Skip to content

Satisfaction ratings

Needs module.csat. No other configuration.

When a repair is closed, the person’s tracking page shows “How did we do?” with a one to five star rating and an optional comment.

Because it asks at the moment the person has just got their device back, on a page they already have open, in one click.

An email survey a week later asks somebody who has moved on, in a message they will not open, with a form they will not finish. The response rates are not comparable.

The dashboard carries a satisfaction summary.

The satisfaction report breaks ratings down by star for a period, and exports.

Ignore it for a term. An average over four responses is not a measurement.

A single one-star is not a trend. Somebody was having a bad day, or their device was away for three weeks for reasons outside your control.

Read the comments. They carry what the average cannot: that people keep being surprised by how long vendor repairs take, which you fix by setting expectations rather than by working faster.

Watch for clusters. Three poor ratings in a fortnight from the same year group, or on the same repair type, is a signal. Spread over a year, it is noise.

What tends to drive scores at a school desk

Section titled “What tends to drive scores at a school desk”

From schools that have run this for a while, in rough order:

  1. How long it took, against what they were told to expect.
  2. Whether they got a loan device.
  3. Whether they were told it was ready, or found out by walking in.

Two of those three are about communication, not repair quality, so the email templates repay the time.

There is no built-in follow-up workflow. A poor rating with a comment is a reason to talk to somebody, and a school desk is small enough for that to stay a conversation.

Leave module.csat off and the tracking page does not ask.

Worth considering if your desk is two people and everybody knows them. Asking for a rating in a school where the technician is somebody’s form tutor produces politeness rather than data.