GuidesAI automation
Automated competitive monitoring with AI
What to watch
- PricesPrices, plans, terms.
- OffersNew products, pages that appear.
- Job adsOpen roles reveal the strategy.
- ReviewsWhat their customers like or criticise.
The flow
- 1CollectionOnce a week, the main text of each page.
- 2ComparisonDifference with the previous version.
- 3SummaryThe model only reads the differences.
- 4DeliveryA short note by email or in Slack.
The summary instruction
Here are the changes detected this week on our competitors' pages.
For each one: what changed, in one sentence, and why it may matter to us.
Ignore layout, date or cookie changes.
The rules
| Rule | Why |
|---|---|
| Public pages only | No account, no workarounds |
| Slow pace | One visit per page per week is enough |
| No personal data | GDPR applies to monitoring too |
| Link to the source | To check before acting |
Choosing the tool: n8n, Make or custom code. The bigger picture: AI automation.
Frequently asked questions
- Is it legal?
- Reading public pages generally is, while respecting the sites' terms of use, their robots.txt and a reasonable pace. No personal data is collected.
- Why compare before summarising?
- So you don't pay the model to reread identical pages, and so the summary only covers what's new.
- Which sources say the most?
- Pricing pages, job ads (they reveal projects), release notes and customer reviews.