Define the comparison before collecting links
Choose the customer segment and question that matter. “How do notes tools describe value to a five-person creative team?” is more specific than “What do our competitors do?” Set a time window and a small relevant company set.
Use the same questions for every company. Otherwise, the competitor with the longest feature page can appear strongest simply because more material was available. Mark missing information as missing.
Separate company claims from other perspectives
Public product pages are useful evidence of how a company presents itself. Documentation can describe features and constraints. A customer review offers a reported experience. None of those sources should be silently treated as the same kind of evidence.
Keep the edition and date beside pricing and capabilities. A current enterprise feature does not establish availability on a lower plan, and an older review may describe a product that has changed.
Build an evidence table
Use rows for Positioning, Product, Pricing, Distribution, Strengths, Weaknesses, and Customer sentiment. For each statement, retain the original link, the observed wording or a faithful paraphrase, the date, and the scope of the claim.
Keep an inference column separate from sourced observations. “The homepage emphasizes quick capture” is an observation about messaging. “They probably have higher retention” is an unsupported inference unless you have relevant evidence.
Avoid false precision
A few public reviews are not a representative customer survey. Similar language across several articles may come from one press announcement. Repetition and availability can distort a comparison if you treat every URL as independent evidence.
Do not estimate confidential revenue, acquisition costs, or conversion rates from a polished website. Keep useful questions open and identify the source that would be needed to answer them.
Turn the comparison into a next step
Summarize the pattern relevant to your segment: which needs are emphasized, which claims require verification, and which questions you want to test with your own audience. Competitor research informs a decision; it does not establish what your customers want.
Create a Competitors project in Donde to organize selected public sources. When the question becomes which positioning or growth experiment to try, use Product Growth for that separate decision. Donde does not automatically monitor companies for changes.
