What is a held-back group?
A share of the guests a campaign targets who do not receive it. Comparing them with the guests who did shows what the campaign changed.
Measurement
Status: BetaCampaign results measured against a held-back group
Purple Engage measures the extra visits and POS revenue a campaign brought in against a group of guests who did not receive it, and filters out the opens and clicks no person made.
How does Engage measure whether a campaign worked?
By comparing the guests who received it with a held-back group who did not. The difference is the campaign's lift in visits, bookings and POS revenue, shown with a confidence range so you can tell a real result from noise. An open rate cannot tell you that.
- Lift in visits, bookings and POS revenueStatus: Beta
- A confidence range on every resultStatus: Beta
Why do my open rates look different in Engage?
Because Engage counts people, not machines. Opens from Apple Mail Privacy Protection and clicks from link scanners are filtered out. Where opens cannot be measured, Engage shows "not measured" instead of 0%, so a quiet campaign and an unmeasurable one never look the same.
- Apple Mail Privacy Protection opens and link scanners filtered outStatus: Beta
- "Not measured" shown instead of 0%Status: Beta
Can I see where guests clicked in an email?
Yes. The click map lays clicks over the email as it was sent, so you can see which offer, button or link drew attention. Clicks from link scanners are already filtered out, so the map shows what people did.
- A click map on the sent emailStatus: Beta
Can I compare campaigns with each other?
Yes. Pick two to six campaigns and Engage lines up their results with confidence intervals, so you can see whether one did better or the difference sits inside the margin of error.
- Compare two to six campaigns with confidence intervalsStatus: Beta
How does Engage decide an A/B test winner?
It tests for significance and gives a verdict. When the gap between versions is not significant, Engage says "too close to call" rather than naming a winner by a fraction of a percent. You can test the subject line, the sender or the whole email.
- A significance verdict, including "too close to call"Status: Beta
- Tests on the subject line, the sender or the whole emailStatus: Beta
How does the AI agent show its suggestions work?
Every campaign the agent proposes is a controlled experiment. Engage holds back 5% of the audience from each agent-proposed send and measures the lift against them. The agent can propose, but a person approves every send before it goes.
- 5% held back from every send the agent proposesStatus: Beta
- A person approves every agent-proposed sendStatus: Beta
Questions
Why does Engage show "not measured" instead of 0%?
Because 0% would say nobody opened. When opens cannot be measured, for example because of Apple Mail Privacy Protection, Engage says so.
Does Engage report POS revenue from campaigns?
Yes, from connected POS systems. Campaign lift includes POS revenue, and the offer report shows the revenue captured from redemptions.
How many campaigns can I compare at once?
Between two and six, with confidence intervals on each.
Find out what your last campaign was worth
The free plan includes 500 Actions a month and unlimited contacts, with no card needed.