Product feedback survey

A vague 'how are we doing?' survey gets vague answers. This one ties every answer to a task the person just tried. You find out where the product helped, where it got in the way, and the one change they want most.

Best for: Product teams collecting in-app feedback to decide what to build next.

Use this template free

1 of 7

What were you trying to get done in the product today?

Made with Enform

A live preview of this template. Click through it. Nothing gets sent.

The questions

1What were you trying to get done in the product today?Long textTies the feedback to a real task instead of a general mood.
2What worked well while you were doing that?Long text
3Where did it get in your way or slow you down?Long textPeople remember the moment they got stuck far better than abstract likes and dislikes.
4If you could change one thing about the product, what would it be?Long textAsking for one change gets you a priority, not a wish list.
5How easy was it to do what you came to do?Opinion scaleA 1-7 ease scale. It gives you a number to watch next to the written answers.
6How often do you use the product?Single selectDaily / Weekly / Monthly / Rarely. Lets you weigh feedback by how much someone uses it.
7What's your role?Short textLets you split answers by group instead of averaging everyone together.

Sample replies, counted

These replies are made up to show the method.

Asked: If you could change one thing about the product, what would it be?

Search. I type the exact project name and it still shows me five wrong ones first.Better search
let me edit more than one row at a timeBulk editing
Make search find things inside comments, that's where half our notes live.Better search
The phone app logs me out every couple of days.Mobile app fixes
Selecting 40 items and changing the status in one go. Right now it's 40 clicks.Bulk editing
Search is slow on big workspaces, like 3-4 seconds.Better search
Mobile, honestly. I can't attach a file from my phone.Mobile app fixes
Dark mode pleaseOther

8 replies, one label each
Label	Replies
Better search	3
Bulk editing	2
Mobile app fixes	2
Other	1

Search comes up in three of eight replies, and each one names a different flaw: ranking, coverage and speed. Start there, then read the bulk editing replies together, since both describe the same repeated clicking.

How to run it well

Show it after a real action, not on a timer. Feedback right after a task is specific. Feedback at a random moment is mood.
Count the 'one change' answers. The change named most often is your next ticket.
Split by how often people use the product before you act. Daily people and first-week people are describing different products.
Paste the exact words about friction into the ticket. A paraphrase loses the detail that tells you what to fix.

When not to use this: If you need a single loyalty number to track every quarter, this is the wrong tool. Use the NPS survey and keep this one for the reasons behind it.

FAQ

How is this different from an NPS or CSAT survey?

NPS and CSAT give you a score. This gives you the reasons, tied to a task. Run them together: the score shows whether things are moving, and this shows what to ship.

How many replies before I act on it?

Enough that the same answers start repeating. With a handful of replies, treat a pattern as something to check in a few follow-up calls, not a roadmap decision.

Won't a single 'one change' question miss things?

That's on purpose. Asking for everything gets you a long flat list. Asking for one change ranks things for you, and the open friction question still catches the rest.

Where the method comes from

Open-Ended vs. Closed Questions in User Research, Nielsen Norman Group (2024)nngroup.com
Writing Survey Questions, Pew Research Centerpewresearch.org