Does anybody have any analytics data on typical engagement with this?
We tried an alternative on Explore careers where we made the survey banner sticky and adjusted the design to accommodate. Currently seeing around 2% of users providing feedback through the Yes / No links. Just wondering how that compares to non sticky.
Full write-up and further research can be found here:
Dropbox Paper audit
On 19th March 2019 the Design System team reviewed a Dropbox Paper document discussing the Feedback links pattern.
The aim was to reduce the number of places containing guidance and code by:
Below is a record of the outcome of that review.
If you need to, you can see the original Dropbox Paper content in the internet archive.
Combine the document history discussion on Dropbox Paper with this issue and remove the original Dropbox Paper page.
Improve your service by helping users tell you what they think of it.
When to use this pattern
All public beta or live services must have a feedback page so you can collect comments and measure user satisfaction.
GOV.UK information page feedback link
Here’s how the feedback links at the bottom of pages on GOV.UK work:
GOV.UK Service Manual page feedback link
On Service Manual pages the feedback form is combined with a simple yes/no link for tracking user satisfaction:
GOV.UK user survey
Every 50 page-loads we insert a banner: “Tell us what you think of GOV.UK – take the 3 minute survey". This takes you to this survey: https://www.surveymonkey.com/s/6HZFSVC
On one service we eliminated the feedback link and put the feedback questions directly on the feedback page. The feedback rate jumped from 2.5% (average over 6 months) to 45% (average over 1 month).
Can anyone think of a significant disadvantage to putting feedback questions directly on the confirmation page?
@terrysimpson99 this pattern is used on many different types of pages, so I imagine putting them inline might not work so well for a generic use case.
However, you might want to consider raising your findings on the confirmation pages pattern page, perhaps we could consider updating that pattern with what you've found?