The 2% kill criterion
Why we flag low-performing widgets, and when to act on the warning.
If a widget's open rate drops below 2% for four consecutive weeks, we flag it in your dashboard. This is the kill criterion — a built-in check that helps you decide whether a widget is earning its keep.
Why we flag this
If only 2% of the visitors who load your widget actually open it, the widget is mostly silent. You're paying for a feature that isn't pulling its weight.
Most SaaS tools won't tell you this — they'd rather you keep paying. We'd rather you fix it or cancel than quietly churn out months later when you finally notice.
When to act on the flag
A flagged widget isn't broken — it's underused. Diagnose before you pull it:
1. Is the bubble actually visible? On mobile, the bubble can be covered by a sticky chat icon or cookie banner. Check on your phone.
2. Is the welcome video doing its job? The auto-looping welcome is what signals "this is interactive, click me." If it's plain, slow, or off-tone, visitors won't engage. Try a shorter, friendlier welcome script and see if the rate recovers.
3. Is the widget on the right page? A widget on your "About" page will see less engagement than one on your pricing page. Move it.
4. Are the questions interesting? Boring questions don't get clicked. Look at the top-questions report — if even your most-clicked question has low absolute numbers, your questions don't match what visitors are looking for.
When to pull it
If two or more of the above don't move the needle, the widget probably isn't right for that page. Options:
- Move it to a different page. A pricing-page widget might thrive on a product-detail page.
- Retire it. Delete the widget. The embed code can be left in place (it'll silently not render) or removed.
- Downgrade your plan. If multiple widgets are flagged, you might not need the higher tier.
How we calculate it
- Open rate = opens ÷ loads
- Sampled over the most recent 30 days
- Threshold: 2%
- Sustained: must be below 2% for four consecutive weeks before we flag
We don't flag new widgets (fewer than 14 days old). They need time to gather data.
Can I turn it off?
Not yet — we think it's important enough to keep visible. If you have a strong reason to ignore it (e.g. you're using the widget for a niche use case where 2% is genuinely fine), tell us; we'll work out how to handle the case.