Add a help center survey to the end of every article, ask readers who search twice or give up, and see what each page’s readers were trying to do.
Place “Was this article helpful?” under your article template once, and every article asks it. Readers who answer No are asked what could be better, readers who answer Yes are done in one tap, and each answer records the article it came from.
Some readers never reach the end of an article. These moments catch them where they lose the thread, each with what opens the survey, who it shows to and a template to start from.
Usertick notices a reader hopping from page to page, searching again with new words or going straight back. Ask what they were trying to do, and you learn which article is missing.
A search with no results is a question your help center can’t answer yet. Ask what they were looking for the moment the empty results appear.
A reader who copies a code sample is about to try it. Ask how the guide could be better while the steps are still on screen.
Ask how hard the tutorial was once most of the video has played, and learn which lessons need a slower pace or a written version.
Opening the same article again in one visit is a sign the first read didn’t solve it. Offer a person before the reader gives up and writes in.
Give readers a way to report an outdated step or ask for a missing topic whenever they find one, with a button that stays on the edge of the page.
One survey covers the whole help center, and each answer records the page it came from.
Results lists every article readers answered on, with its share of the answers. Select one, and every chart shows that article’s answers alone.
Add a rating question, and each article gets its own score beside the overall one, with a mark on the articles that score clearly lower.
Group answers by page on a dashboard card, and keep the articles that need work in front of the people who write them.
AI reads every comment, and workflows send each gap to the people who write the articles.
AI files each comment under a theme such as single sign-on or invoices, in any language, so the missing topics show up as counts.
A workflow posts each No with a comment to your docs team’s Slack channel, or opens a Linear or Jira issue, on every plan.
When a reader asks for help, Zendesk gets a ticket with the answers in an internal note, on every plan.
Paste the install code into your help center’s theme, or add it through Google Tag Manager. Any help center or docs site that lets you add a script works.
Choose Embedded, name the element at the end of your article template, and the survey appears under every article that has it.
Set the pages to start with /help/, publish, and every article starts collecting answers within minutes.
Readers who reach the end can judge the whole article. Put the question under the last step, never at the top.
Start with a question that takes one tap, and ask what could be better only to readers who answered No.
A task like “turn on single sign-on” tells you what to write next, and a score alone never does.
Ask when search comes back empty or when a reader can’t find the answer. Those answers come from readers who never reach the end of an article.
Filter by page, start with the articles that get the most No answers, and check them again after each rewrite.
Update the article, mark the answers Resolved, and reply to the readers who left an email.
Start from a survey written for articles, failed searches or tutorials, or from a single question that fits the page.
Create one survey from the Help article feedback template, choose Embedded and name the element at the end of your article template. Set the pages to start with your help center’s path, and the question appears under every article, with each answer recording the page it came from.
It works on any help center or docs site where you can add a script to the pages, either in the theme or through Google Tag Manager. A hosted help center that doesn’t allow your own scripts can link to the survey’s own page instead.
Yes. Results lists every page readers answered on, and selecting one filters every chart to that article. With a rating question, each page also gets its own score, and the pages that score clearly lower are marked.
Yes. Branching sends readers who answer No to the next question, such as what could be better, and ends the survey for readers who answer Yes.
Yes. Answers carry no name unless your help center tells Usertick who is signed in, and Anonymous responses on the survey keeps every answer anonymous even then. A Contact question can ask for an email from readers who want a reply.
No. The widget loads after the page without holding it up and draws in its own isolated layer, so your theme’s styles stay as they are. If Usertick can’t be reached, the article works as usual.
Ask whether they’d like someone to help, and send the Yes answers where your team sees them first: a Slack or Microsoft Teams channel, or a Zendesk ticket with the answers in an internal note, on every plan.
The embedded question, the feedback button, page rules and the Zendesk, Jira and Linear connections work on the Free plan, and so do the no-results, copied code and video triggers on 1 survey. Starter adds those triggers on every survey. Can’t find it and Hesitation are on the Pro plan and above.
Start free, see what your customers mean, and get each answer to the person who can act on it.
