# How to sell Jira and issue tracker history for AI training

URL: https://sellmydatatoai.net/guides/sell-jira-and-issue-history-for-ai-training
Reviewed: 2026-11-08

You can license Jira and other issue tracker history to AI companies, usually alongside the code it describes. Labelbox Prism and SellMyData name Jira data, and Mercor's published listing puts software companies at $50,000 to $500,000. Tickets linked to commits, reviews and outcomes are what buyers pay for. Strip customer data, secrets and client project tickets before you export anything.

## Key takeaways

- Labelbox Prism and SellMyData name Jira, and DataVendor and SimpleClosure AssetHub buy the code and tests that issue history explains.
- Mercor's listing puts software companies at $50,000 to $500,000 for 15 to 500 staff, while one closed startup received about $25,000 for its code.
- Issues linked to commits, pull requests and passing tests are worth far more than a backlog of open tickets.
- Jira Cloud's backup manager exports XML plus attachments, and CSV export stops at 10,000 work items per batch.
- Customer data pasted into bug reports and secrets pasted into comments must come out before any sample is drawn.

Issue history explains why software changed. A ticket that describes a bug, links to the commit that fixed it and records the test that now passes is a complete task with a verified answer, which is the shape coding agents and evaluation sets need. OpenAI's SWE-bench Verified, published in August 2024, used 93 software developers to screen tasks for well-specified issues and appropriate tests, and kept 500 verified samples.

## Who buys Jira and issue tracker data?

Jira and issue tracker data is bought by training-data companies, code marketplaces and brokers, nearly always together with the repositories the tickets describe.

| Buyer | What it lists | Published terms |
| --- | --- | --- |
| Labelbox Prism | Operational data from Salesforce, Jira and SAP, plus code generation and edit traces | From around $50,000 to more than $1 million, exclusive by default |
| SellMyData | Code from GitHub, GitLab and Jira | No minimum headcount, commission when a sale closes |
| DataVendor | Codebases, tasksets with their RL environments and raw data | Vendor sets the price, 20% of earnings on withdrawal |
| SimpleClosure AssetHub | Repositories, documentation and tests | No upfront cost, a percentage on a completed sale |
| micro1 | Engineering documentation among operational records | $100,000 to $500,000 for a qualified partnership |

Our profiles of [DataVendor](https://sellmydatatoai.net/data-buyers/datavendor) and [SimpleClosure AssetHub](https://sellmydatatoai.net/data-buyers/simpleclosure-assethub) explain how each lists code and its history.

## How much is Jira history worth?

Jira history has no published price per issue or per project. It is priced as part of a software company's engineering record.

- Mercor's listing puts software companies at $50,000 to $500,000 for 15 to 500 staff, as reported by ddiy.co in October 2026
- A shut-down crypto startup received about $25,000 for its codebase, with about $15,000 more pending for workspace data, according to MIT Technology Review in October 2026
- SimpleClosure's typical payments run $10,000 to $100,000 per company across code and data, according to Fortune in September 2026

DataVendor cautions that a listed price does not establish what an asset will earn. Our page for [software companies](https://sellmydatatoai.net/sell-your-data/software-companies) covers the rest of the engineering archive buyers ask for.

## What makes issue history more valuable?

Issue history is more valuable when each ticket can be traced to the code change that resolved it and to a check that proved the fix worked.

| Factor | Worth more | Worth less |
| --- | --- | --- |
| Links to code | Issue keys in commit messages and pull requests | Tickets with no trace to the code |
| Resolution | Resolution, fix version and linked test on each ticket | Tickets closed without a reason |
| Change history | Full changelog of status changes and comments | Current state only |
| Incidents | Incident tickets linked to postmortems and runbooks | Alerts with no follow-up |
| Specification | Acceptance criteria and design discussion | One-line titles |

Linked tickets and code can be turned into [agent trajectory data](https://sellmydatatoai.net/buy-ai-training-data/agent-trajectory-data) or evaluation tasks, which some buyers price separately from raw history.

## How do you export Jira Cloud data?

You export Jira Cloud data with the backup manager for a full copy, or with CSV export for selected fields, according to Atlassian's support pages.

| Method | What you get | Limits |
| --- | --- | --- |
| Backup manager | A ZIP file with entities.xml and activeobjects.xml, plus attachment, avatar and logo folders | Atlassian recommends up to 25 GB of data, and 48 hours between backups that include attachments |
| CSV export | Work items with the fields you choose | Up to 10,000 work items per export, so split larger sets into batches with JQL |

Atlassian notes that backups leave out automation flows and third-party app data, so export anything held in apps separately. GitHub Issues, GitLab and Linear have their own exports. Whatever the tool, keep issue keys, timestamps and links so the history can be joined to the repository.

## What must you remove from issue history?

You must remove customer data, secrets and live security details from issue history, because engineers paste all three into tickets while fixing problems.

| Item | Where it hides | What to do |
| --- | --- | --- |
| Customer data | Bug reports, pasted logs, screenshots and attachments | Replace with synthetic values or remove |
| Secrets | Comments, config snippets and attachments | Scan, remove and rotate every credential found |
| Open security issues | Vulnerability tickets for live systems | Exclude until fixed |
| Client projects | Projects run for clients under services contracts | Check each contract, usually exclude |
| Staff identities | Assignee, reporter, watchers and @mentions | Rewrite to consistent pseudonyms |

Rotate any credential that ever appeared in a ticket, even after deleting it. Once a copy leaves your control it cannot be recalled.

## Who owns the issue history?

Your company owns the issue history its employees wrote. The US Copyright Office's Circular 30 says a work prepared by an employee within the scope of employment is a work made for hire, owned by the employer.

Contractors are different. A commissioned work counts as made for hire only if it falls in one of nine categories and both sides signed a written agreement saying so, so check that contractors assigned their work to you in writing. If you run a software service, tickets may also contain your customers' data, which your contracts with them govern. Our guide to [selling source code](https://sellmydatatoai.net/guides/sell-source-code-to-ai-companies) covers the code side of the same checks.

## If you hold UK or EU data

If you run a software service for UK or EU customers, you are usually their processor for personal data in logs and tickets, and Article 28(3)(a) of the UK GDPR limits a processor to the controller's documented instructions. Under section 11 of the Copyright, Designs and Patents Act 1988, contractors own what they write unless they assigned it to you.

## How do you sell issue history step by step?

1. **Pick the projects**: Choose long-running product projects whose tickets link to repositories you also own.
2. **Check ownership**: Confirm contractor assignments and exclude projects run for clients.
3. **Export a fixed copy**: Take a full backup or batched CSV exports, plus the matching repository history.
4. **Clean the history**: Remove customer data and open security issues, rotate secrets and pseudonymize staff.
5. **Write the manifest**: Record projects, issue counts, date ranges, link rates to commits and exclusions. See [how to build a data manifest](https://sellmydatatoai.net/guides/how-to-build-a-data-manifest).
6. **Share a sample**: Buyers you approve see a redacted sample under a mutual NDA before you sign directly with one.

For a view of likely buyers and a realistic range for your engineering archive, request a [free data valuation](https://sellmydatatoai.net/data-valuation).

## Frequently asked questions

### Is Jira data worth anything without the source code?

Jira data on its own is worth less, because buyers training coding agents want to see the change that resolved each ticket. Issue history can still help with project management and planning models, and Labelbox Prism names Jira among the systems it takes operational data from. Pairing tickets with the repository usually raises interest and price.

### Can I include tickets from client projects?

Tickets from client projects are usually excluded, because services contracts often give the client ownership of deliverables and impose confidentiality on project information. Read each contract. Where the client agrees in writing, you can include the project after removing the client's name, systems and data. Record every excluded project in the data manifest.

### How do I export more than 10,000 issues from Jira Cloud?

Atlassian's support article says CSV export handles up to 10,000 work items at a time, and recommends splitting larger searches into batches with JQL, for example by project or date range, then merging the files. For a complete copy with comments and attachments, use the backup manager, which produces XML files and attachment folders in a ZIP.

### Do ticket comments and attachments need cleaning?

Ticket comments and attachments need the most cleaning, because that is where engineers paste logs, customer records, screenshots and credentials. Scan comments for secrets and personal data, rotate any credential found, and either redact attachments or leave them out. Test a random sample of tickets after cleaning, as our vetting standard requires.

## Sources

- [Introducing SWE-bench Verified](https://openai.com/index/introducing-swe-bench-verified/), OpenAI, accessed November 2026
- [Prism](https://prism.labelbox.com/), Labelbox, accessed November 2026
- [SellMyData homepage and FAQ](https://www.sellmydata.com/), SellMyData, accessed November 2026
- [AI training data marketplaces](https://datavendor.ai/blog/ai-training-data-marketplaces), DataVendor, accessed November 2026
- [AssetHub](https://simpleclosure.com/product/assethub/), SimpleClosure, accessed November 2026
- [Data partnerships](https://www.micro1.ai/data-partnerships), micro1, accessed November 2026
- [How to sell company data to AI companies (2026 guide)](https://ddiy.co/sell-company-data-to-ai/), ddiy.co, accessed November 2026
- [AI labs are buying data from dead startups to train their models](https://www.technologyreview.com/2026/10/13/1145755/ai-labs-buying-data-from-dead-startups-to-train-models), MIT Technology Review, accessed November 2026
- [micro1 tries to beat Google's bid to buy bankrupt Spirit Airlines' data](https://fortune.com/2026/09/14/micro1-tries-to-beat-googles-bid-to-buy-bankrupt-spirit-airlines-data-for-12-5-million-highlighting-value-of-data-to-ai-companies/), Fortune, accessed November 2026
- [Export data from Jira Cloud](https://support.atlassian.com/jira-cloud-administration/docs/export-issues/), Atlassian Support, accessed November 2026
- [Export over 10,000 work items in Jira Cloud](https://support.atlassian.com/jira/kb/export-over-10-000-work-items-in-jira-cloud), Atlassian Support, accessed November 2026
- [Circular 30: works made for hire](https://www.copyright.gov/circs/circ30.pdf), US Copyright Office, accessed November 2026
- [UK GDPR, Article 28: processor](https://www.legislation.gov.uk/eur/2016/679/article/28), legislation.gov.uk, accessed November 2026
- [Copyright, Designs and Patents Act 1988, section 11](https://www.legislation.gov.uk/ukpga/1988/48/section/11), legislation.gov.uk, accessed November 2026

---

Book a free consultation: https://sellmydatatoai.net/book-a-consultation