Product discovery on top of Linear

Linear is a powerhouse for tracking issues. It's modern, it's fast, it's opinionated, it's easy to get started with, the attention to detail is amazing, it's got all the integrations you need, the API is well documented… I could go on and on. It’s a great product.

But when it comes to managing product feedback, it's feels like trying to use a hammer for a screw. It's not that it can't be done, but it's certainly not efficient. The disconnect between customer feedback and product delivery is a gap many of us face. You might think, "But wait, Linear has Slack and Intercom integrations. Can't I just zap feedback into an issue?" It seems straightforward – until it isn't.

Anyone who's tried knows the pitfalls:‍

• ‍Multiple feedback items sent together in one big chunk of feedback.‍

• Overlapping feedback from different users.

• Feedback that's more of a problem statement than a clear feature request.

• Feedback shared by an internal team member (sales/support) on behalf of a customer. How do you keep track of both the reporter and the customer (which are usually both different from the issue’s assignee, who will be an engineer).

• The need to maintain different wordings: raw feedback, release notes, and specs.

• What if the feedback is relevant, but not something to be prioritized any time soon? Do you really want a delivery ticket for that now? And what if the feedback is actually not a bug report but a usability issue? Do you want an issue for that?

• How do you create a customer voice report that shows you what customers have been asking over time?

• How do you capture outbound feedback like call recordings and user research notes? How do you bring these into a single source of truth?

• Finally, how do you close the feedback loop at each release? What do you share? The title of the issue? That title was written for internal implementation, not for external understanding of the value that was shipped.

I'm not even exaggerating. Let's take a real-life example your team surely experienced before: A customer success manager hops off a call with a user and sends a Slack message in a #product-feedback channel with a bullet list of 2 bug reports, 3 feature requests, 1 kudo, and 2 broad problems that need to be fixed.

How do you act upon that feedback?

It. Doesn't. Work.

It’s just not the right data model.

Components of a feedback system that works

So here's the solution. You need a feedback system. With a powerful data model that solves all the issues above. You need the below concepts:

• Feedback: A rich editor to ingest and document all types of feedback, ranging from user interviews to WhatsApp screenshots. The editor must also come with AI features to help with translations or generating summaries.

• Insight: A way to breakdown feedback into actionable & atomic customer quotes.

• Problem: To aggregate related insights and then be broken down into features (no need to send to Linear before it's a scoped feature).

• Feature: PRDs that can be synced with Linear.

• Bug: Your backlog of bugs, before it hits Linear issues.

• Kudo: Sharing kudos matters for team's morale, it's not just about the negative feedback.

• Customer: People & companies, linked to feedback and their insights.

• Feedback assignee: Customer-facing person in charge of closing feedback loops.

• Release notes: Show don't tell. Release notes over roadmaps.

With Cycle, each piece of feedback is meticulously categorized, tracked, and acted upon. Bugs and features find their way to Linear, while broader problems are scoped and prioritized without cluttering your engineers' work environment. Kudos spread positivity, and every piece of feedback is a step towards a more beloved product.

Cycle transforms the feedback loop into a strategic asset, making product discovery a seamless extension of issue tracking. It's not just about fixing what's broken; it's about building what's next with a clear vision driven by customer insights. For teams looking to bridge the gap between feedback and delivery, the Cycle + Linear combo is what you've been looking for.