{"id":9096,"date":"2025-03-12T14:24:01","date_gmt":"2025-03-12T18:24:01","guid":{"rendered":"https:\/\/canny.io\/blog\/?p=9096"},"modified":"2026-03-16T15:23:57","modified_gmt":"2026-03-16T19:23:57","slug":"autopilot-experiment-with-typeform","status":"publish","type":"post","link":"https:\/\/canny.io\/blog\/autopilot-experiment-with-typeform\/","title":{"rendered":"How we automatically uncovered 100+ feature requests in Typeform\u2019s support tickets"},"content":{"rendered":"\n

Product feedback comes from everywhere.<\/p>\n\n\n\n

It\u2019s living in the conversations your support team are having with customers.<\/p>\n\n\n\n

It\u2019s sitting in the sales demo conversations that your AEs are having with leads.<\/p>\n\n\n\n

It\u2019s in the reviews that your customers are leaving on sites like G2 and Capterra.<\/p>\n\n\n

\n
\"Customer
Example of product feedback in G2<\/figcaption><\/figure>\n<\/div>\n\n\n

You already have these insights waiting for you. That\u2019s great. Tracking and organizing customer feedback<\/a> from many places can be a lot of work and take a lot of time, though.<\/p>\n\n\n\n

Here\u2019s why:<\/p>\n\n\n\n

Feedback from different places adds up fast. You need to read, understand, and sort each piece. It can take hours per week to get through it all.<\/p>\n\n\n\n

It also comes from many channels. Since different teams handle different channels, tracking it all becomes very hard.<\/p>\n\n\n

\n
\"Inside
Inside Canny Autopilot \u2013 capturing and organizing feedback from multiple sources<\/figcaption><\/figure>\n<\/div>\n\n\n

New feedback keeps coming in, so your feedback board<\/a> needs regular updates and deduplication. Managing customer feedback can feel like a job that never ends.<\/p>\n\n\n\n

Is all this time and effort spent on managing customer feedback worth it?<\/p>\n\n\n\n

100% yes.<\/p>\n\n\n\n

Reviewing customer feedback is important.<\/a> It shows you what customers want and helps you decide what to build first.<\/p>\n\n\n\n

Let\u2019s say you capture feature requests from your support tickets, sales calls, and review sites<\/a>. You’ll see what features people want most. Aggregating all this feedback helps you see what\u2019s really important and likely to drive impact.<\/p>\n\n\n

\n
\"canny
Feature requests from Intercom captured by Autopilot<\/figcaption><\/figure>\n<\/div>\n\n\n

When you update customers on feature requests, it shows that you value their feedback. This builds brand loyalty. Customers feel heard and are more likely to share feedback in the future. This starts a feedback loop<\/a> that helps you keep improving your product in the right direction.<\/p>\n\n\n\n

Getting your time back<\/h2>\n\n\n\n

You don\u2019t have to spend hours every week tracking and organizing feedback from all these sources.<\/p>\n\n\n\n

With Canny Autopilot<\/a>, capturing feedback changes from a boring manual job to something that runs in the background. Autopilot automatically captures feedback wherever it lives. It also merges duplicates and keeps feedback organized, helping you focus on the insights behind it. Then you can focus your time on prioritizing<\/a> and implementing the insights you\u2019ve gained.<\/p>\n\n\n\n

How long does it take Canny Autopilot to sort through 1,611 tickets?<\/h2>\n\n\n\n

Typeform is one of our valued customers at Canny. <\/p>\n\n\n\n

Their team didn\u2019t have a good way to capture and organize feedback<\/a> from support tickets. They needed a way to capture feature requests from the customer support tool they use, Zendesk.<\/p>\n\n\n\n

They were keen to try Canny Autopilot but worried about its accuracy\u2014they didn\u2019t want any important feedback to slip through.<\/p>\n\n\n\n

So we put it to the test: <\/p>\n\n\n\n

We ran an experiment. They had 1,611 support tickets from Zendesk to analyze.<\/p>\n\n\n\n

The team at Typeform read the support tickets and tagged conversations with feature requests.<\/p>\n\n\n\n

While they did that, we ran Autopilot\u2019s Feedback Discovery. It analyzed these same 1,611 support tickets and captured feature requests it found. The entire process took 53 minutes.<\/p>\n\n\n\n

The results came in:<\/p>\n\n\n\n

76 of the 1,611 tickets<\/strong> were manually tagged as containing feature requests <\/strong>by <\/strong>someone on Typeform\u2019s team.<\/p>\n\n\n\n

And drumroll please\u2026<\/p>\n\n\n\n

Autopilot\u2019s Feedback Discovery tagged 109 of the 1,611 tickets<\/strong> as having a feature request. Typeform\u2019s team agreed that Autopilot\u2019s results were accurate and not false positives in this experiment.<\/p>\n\n\n\n

Autopilot\u2019s deduplication also suggested what feature requests should be merged together. The client\u2019s team accepted 98.3% of these suggestions, so it proved to be very accurate.<\/p>\n\n\n\n

\n

“The feature requests from Canny are generally spot on. I was really impressed by the number of times that Canny caught a feature request that our team didn\u2019t.<\/em>“<\/p>\n\n\n\n

– <\/em>Lauren Corboy, Lead Onboarding Consultant at Typeform<\/p>\n<\/blockquote>\n\n\n\n

If they only relied on tickets tagged by their team, about 30% of their valuable customer feedback would have been missed.<\/p>\n\n\n\n

\n

“Overall, Autopilot identified 93% of the feature requests correctly, and our live team caught about 63% of those. So Autopilot is about 30% better than our live team at the moment in flagging those requests.<\/em>“<\/p>\n<\/blockquote>\n\n\n\n

The test showed that Canny Autopilot\u2019s Feedback Discovery is more accurate than manual tagging. It\u2019s also faster and frees up their time.<\/p>\n\n\n\n

They walked away with about 30% more feedback\u2014already sorted and organized.<\/p>\n\n\n\n

Their team can now see the most requested features right away on their feedback board. Now they can start building them.<\/p>\n\n\n\n

\n

“Using the Autopilot feature has just made it so much easier for us to track and manage requests that are coming at scale through those tickets. We know that requests that are coming in through those tickets (7,500 a month) are being tracked accordingly. And it\u2019s a lot less effort for us to parse those into the correct existing posts. <\/em><\/p>\n\n\n\n

Because they\u2019re coming from our ticketing system, we can identify who\u2019s asking, and the Autopilot tool is giving a really great summary of what the customer request is.<\/em>“<\/p>\n<\/blockquote>\n\n\n\n

And it\u2019s not just this one experiment that highlights Canny Autopilot\u2019s accuracy. Here are some more Canny users who\u2019ve saved time and added value with Autopilot:<\/p>\n\n\n

\n
\"Testimonials
Testimonials from Canny’s Autopilot customers<\/figcaption><\/figure>\n<\/div>\n\n\n

Conclusion: Capture more feature requests with Autopilot<\/h2>\n\n\n\n

Don\u2019t miss out on capturing the valuable feature requests that are sitting in your support tickets.<\/p>\n\n\n\n

Autopilot\u2019s Feedback Discovery<\/a> can accurately detect and organize feature requests coming from your customer conversations. Try it today with our Intercom, Help Scout, Freshdesk, and Zendesk integrations and see how much feedback is out there.<\/p>\n\n\n\n

They can see the most requested features right away on their feedback board. Now they can focus on building them.<\/p>\n\n\n\n

Start capturing feature requests in your support tickets with Canny Autopilot:<\/p>\n\n\n\n

\n \n \"try-autopilot\n <\/a>\n<\/div>\n\n\n\n

<\/p>\n","protected":false},"excerpt":{"rendered":"

Autopilot detects 30% more feature requests than humans. See how our experiment with Typeform proved it and learn how you can capture more feedback effortlessly.<\/p>\n","protected":false},"author":45,"featured_media":9127,"comment_status":"open","ping_status":"open","sticky":true,"template":"","format":"standard","meta":{"content-type":"","_lmt_disableupdate":"","_lmt_disable":"","footnotes":""},"categories":[8,7],"tags":[78,66,62,37,176,15,10,25,77],"class_list":["post-9096","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-customer-feedback","category-product-management","tag-customer-feedback","tag-feedback","tag-feedback-management","tag-prioritization","tag-product-development","tag-product-management","tag-product-manager","tag-saas","tag-user-feedback"],"aioseo_notices":[],"modified_by":"Eric Hoppe","_links":{"self":[{"href":"https:\/\/canny.io\/blog\/wp-json\/wp\/v2\/posts\/9096","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/canny.io\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/canny.io\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/canny.io\/blog\/wp-json\/wp\/v2\/users\/45"}],"replies":[{"embeddable":true,"href":"https:\/\/canny.io\/blog\/wp-json\/wp\/v2\/comments?post=9096"}],"version-history":[{"count":14,"href":"https:\/\/canny.io\/blog\/wp-json\/wp\/v2\/posts\/9096\/revisions"}],"predecessor-version":[{"id":10964,"href":"https:\/\/canny.io\/blog\/wp-json\/wp\/v2\/posts\/9096\/revisions\/10964"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/canny.io\/blog\/wp-json\/wp\/v2\/media\/9127"}],"wp:attachment":[{"href":"https:\/\/canny.io\/blog\/wp-json\/wp\/v2\/media?parent=9096"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/canny.io\/blog\/wp-json\/wp\/v2\/categories?post=9096"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/canny.io\/blog\/wp-json\/wp\/v2\/tags?post=9096"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}