Enter how many people you invited, how many the invitation reached, how many started, and how many finished.
Response rate measures how many of the people you invited actually took part. It's usually reported as completion rate — completed responses divided by delivered invitations — but "start rate" and "completion among starters" tell you different things: a low start rate points to a weak invitation or subject line, while a low completion-among-starters rate points to a survey that's too long or confusing once people are in it.
| Metric | What it tells you |
|---|---|
| Completion rate | The headline number — the share of your delivered audience who gave you a usable, complete response |
| Start rate | How compelling your invitation and subject line were at getting the click |
| Completion among starters | How well the survey itself held attention once someone opened it |
You send 1,000 invitations. 960 are delivered (40 bounce). 600 people open and start the survey. 505 finish it.
It depends heavily on channel and relationship: internal employee surveys often see 60–90%, customer email surveys typically land between 10–30%, and cold outreach to strangers can be under 5%. Compare your rate against your own past surveys and channel, not a single universal benchmark.
Delivered is the more honest number — invitations that bounced or were never received couldn't possibly have been answered, so including them in the denominator understates your true response rate.
That pattern means your invitation is working (people are opening the survey) but the survey itself is losing people — usually because it's too long, has a confusing early question, or asks for sensitive information too soon.
A low response rate is only a problem for accuracy if the people who responded are systematically different from those who didn't (nonresponse bias) — a small but representative sample can still be highly accurate.