How we built a real-time duplicate issue detector for Jira (and why it was harder than expected)

Alaeddine MELLITI — Tue, 09 Jun 2026 12:55:18 +0000

Every team I've talked to has the same silent problem in their Jira backlog.

A developer opens a ticket to investigate a bug. Spends 2 hours on it. Closes it.

Three weeks later, someone else opens the exact same ticket.

Nobody searched. Nobody knew it already existed. The backlog keeps growing, and the same problems keep getting solved twice.

Why this happens

Jira has no built-in mechanism to surface similar issues at the moment a new ticket is created. By the time someone thinks to search, they've already written the description, assigned it, and moved on.

The problem isn't people being careless — it's a missing UX layer.

What we built

I'm part of a small team at Lotus INFO and we decided to fix this with a Jira Cloud app: Issue Duplicate Finder for Jira.

The idea is simple: the moment a user starts creating a new issue, the app scans the backlog and surfaces similar issues in real time — with matching words highlighted so the comparison is instant.

No configuration. No onboarding. Just works.

The technical challenges

Building on the Atlassian Forge platform has its own constraints. A few things that took longer than expected:

Similarity matching in real time — we needed it to feel instant, not like a search query. Tuning the matching algorithm to balance precision and recall without slowing down the issue creation dialog took several iterations.
Cross-project search — surfacing duplicates across projects (not just the current one) required careful handling of permissions and project scope.
Handling resolved issues — we wanted to show resolved duplicates too (they're often the most useful), but display them differently. We ended up using strikethrough so they're visible but clearly closed.

Who it's for

We built it primarily for three types of teams:

ITSM / Support teams — stop re-investigating incidents that were already resolved. Surface the existing resolution instantly.

QA teams — works with Xray and Zephyr. Avoid creating duplicate test cases across sprints or projects.

Dev teams — catch duplicate user stories and bug reports before they create technical debt.

Pricing — free for small teams

If your Jira team has fewer than 10 users, the app is completely free. No trial period, just install and enjoy it.

For larger teams it's a paid plan via Atlassian, but anyone can start for free and see if it solves the problem before committing to anything.

Try it

The app is live on the Atlassian Marketplace:

👉 Issue Duplicate Finder for Jira

Here's a short demo if you want to see it in action before installing:

We're actively looking for feedback from Jira teams — especially around edge cases, performance on large backlogs, and use cases we haven't thought of yet.

What's your current workaround for duplicate issues in Jira?