# Appen enterprise data review: is Appen legit, and how its company-data program pays

> Appen enterprise data review: what Appen buys from companies, payment on acceptance after sample approval, its ASX listing and history, and what to ask first.

Canonical: https://origindatapartners.com/guides/appen-enterprise-data-review
By: Origin Data Partners
Status: Published
Published: 2026-11-06
Updated: 2026-11-06
Sources checked: 2026-11-06

## Quick answer

Appen, an AI training-data company founded in Sydney in 1996 and listed on the Australian Securities Exchange (ASX: APX) since 2015, runs an enterprise data program that licenses companies' operational records to AI labs. It publishes no price range: each deal is quoted individually, terms are agreed in writing before extraction, you approve a sample pack first, and payment is made on acceptance, with no costs or deductions to you. The main catch is transparency: there is no price range, no stated payment timing and nothing on exclusivity, so you learn the real terms only in the offer. Appen is a public company with long lab relationships, though it lost Google, nearly a third of its revenue, in 2024.

## Key points

- Appen buys three kinds of records: operational documentation, decision and escalation records, and systems of record such as ERP and contracts.
- No published price. Each deal is quoted individually, and payment is made on acceptance, not on submission.
- You approve a representative sample pack before anything is licensed, and Appen covers extraction and anonymization costs.
- The counterparty is a listed company with 30 years in AI data; the program itself has no public seller reports yet.

## Appen enterprise data review: the short version

[Appen's enterprise data program](https://www.appen.com/enterprise-data) invites companies to "license the workflows and operational know-how your company already runs on to the AI labs building the next generation of models." Appen handles extraction, anonymization and the conversations with the labs, and you keep ownership.

It suits a company that values a long-established, publicly listed counterparty and control over what leaves, and that is willing to go through scoping before seeing a number. It suits a company less well if it wants a published price or payment date before investing time.

Because Appen publishes no range, it helps to [see what your data is worth](https://origindatapartners.com/check-fit) before the first call, so you can judge the quote when it arrives.


## What does Appen's enterprise data program buy?

Appen says it scopes every deal against three categories of operational data, set out in the table below.

Appen says the labs want "enterprise task competence", "failure and recovery paths" such as reopened tickets and rejected approvals, and the reasoning behind decisions. It adds that "connected systems are worth considerably more than isolated exports" and that winding-down companies follow a faster path and can often license their complete operational history.

| Category | Typical sources (as Appen lists them) | Appen's note |
| --- | --- | --- |
| Operational documentation | Docs and wikis, cloud storage, intranet, training material, QA procedures | Usually the fastest to scope |
| Decision and escalation records | Support desk, ITSM, CRM and sales, project tracking, messaging and meetings | "This is the category frontier labs ask for most often." |
| Systems of record | ERP and finance, data warehouse and BI, contracts and e-signature, HR and payroll, procurement | Connected systems are worth considerably more than isolated exports |

Source: [Appen: enterprise data](https://www.appen.com/enterprise-data)

## How does Appen pay for company data?

"Each partnership is quoted individually, terms are agreed in writing before anything is extracted, and payment is made on acceptance," according to [Appen](https://www.appen.com/enterprise-data). It says price depends on uniqueness, structural depth (how connected the systems are) and volume and longevity, and that "a decade of records is worth more than a single quarter".

Paying on acceptance rather than submission means Appen pays for data it has checked, not for whatever is delivered. That protects Appen; for you it means the acceptance criteria matter as much as the price. Ask for them in writing.

| Term | What Appen publishes | What to confirm |
| --- | --- | --- |
| Price | Not published; quoted individually | The quote, and which records drove it |
| Payment trigger | On acceptance rather than submission | Written acceptance criteria and how long acceptance can take |
| Payment timing | Not stated | The number of days from acceptance to payment |
| Sample | You review a representative sample pack before anything is licensed | That the sample matches what is finally delivered |
| Costs | No upfront costs, setup fees or deductions | That this holds if a dataset is not accepted |
| Ownership | A defined license for a defined use; no transfer of ownership | The permitted uses and the term |
| Retention and deletion | Agreed in writing before extraction | What happens to copies when the license ends |
| Exclusivity | Not stated | Whether the license is exclusive, and for how long |
| Access | Read-only, scoped credentials, isolated pipelines | Who at Appen can access the systems, and for how long |
| First contact | Reply within two business days; mutual NDA before further discussion | The NDA terms |

Source: [Appen: enterprise data](https://www.appen.com/enterprise-data)

## Is Appen legit?

As a company, yes. Appen was founded in Sydney in 1996 ([Appen](https://www.appen.com/about-us)) and listed on the Australian Securities Exchange as APX on January 7, 2015 ([Appen](https://www.appen.com/press-release/appen-apx-on-asx)), so it files regular reports with the ASX. In the third quarter of 2026 it reported revenue up 34% year on year to US$73.9 million and said it now has seven of the eight biggest US model developers as clients, [Slator reported](https://slator.com/appen-q3-2026-growth-accelerates/) from its [quarterly filing](https://announcements.asx.com.au/asxpdf/20261029/pdf/075yd0l9znkw74.pdf).

It has also had hard years. In January 2024 Alphabet ended its contract with Appen; Alphabet work had brought in $82.8 million of Appen's $273 million in 2023 sales ([CNBC](https://www.cnbc.com/2024/01/23/alphabet-ends-contract-with-appen-which-trained-bard-google-search.html)). After the October 2026 results, its shares fell around 20% when it said full-year revenue would come in at the lower end of its forecast ([Slator](https://slator.com/appen-q3-2026-growth-accelerates/)).

What is not yet visible is the program's track record. Appen's page gives no launch date, price range or number of companies paid, and we found no public report from a company that licensed data through it. Our view: a credible, accountable counterparty with careful process terms, but the least transparent on price of the three programs we reviewed.

Source: [Appen: about us](https://www.appen.com/about-us)

## Looking for Appen jobs instead?

Many people who search "is Appen legit" are asking about paid tasks as a contributor, not about licensing company data. That is a different part of Appen. Its [scam awareness page](https://www.appen.com/legal-policies/scam-awareness) says "we will never ask you to send money to apply or start a job with us", and that official emails come only from @appen.com addresses.

This page covers the enterprise data program for companies that own records.

Source: [Appen: scam awareness](https://www.appen.com/legal-policies/scam-awareness)

## How Appen compares with other company-data programs

Appen is the only listed company of the three programs we reviewed, and the only one that publishes no price range. Its sample sign-off before licensing is the most explicit seller control of the three. Figures are as each program publishes them, read November 6, 2026.

Origin also covers the programs run by Mercor, micro1, Handshake AI and Turing. See [company data buyers compared](https://origindatapartners.com/guides/company-data-buyers-compared) for their published terms side by side, [Mercor competitors](https://origindatapartners.com/guides/mercor-competitors) for the wider field, and [exclusive vs non-exclusive data licenses](https://origindatapartners.com/guides/exclusive-vs-non-exclusive-data-license) for what each license type costs you in future deals.

| Program | Published pay | How you are paid | Exclusivity (as published) | Timing (as published) | Who is behind it |
| --- | --- | --- | --- | --- | --- |
| [Labelbox Prism](https://origindatapartners.com/guides/labelbox-prism-review) | About $50K to $1M+ | Upfront fee at signing plus a revenue share on downstream use | Exclusive by default | About 21 days from first call to first wire (Prism's claim) | Labelbox, a privately held AI data company founded in 2018 |
| [Polyshares](https://origindatapartners.com/guides/polyshares-review) | $100K to $2M and above, typically | Paid per license, net 30 to 60 days after approval, anonymization and delivery | Exclusive by default unless agreed otherwise; its four published deals are perpetual exclusive | Offer in one to two weeks | Polyshares, Inc., run by two managing partners; it does not publish its founding date or investors |
| [Appen enterprise data](https://origindatapartners.com/guides/appen-enterprise-data-review) | Not published; each deal quoted individually | Paid on acceptance, after you approve a sample pack | Not stated | Reply within two business days; payment timing not stated | Appen Limited, listed on the ASX (APX), founded in 1996 |


## Questions to ask Appen before you sign

Ask for the answers in the written terms Appen agrees before extraction. Our [data licensing agreement guide](https://origindatapartners.com/guides/data-licensing-agreement) explains each clause.

- What is the quote, and what are the written acceptance criteria for the dataset?
- How many days after acceptance will we be paid, and is any part paid if only some data is accepted?
- Is the license exclusive? If so, for how long and for which uses?
- Which AI labs or kinds of buyer will receive the data, and can we exclude competitors?
- How does your anonymization handle names inside free text, attachments and call recordings?
- What are the retention and deletion terms, and what happens to models already trained if the license ends?
- How many companies have licensed data through this program so far?


## Next step

Get a sense of value before Appen quotes. Compare its process with [Labelbox Prism](https://origindatapartners.com/guides/labelbox-prism-review), which publishes a range and a revenue share, and [Polyshares](https://origindatapartners.com/guides/polyshares-review), which publishes example deals. Read [is selling company data to AI legit?](https://origindatapartners.com/guides/is-selling-company-data-to-ai-legit) for the ownership and privacy checks every buyer runs.

Origin Data Partners works with companies that own data and helps get it bought. We are independent of the services on this page, and a buyer may pay us when a deal closes ([how Origin is paid](https://origindatapartners.com/referral-disclosure)). [See what your data is worth](https://origindatapartners.com/check-fit).


## Questions and answers

### Is Appen legit?

Yes, as a company. Appen was founded in Sydney in 1996, has been listed on the ASX as APX since January 2015, and reported US$73.9 million of revenue in the third quarter of 2026. Its enterprise data program is newer, publishes no prices, and has no public seller reports yet.

### Does Appen buy company data?

Yes. Appen's enterprise data program licenses companies' operational records to AI labs: documentation and procedures, decision and escalation records from support and CRM systems, and systems of record such as ERP, contracts and finance. You keep ownership; Appen licenses a defined use.

### How much does Appen pay for company data?

Appen publishes no price range. Each partnership is quoted individually, based on how rare the knowledge is, how connected the systems are, and how much history there is. Appen says there are no costs or deductions from what you are paid.

### When does Appen pay?

On acceptance. Appen says "payment is on acceptance rather than submission", after you have approved a representative sample pack and the dataset has been checked. It does not publish how many days after acceptance payment arrives, so ask for that in writing.

### What is an Appen data partnership?

It is a license of a company's operational data to Appen for a defined use. Appen scopes the systems with you, extracts read-only, anonymizes the data and supplies AI labs. Ownership does not transfer, and retention and deletion terms are agreed in writing before extraction.

### Is Appen listed on the stock exchange?

Yes. Appen Limited trades on the Australian Securities Exchange under the code APX. It listed on January 7, 2015, and reports quarterly to the ASX.

### Is an Appen enterprise data license exclusive?

Appen's enterprise data page does not say. It describes a defined license for a defined use with ownership retained, so ask whether exclusivity applies, for how long and for which uses before you agree terms.

### Can a company that is closing license data to Appen?

Yes. Appen says wind-downs follow a faster path and can often license the complete operational history rather than a scoped subset. It asks you to flag a wind-down so it can prioritise the response, because records can become unrecoverable once systems are shut down.

## Sources

- [Appen: enterprise data](https://www.appen.com/enterprise-data)
- [Appen: about us](https://www.appen.com/about-us)
- [Appen: Appen listed on the Australian Stock Exchange (Jan 2015)](https://www.appen.com/press-release/appen-apx-on-asx)
- [Appen: Q3 FY26 quarterly activity report, ASX announcement (Oct 2026)](https://announcements.asx.com.au/asxpdf/20261029/pdf/075yd0l9znkw74.pdf)
- [Slator: Appen Q3 revenue growth accelerates, shares still fall (Oct 2026)](https://slator.com/appen-q3-2026-growth-accelerates/)
- [CNBC: Alphabet ends contract with Appen (Jan 2024)](https://www.cnbc.com/2024/01/23/alphabet-ends-contract-with-appen-which-trained-bard-google-search.html)
- [Appen: scam awareness](https://www.appen.com/legal-policies/scam-awareness)

## Related reading

- [micro1 vs Mercor vs Handshake vs Turing: company data buyers compared (2026)](https://origindatapartners.com/guides/company-data-buyers-compared)
- [Mercor Competitors: Who Else Pays Companies for Data (2026)](https://origindatapartners.com/guides/mercor-competitors)
- [Data licensing agreement: what to include](https://origindatapartners.com/guides/data-licensing-agreement)
- [Is selling company data to AI companies legit? How to tell a real buyer](https://origindatapartners.com/guides/is-selling-company-data-to-ai-legit)
- [Exclusive vs non-exclusive data licenses: what sellers should know](https://origindatapartners.com/guides/exclusive-vs-non-exclusive-data-license)
- [Defunct company data: who buys it and on what terms](https://origindatapartners.com/guides/sell-defunct-company-data)