Research a competitor's public pages

The output is a short table of facts, each dated, each with the URL it came from. What makes it useful six weeks later is that every row can be re-checked; what makes it worthless is a single unsourced claim mixed in with the rest.

Steps

1. List the pages worth reading before reading any: pricing, the product pages for the areas that overlap, documentation, changelog or release notes, and the careers page. Changelog and careers are the two that tell you where they are going rather than where they are.
2. Fetch each page's content rather than reading a search summary of it. Summaries are where stale prices come from.
3. Record for each fact: the value, the exact URL, and the date you fetched it. Pricing especially — it changes without announcement and an undated price is a liability in any internal document.
4. Note what is absent. A missing page, a removed feature, a pricing tier that disappeared since the last time somebody looked. Absence is often the most informative finding and it is invisible unless you were looking for it.
5. Separate what they say from what it does. Marketing copy goes in quotes and is labelled as a claim. What their documentation says is achievable is a different column.
6. Write the comparison as a table of dimensions that matter to a buyer, with a row for each, and a short paragraph under it of what actually differs. Where they are genuinely better, say so in that paragraph — a comparison that finds no strengths is one nobody internally will trust.

Stay on the public side

• Only pages anyone can reach. No signing up under a false name, no using somebody else's login, no pages behind a customer portal.
• Respect the site's crawling rules and fetch politely. This is research, not a load test.
• Do not copy their text into anything the company publishes.