Canny

Turn every customer conversation into prioritized Canny feedback

Customers tell your AI agent what they need. The agent logs it as a Canny feature request, records the vote, and shows them where it sits on the roadmap. Your product team wakes up to structured, voted, categorized feedback instead of scattered support tickets.

Chosen by 800+ global brands across industries

Product feedback, captured mid-conversation

Your AI agent listens to what customers want, logs requests in Canny, tracks votes, and publishes changelogs, keeping your product roadmap informed by real demand.

Canny

Use Cases

Feedback loops that close themselves

Watch how AI captures feature requests, tracks demand signals, and keeps customers informed about what you ship next.

Feature Requests Straight from Support Chats

A customer tells your chatbot, 'I really need calendar sync.' Your AI Agent searches Canny for existing posts about calendar sync, finds one with 47 votes, casts a vote on the customer's behalf with 'Must have' priority, and tells the customer it is in the planned stage. The customer feels heard. Your product team sees demand rising on an already-tracked request, no manual triage required.

Automated Changelog Announcements After Shipping

Your engineering team marks a feature complete. Your AI Agent creates a Canny changelog entry with the release details, links the 12 original feature request posts, sets the entry type to 'new,' and triggers email notifications to all voters. Customers who asked for the feature learn it shipped within minutes. Your product team skips the manual update-and-email routine.

Preventing Duplicate Feature Posts Across Channels

Three customers mention the same idea in separate WhatsApp conversations. Each time, your AI Agent searches existing Canny posts before creating a new one. It finds the match, adds a vote with priority context from each customer, and replies with the current status. Instead of three duplicate posts cluttering your board, you have one post with three additional votes and richer priority data.

Try
Canny

Canny

FAQs

Frequently Asked Questions

How does the AI agent log a feature request in Canny during a conversation?

When a customer describes something they need, the agent calls Canny's Create Post API with the board ID, customer's user ID, a title summarizing the request, and the full details. The post appears on your Canny board immediately, attributed to the right user, and ready for team review and voting.

Can the agent vote on behalf of a customer if the request already exists?

Yes. The agent first searches Canny posts by keyword. If a matching post exists, it calls Create Vote with the customer's voter ID and a priority level (Nice to have, Important, or Must have). The customer gets confirmation, and the post's vote count and priority data update in real time.

What Canny API key permissions does Tars need to function?

Tars uses your Canny secret API key found in your company admin settings. This key grants access to posts, boards, users, votes, comments, changelog entries, and tags. There is no granular scope control, so the key provides full API access. You can regenerate it anytime to revoke access.

Does Tars store any of my Canny feedback data?

No. The agent queries Canny's API live during each conversation. Post details, vote counts, user profiles, and changelog entries are fetched in real time and used only to serve the current interaction. Tars does not replicate or cache your Canny database.

Will customers get notified when the status of their request changes?

You control this per update. The Change Post Status endpoint includes a shouldNotifyVoters parameter. When set to true, all non-admin voters receive an email about the status change. You can also include a comment explaining the update. Set it to false for internal status moves that do not need customer visibility.

How does the agent avoid creating duplicate feature request posts?

Before creating a new post, the agent searches existing Canny posts using the List Posts endpoint with a keyword query. If a relevant match is found, the agent casts a vote on the existing post instead. This deduplication logic runs automatically during conversation, keeping your feedback boards clean.

How is this different from Canny's own feedback portal?

Canny's portal requires customers to visit a separate page, create an account, and submit requests manually. The Tars AI Agent captures feedback inside conversations customers are already having, on your website, WhatsApp, or in-app chat. Participation goes up because the barrier drops to zero. The agent also handles voting, deduplication, and profile sync automatically.

Can the agent publish changelog entries and notify voters when a feature ships?

Yes. The agent calls Canny's Create Changelog Entry endpoint with the release title, markdown-formatted details, entry type (new, fixed, or improved), and linked post IDs. Setting the notify parameter to true sends an email to voters when the entry publishes. You can also schedule entries for future publication using the scheduledFor parameter.

How to add Tools to your AI Agent

Supercharge your AI Agent with Tool Integrations

Don't limit your AI Agent to basic conversations. Watch how to configure and add powerful tools making your agent smarter and more functional.

Privacy & Security

We’ll never let you lose sleep over privacy and security concerns

At Tars, we take privacy and security very seriously. We are compliant with GDPR, ISO, SOC 2, and HIPAA.

GDPR
ISO
SOC 2
HIPAA

Still scrolling? We both know you're interested.

Let's chat about AI Agents the old-fashioned way. Get a demo tailored to your requirements.

Schedule a Demo