Productboard logo Productboard automation · runs daily

Triage Productboard feedback and link it to features

Feedback from calls, emails and the portal piles up on the insights board. Every item still needs someone to read it, pick the quote that matters, link it to a feature and mark it processed.

Cotlet does it in your own Productboard account with list_feedback, get_feedback, search_features, link_feedback_to_feature and mark_feedback_processed, called by your AI agent, an n8n workflow or a script.

Sign up free All Productboard tools

Hosted · nothing to install or maintain

Can Productboard do this on its own?

Productboard has AI auto-linking, which needs the Enterprise plan with the AI add-on, and its help center says Productboard AI is not available for purchase at this time (checked 22 Oct 2026). API v2 has relationship endpoints that link a note to a feature, and its tokens need a workspace admin on the Pro plan or higher. Nothing applies your own rule about which quote counts and how important it is.

How you do it in Productboard today

  1. A maker opens any insights board in the workspace and clicks a feedback item to open its sidebar.
  2. In the feedback body, the maker selects the most relevant text or images and clicks Link to.
  3. The maker searches for the idea and picks a feature from the hierarchy, or adds a new feature at the right spot in the hierarchy menu.
  4. The maker sets the importance, which adds +1 for unknown importance, +1 for nice-to-have, +2 for important or +3 for critical to the feature.
  5. Mark processed in the bottom-right corner of the item takes it out of the queue. For many items, the maker ticks checkboxes and chooses (Un)process in the sidebar.
  6. The queue refills every day, and an item left unprocessed looks the same as one nobody has read.

How to triage Productboard feedback and link it to features

Prompt

Go through unprocessed feedback in Productboard. For each item, find the feature it is about, link the most relevant quote with an importance, and mark it processed. Leave the ones you are unsure about for me.
  1. 01

    Read the queue

    The agent calls list_feedback with processed set to false and follows the cursor to the end, so a long weekend is judged as one batch. Each row already shows how many insights the item has, so items linked by a colleague are skipped.

  2. 02

    Read each item in full

    get_feedback brings the whole text and the company or user. A feedback item without a company or user goes on your list instead of being linked blind, because the customer decides how much a link should weigh.

  3. 03

    Find the feature

    search_features runs with the product word from the quote. One clear match is linked. Two or more near matches, or none, leave the item for you with the candidates named, because the agent never creates a feature.

  4. 04

    Link the quote with an importance

    link_feedback_to_feature sends the exact sentence as the highlight and one of unknown, nice_to_have, important or critical, chosen by the rule in your prompt, for example "critical when the customer names a cancellation". The link can be undone with unlink_insight, so it does not wait for your OK.

  5. 05

    Mark processed last

    mark_feedback_processed runs only after the link went through, so an item that failed to link stays in the queue for the next run.

What you get back

Example data

Unprocessed feedback: 15
Linked and processed 11, e.g. "Export to CSV is too slow" to CSV export, important
Left for you 4:
- "Would love dark mode" (two features match)
- "Call notes, Acme" (no company or user)

Other ways to ask

  • Every Friday, link feedback from enterprise customers as critical when it names a deadline.
  • Show me unprocessed feedback grouped by the feature it would link to, but link nothing.
  • Link only feedback tagged "interview" and leave everything else in the queue.

What Productboard allows, and what to know first

  • Productboard scores each link: +1 for unknown importance, +1 for nice-to-have, +2 for important and +3 for critical.
  • Productboard AI auto-linking runs every three hours and skips feedback under 100 characters, and a feature needs both a title and a description before it can be a target.
  • Productboard suggests the owner of a feedback item is the person who processes it, and a comment that @-mentions a colleague tells them you did.
  • Mark processed is a toggle in Productboard: clicking it again takes the item back to unprocessed.

link_feedback_to_feature and mark_feedback_processed change only your own Productboard data and reach no one else, so they run without asking first.

Sources: Productboard Help: Link feedback to feature ideas via insights · Productboard Help: Link insights automatically with Productboard AI · Productboard developers: API v2 overview

Call link_feedback_to_feature from a script

curl

curl -X POST https://cotlet.dev/api/v1/productboard/features/ft_4d7e10/feedbacks/fb_81c2a9/to-feature/link \
  -H "Authorization: Bearer $COTLET_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"quote":"Exporting 50,000 rows to CSV takes over ten minutes and then times out.","importance":"important"}'

Example response

{
  "feedback_id": "fb_81c2a9",
  "feature": {
    "id": "ft_4d7e10",
    "name": "CSV export"
  },
  "quote": "Exporting 50,000 rows to CSV takes over ten minutes and then times out.",
  "importance": "important",
  "linked": true
}

MCP: add https://cotlet.dev/mcp/v1 with the same key; the call is productboard_write → link_feedback_to_feature. Example data.

Schedule it in n8n, Make, Zapier or cron

Productboard tools this automation uses

Cotlet Productboard tools used to triage Productboard feedback and link it to features
Tool Parameters
list_feedback processed, query, cursor
get_feedback feedback_id
search_features query, status, cursor
link_feedback_to_feature feedback_id, feature_id, quote, importance
mark_feedback_processed feedback_id, processed
All 12 Productboard tools with descriptions

Frequently asked questions

Is it safe to connect Productboard to Cotlet?

Yes. You sign in to Productboard yourself, in your own Cloud Browser in the EU, and we never save your password. Your agents only get Productboard once you enable it, every background request is listed, and the browser profile is stored encrypted with AES-256-GCM, the cipher 1Password uses for its vaults. The AI model never sees a login, posts and messages wait for your OK, and Disable for agents cuts access in one click. Independent scanners grade cotlet.dev A+ on Qualys SSL Labs, securityheaders.com and Mozilla Observatory; live results are on https://cotlet.dev/security.

Give your agent access to Productboard.

Free 7-day trial with 2,000 credits, no card needed. Cotlet is independent and not affiliated with Productboard.

Sign up free