Competitor research becomes useful when it is repeatable. A monthly checklist prevents teams from collecting screenshots without comparing them to the previous period.
Use this agenda to produce a concise evidence brief and a small test backlog.
Confirm scope and sources
Review why each store is tracked and whether it still overlaps with the target customer, category, geography, or channel. Remove sources that repeatedly fail to inform decisions.
Check that links, timestamps, currencies, and comparison units are present before interpreting changes.
Review material changes
Compare current and previous snapshots for launches, removals, price moves, discounts, availability, collection emphasis, reviews, and landing-page messages.
Group changes by category and competitor role. Highlight patterns involving multiple stores rather than presenting every event equally.
- New and removed products
- Material price and discount changes
- Restocks and persistent unavailability
- Recurring review themes
- Offer, proof, and CTA changes
Separate facts, hypotheses, and estimates
Facts should link to observed public evidence. Hypotheses explain possible meaning. Estimates need a method and uncertainty label.
This separation protects the meeting from confident stories built on incomplete public data.
Assign no more than three actions
For each action, name an owner, customer or commercial hypothesis, metric, smallest test, and decision date.
Record deliberate no-action decisions too. They stop the same competitor change from reopening an unproductive debate next month.
Frequently asked questions
How long should a monthly competitor review take?
A focused review can fit within 45 to 60 minutes when snapshots and evidence are prepared in advance and discussion centers on material changes.
Who should attend?
Include the smallest cross-functional group needed for the decisions, often merchandising or product, growth, and an operator responsible for economics or fulfillment.
What is the output of the review?
A short evidence brief, explicit no-action decisions, and no more than three owned tests with metrics and review dates.