Can I import my Canny data into FeedbackFlow?

Not with a one-click importer, because we do not have one. Canny will export your posts to CSV, and you can script those rows into the FeedbackFlow feedback API using your project API key, which accepts a message, a category, a page URL, and a requester identity. What it cannot accept is a vote count, a comment thread, a post status, or an original creation date, so an imported archive arrives undated and unvoted. We generally recommend against importing for that reason and suggest hand-filing your top requests instead.

Will I lose my Canny vote counts?

Yes. FeedbackFlow has no voting mechanism at all, so there is no field for a vote score to land in. Ranking is done by AI on sentiment, urgency, and how many separate people raised the same thing, rather than by customers voting. This is the single biggest thing to weigh before switching: if years of accumulated vote data is how you decide what to build, keep the Canny board alive read-only rather than trying to migrate it.

How long does switching from Canny take?

Setup is an afternoon: install the widget with two script tags, connect Slack, Intercom, or Zendesk, and connect Jira or Linear. The honest total is longer, because you should run both tools in parallel for two to four weeks while real feedback accumulates in FeedbackFlow and you decide whether the loss of the public board matters to your customers. There is no reason to cut over in a single day.

Can I keep my Canny board and still use FeedbackFlow?

Yes, and for a lot of teams that is the right answer. They solve different halves of the problem: Canny gives your customers a public place to post and vote, while FeedbackFlow reads feedback from your widget, Slack, Intercom, and Zendesk and files ranked tickets to Jira or Linear. Canny’s free plan covers 25 tracked users (as of July 2026), so keeping a low-traffic board live while the triage work moves across costs very little.

Switching from Canny: What Actually Moves, and What Doesn't

By The FeedbackFlow Team · Last updated August 4, 2026

How we compared: We reviewed each tool's official pricing and documentation, plus public reviews, in July 2026. Figures are stamped “as of” their check date and should be re-verified before purchase. FeedbackFlow is our own product, so we say so wherever it appears and concede where competitors are stronger.

The short version

Straight answer first: FeedbackFlow has no Canny importer. There is no one-click migration, and our feedback API cannot backdate an item, so anything you replay into it arrives stamped today. What you are actually moving is the inflow, not the archive. Canny's CSV export will hand you your posts, and a script can push them in, but vote counts, comment threads, and the original timeline do not survive the trip. Our honest recommendation is to skip the archive import entirely and let a few weeks of real inflow rebuild the signal. Here is exactly what moves, what does not, and how to run both tools while you decide.

First, make sure you are switching for the right reason

Most people leave Canny for one of three reasons, and only one of them points at us. Being blunt about this saves you a migration you would regret.

If you are leaving over price, you want Featurebase, UserJot, or Sleekplan. They are boards, like Canny, with billing that does not scale with your audience. FeedbackFlow would also cut your bill, but it would take your public board away in the process, which is a much bigger change than you were shopping for.

If you are leaving over thin reporting, you want Productboard or Enterpret. Analytics depth is their whole pitch.

If you are leaving because the board fills up faster than anyone can sort it, that is the one we are built for. FeedbackFlow has no board. It captures feedback from an in-app widget plus Slack, Intercom, and Zendesk, dedupes across all of them, ranks what it finds, and drafts ticket-ready Jira or Linear issues. The triage step stops being somebody's Friday afternoon.

If you are in the first two groups, the rest of this page is not for you, and that is fine.

What you can get out of Canny

Canny is good about letting you take your data. You have three routes, per Canny's export documentation (as of August 2026):

A full board export to CSV from the Boards settings tab.
A filtered export from the Feedback page, so you can take only the posts matching your current filters rather than everything.
The Canny API, if you want programmatic access or something more selective than the CSV gives you.

The CSV columns are: Post Title, Post Detail, Score (Total Votes), Comment Count, date created, Image URLs, Post Status, Post URL, Tags, Post Author, Category, Post Owner, and Associated MRR. Voter lists export separately, per post. No plan restriction on exporting is documented, so this works whether you are on Free or Enterprise.

So getting your data out is the easy half.

What survives the move, and what does not

This is the part most migration guides skip. There is no importer, so the only way to get an archive into FeedbackFlow is to script it against the same feedback API our widget uses, authenticated with your project API key. That endpoint accepts a message, a category, a page URL, and a requester identity. It does not accept a vote count, a comment thread, a status, or a creation date.

What Canny exports	Comes across?	What actually happens
Post title and detail	Yes, if you script it	Combined into one message field, capped at 2,000 characters
Post author	Yes	Arrives as the requester identity and links to a customer record
Category	Yes	Maps directly to the category field
Post URL	Yes	Stored as the source page URL on the item
Vote score	No	FeedbackFlow has no voting. Ranking is AI-scored, not customer-scored
Comment threads	No	The CSV carries a comment count, not the comment bodies
Original post dates	No	The API stamps items on arrival, so your whole archive lands dated today
Post status	No	Status lives in Jira or Linear once an item is filed
Tags and Associated MRR	No	No equivalent fields
Your public board and changelog	No	FeedbackFlow does not have either, by design

Why we would tell you to skip the archive import

Look at that table and notice which row matters most. The single most valuable thing on a mature Canny board is usually not the text of the posts. It is the vote counts: years of customers telling you which requests they actually care about. That is the one column with no destination.

So a scripted import gives you a few thousand undated, unvoted, unranked text blobs, all arriving at once, all triggering AI triage simultaneously, and all timestamped the day you ran the script. The AI will dutifully group and rank them, but it is ranking them on language and sentiment, not on the customer demand signal you spent years accumulating. You get the noise of the archive without the part that made it worth keeping.

Our advice, which costs us a "seamless migration" bullet point on a sales page: do not import it. Do this instead.

Leave the Canny board up, read-only. Your history stays queryable exactly where it already is, with the votes intact. You have lost nothing.
Point your top requests at your tracker by hand. In practice there are ten to thirty things on a mature board that genuinely matter. File those into Jira or Linear yourself, once. It is an afternoon, and you get to apply judgment the AI cannot.
Let the inflow rebuild the signal. Once the widget and your support channels are connected, real feedback starts arriving immediately, correctly dated and correctly attributed. Within a few weeks you have a live ranked list built from current demand rather than a five-year backlog.

If you still want the archive in for searchability, script it against the feedback API with your project API key and accept the caveats above. We would rather tell you the trade-off than sell you a migration.

Skip the board. Let AI triage feedback into Jira or Linear.

Try FeedbackFlow free

What switching actually looks like

Realistically, an afternoon of setup and a few weeks of overlap.

Create a project and install the widget. Two script tags before the closing body tag, plus your API key. See the widget installation guide. No build step.
Connect the channels where feedback already lands. Slack, Intercom, and Zendesk each take a few minutes: see the setup guides for Slack, Intercom, and Zendesk. This is the step that does the work, because it captures the feedback that never made it to your Canny board in the first place.
Connect Jira or Linear. An API token and a project key for Jira, or an API key and team ID for Linear.
Hand-file your top Canny requests. As above. Ten to thirty items, once.
Run both for two to four weeks. Keep Canny live while the new inflow accumulates. Nothing forces you to cut over on a single day, and you will want the comparison.

The part your customers will notice

Be honest with yourself about this one: switching from Canny to FeedbackFlow removes something your customers could see. No public board, no voting, no customer-facing roadmap, no changelog. If people currently visit your Canny board, follow requests, and get notified when something ships, that stops.

For a lot of teams that is a relief, because the public board was generating expectation-management work nobody signed up for. For others it is a real loss of a trust-building surface, and no amount of AI triage compensates.

Two sane ways to handle it:

Keep Canny for the board, use FeedbackFlow for triage. These are not mutually exclusive. Canny's free plan covers 25 tracked users, so a low-traffic public board can stay live for very little while the triage work moves.
Tell people where feedback goes now. If you are closing the board, say so on it, and point people at the in-app widget. Silently deleting a board your customers used is the version of this that costs you goodwill.

When you should not switch

Stay on Canny if a public voting board is load-bearing for how you work with customers, if your audience is small enough that tracked-user pricing does not bite (the free plan covers 25 tracked users, and Pro starts at $79 a month billed yearly for up to 100, as of July 2026), or if your customers already know and use the board. Canny's Autopilot AI is included free on every plan and does a genuinely useful job of surfacing feedback from connected support tools, so if triage volume is only a mild irritation rather than a real bottleneck, switching tools is churn for its own sake.

Leave for a specific, nameable reason. "The board fills up faster than we can sort it" is one. "The grass looks greener" is not.