Feedback

Feedback comes in two shapes. One is triggered by an event: a support ticket, a purchase, a bug report. The other is a regular measurement that goes to the same audience every month or quarter so you can see a trend. The first needs a short form on one screen. The second needs the same questions every time, or the numbers stop being comparable.

When you cannot predict why people will write in, start with the branching feedback form. The respondent picks a reason first (a bug, an idea, praise, a complaint or a question) and only sees the questions for that branch. For loyalty there is the classic NPS survey: one 0 to 10 question, the reason behind the score and an optional email. The maths is simple: the share of 9–10 scores minus the share of 0–6 scores.

Support quality is better measured per ticket, so the CSAT survey rates one interaction: an emoji scale, a matrix across aspects of the reply and a check on whether the issue was solved first time. For a fuller picture of the product, the product review template combines stars, NPS, a matrix of characteristics and a ranking of features by importance.

Where the form appears matters more than its wording. A link in a post-purchase email reaches buyers. A widget on the help page catches people while something is broken. A QR code on a receipt or a table card covers offline visits. For a recurring measurement, keep the template and the audience fixed: if you asked every customer in March and only active ones in June, a higher score means nothing.

Why NPS scores often mislead, and how to avoid it, is covered in NPS without lying to yourself. The full job, from questions to putting the form on your site, is on the feedback form page, and loyalty tracking on its own has a separate NPS survey page.