Ready to get started with Harvestr?
Follow these five steps and you'll have your feedback loop running in no time.
Step 1. Connect your tools
Where your feedback comes from
To centralize all your product feedback, we've built native integrations with the tools your teams already use:
Slack, Microsoft Teams: collect feedback from teams working in chat
Salesforce, HubSpot: let sales create and track feature requests without leaving their CRM
Chrome extension, feedback widget, email: capture feedback from anywhere else
Zapier: connect thousands of other apps
Where your work happens
Figma: attach design work to your Discoveries
Your AI tools
Connect Claude, Cursor, or any MCP-compatible assistant to your workspace with our MCP server.
Step 2. Organize your product space
Customer feedback becomes a mess fast if it isn't organized. Two concepts keep it in order: Discoveries and Components.
What is a Discovery?
A Discovery is where you aggregate all the feedback about the same topic.
Discoveries can be customer problems, improvement opportunities, or feature ideas. Each one has:
a title (e.g. "Slack integration")
a description (e.g. "Our customers need a Slack integration to…")
tags (e.g. "Feature request", "Bug")
a state (e.g. "Not for now" or "In development")
custom fields (goal, revenue, effort, and so on)
What is a Component?
Components are the categories your Discoveries live in. They can be product areas (Onboarding, Payment, Settings), Jobs to be Done, team scopes, or whatever matches how your product organization thinks.
At Harvestr, we organize our Discoveries around our main product areas and Jobs to be Done: Feedback, Integrations, Discovery, and so on. Every piece of feedback lands in the matching Discovery, inside the right Component.
You can also nest sub-Components inside Components.
Best practice: start simple with one or two levels of Components and refine later if you need to.
Step 3. Import existing data
Feedback
Your integrations start bringing feedback in as soon as they're connected. For historical feedback, go to Settings > Import data and use our CSV and Excel import.
Customer data
Set up customer data synchronization from HubSpot, Salesforce, or Intercom (Salesforce is available on the Elite plan). This is what later lets you prioritize on revenue rather than volume.
Discoveries
If you're migrating from another tool, import your existing Discoveries by CSV too.
You can import feedback, customers and Discoveries together, with the links between them preserved, using our multi-object import (Everything at once in Settings > Import data).
Step 4. Process incoming feedback
Now you can start categorizing feedback from every source you've connected.
Let AI do the first pass
HAI automatically categorizes new feedback by suggesting relevant Discoveries. You review the suggestions and accept, reject, or reassign them, individually or in bulk. This is what makes it possible to stay on top of feedback at scale instead of reading every item by hand.
Categorize feedback yourself
You can always do it manually. When you spot something valuable in your inbox, you can:
Select specific verbatims and link them to a Discovery
Or click "Link to" to link the whole item
Either way, a popup opens showing all your Components and Discoveries. Search and click a Discovery to link the feedback to it.
If the topic is new, create a Discovery with the + button next to the relevant Component.
Step 5. Prioritize what to build next
Prioritization is the hardest part of the job. Harvestr takes you from raw feedback to informed decisions by combining qualitative insight with criteria that match your strategy.
Define your prioritization criteria
Decide which factors matter to your team, then create the matching Discovery fields. Common criteria:
Feedback volume: how much feedback did we get on this?
Customer segment: is the request coming from target accounts?
Opportunity size: how much revenue is tied to this problem?
Effort or complexity: how hard is it to deliver?
Strategic fit: does it align with our goals?
Then enrich your Discoveries with Rollup fields, which compute values from the customers and feedback linked to each one: cumulated ARR, number of Enterprise accounts, count of feedback flagged as churn risk.
Filter and sort to surface top priorities
Once Discoveries are categorized and enriched, use filtering and sorting to find what matters:
Filter by product area, customer segment, or feedback source
Sort by linked feedback count, aggregated revenue, or any custom field
Combine filters to focus on what matters now, for example "Enterprise customers" + "Billing"
Best practice: save your most-used filter and sort combinations as Views, so you can answer questions like "what are the most requested features this month?" in one click.
You're all set
Your feedback loop is running. From here, the natural next steps are building your roadmap and closing the loop with the people who gave you feedback in the first place.