scores hide tradeoffs and are unfalsifiable. Behavior-anchored marks with observed evidence per cell preserve the tradeoffs, which are themselves the finding that drives the design implication.

Not without interrogation. Ask whether the pattern actually works, why it exists, and whether it transfers to your audience and business model, since many patterns exist for monetization or legacy reasons rather than user value.