Skip to content
English
  • There are no suggestions because the search field is empty.

Suspect clicks

If you've ever looked at your analytics and seen weirdly high click counts that don't match your audience size — those are likely suspect clicks. This...

If you've ever looked at your analytics and seen weirdly high click counts that don't match your audience size — those are likely suspect clicks. This article explains what they are and how rasa.io handles them.

What's a suspect click?

A suspect click is a click that's likely not from a human reader. It comes from:

  • Bots — automated tools crawling email content
  • Corporate security scanners — many enterprise IT setups scan every link in every incoming email to check for phishing or malware
  • Email previewers — services that pre-fetch content for inbox previews
  • Caching systems — automated systems that hit links to cache content

These all look like clicks to a tracking system, but no human ever read your article.

Why this matters

If you can't separate human clicks from bot clicks, your analytics get noisy:

  • Click rates look artificially high
  • Top Articles get distorted by scanner activity
  • Comparisons across time periods become unreliable
  • Your engagement signals to the AI personalization engine get noisy

Filtering out suspect clicks gives you a more honest picture of what's actually working.

How rasa.io identifies suspect clicks

rasa.io looks at signals like:

  • Click timing (multiple clicks within milliseconds of each other usually = bot)
  • User agent strings (some scanners identify themselves)
  • IP patterns
  • Click sequence (clicking every link in an email is bot behavior; clicking 1-2 is human)
  • Known bot signatures

If a click matches enough suspect patterns, it gets flagged.

Where to find the toggle

Every analytics report in rasa.io has a Without Suspect Clicks toggle:

  • On (Without Suspect Clicks) — shows you only the cleaner, more human signal
  • Off (All Clicks) — shows you everything including bot traffic

The default varies by report — some default to filtered, some to unfiltered. Look for the toggle at the top of the report.

Which view should you use?

Use Without Suspect Clicks for:

  • Reporting to leadership or stakeholders
  • Comparing performance over time
  • Evaluating which content actually resonates with readers
  • Most "is this newsletter working?" questions

Use All Clicks for:

  • Diagnosing odd patterns
  • Investigating deliverability (sometimes a flood of bot activity indicates a specific issue)
  • When you suspect the suspect filter might be too aggressive
  • Total inbox-level activity questions

Some common situations

"My click rates suddenly dropped after switching to Without Suspect Clicks."

That's normal — the previous numbers were inflated by bot traffic. The new numbers are more accurate.

"My click rate doubled overnight."

Could be:

  • New B2B audience added (more corporate security scanning)
  • A specific subscriber's organization changed their security settings
  • An issue with your send infrastructure

Switch to Without Suspect Clicks to see if it's a real change or just scanner activity.

"Articles I never expected to do well are showing huge click counts."

Check Without Suspect Clicks. If the article still shows high clicks, it's a real surprise hit. If it drops dramatically, it was scanner-driven.

How AI personalization handles suspect clicks

For personalization purposes, rasa.io's AI primarily weights real human engagement. Suspect clicks have less influence on what gets recommended to subscribers. That's intentional — personalization should reflect human interest, not bot patterns.

Best practices

  • Default to Without Suspect Clicks for most strategic decisions
  • Check both views when something looks unusual
  • Don't celebrate spikes before checking that they're not bot-driven
  • Don't panic about drops in click rates — they might just be a filter change

What's next

  • Analytics overview — the full map of reports
  • Metrics defined — all the metrics explained
  • Email Health Report — overall list health