Helpdesk archives, licensed for AI training. Not event tickets.
Sell Support Tickets
Years of Zendesk, Intercom or Gorgias history show how real problems get solved. AI data buyers publish what they pay for that and who they take it from. Tell us about your archive and we name the buyers it fits.
- 1 Raw export
Priya ShahOrdered a medium, got a large. Order 55-20931.
Tom, agentSorry about that. A prepaid return label is on its way to priya.s@mail.com.
- 2 Redacted
CustomerOrdered a medium, got a large. Order .
AgentSorry about that. A prepaid return label is on its way to .
- 3 Licensed
BuyerYour pick
ScopeSolved tickets, 6 years
ExclusiveAs agreed
Signed byBoth sides
- 4 Paid
30 days
micro1's published payment window, counted from acceptance.
Labelbox Prism says first call to wire can take under 21 days.
Invented example. The names, order number and email in step 1 are made up.
What counts
The conversation is the asset, the customer is not
A buyer wants the problem in the customer's own words, every reply in order, the internal note where an agent worked it out, and the outcome. That sequence is what an AI support agent has to learn, and it is hard to fake. Long histories help: Labelbox Prism lists longitudinal customer support as a data type it wants.
Buyers keep
- Thread in order
- Internal notes
- Tags and status changes
- Timestamps
- Macro text
- Channel
- CSAT rating
You strip first
- Customer and agent names
- Emails and phones
- Order and account numbers
- Card digits, even the last four
- Health details
- Attachments
Buyers
Four buyers name support tickets on their own sites
Each prints a range per company, not per ticket. Read every range as the top of what is advertised: none of them is an offer.
- Labelbox PrismAbout $50K to $1M+
Wants longitudinal support data, preferably from busy consumer brands. Signing fee, then a revenue share.
- micro1$100K to $3M+
Lists support records among its operational data. Three published tiers; pays within 30 days of acceptance.
- Nyne$100K+ typical
Names support tickets outright, next to CRM, dispatch and service history.
- Scale AI$10K to $1M+, illustrative
Lists support records and chat histories. Anonymizes inside your own systems before extraction.
Disclosure: any buyer listed here could owe us a fee if our introduction leads it to a signed license. They are listed A to Z, and a possible fee never changes how we describe a buyer.
Fit
Headcount decides the shortlist before volume does
We match each buyer's published rules to what you tell us: headcount first, then years in business, the language of the tickets, where you are based, and whether your customers are consumers.
Two questions decide whether a fit is worth anything. Can you license this data under your own policy and terms? Could the tickets hold card or health details that have to come out? If either is unclear, that is the first thing we say back.
Example company34 staff, 5 years, US, English
- NyneTwenty or more US employees on W-2, trading three yearsClears
- micro130+ full-time, mostly EnglishClears
- Scale AI40+ preferred, each judged on its ownCase by case
- Labelbox PrismNo floor; favors high-volume consumer supportClears
Invented company, set against each buyer's published entry rule.
Platforms
Your helpdesk shapes what the archive looks like
A ticket-first tool and a chat-first tool export very different records. Formats and plan limits are in each platform's own documentation; these are the six we hear about most.
Zendesk
Ticket-firstPublic replies and internal notes, tags, custom fields.
Freshdesk
Ticket-firstConversations, private notes, statuses and groups.
Intercom
Chat-firstBot turns, customer messages and agent replies in one session.
Help Scout
Mailbox-firstShared email threads with notes and saved replies.
Jira Service Management
Issue-firstRequests are Jira issues, often linked to engineering tickets.
Gorgias
Store-firstTickets sit next to the customer's store orders.
Five rules that come before any price
Whether you can license at all is settled by your own paperwork, not by the buyer.
Your privacy policy and customer terms
They say what you told customers you would do with their messages. If licensing for AI training is not covered, that comes before any price.
Ownership
You must own the tickets or hold the right to license them. Support run for a client under contract is usually the client's data.
Personal data out first
Every buyer here says it de-identifies, but the safest archive is one where names and numbers never leave your systems.
Your own lawyer
Counsel on your side reads every draft, with care over exclusivity. Labelbox Prism licenses are exclusive unless agreed otherwise.
Estimates are never offers
Advertised ranges, and the estimators some buyers run, commit nobody to anything. The real number comes after scoping, and it can be lower or zero.
From first call to payment in five stages
- 1
Scoping
A buyer asks what you hold: platform, years, volume, channels, languages. No data moves yet.
- 2
Sample review
Look at real output before you agree. micro1 lets sellers look over representative samples; ask every buyer for the same.
- 3
De-identification
Identities come out. At Scale AI that work runs on your own systems, and no package leaves without your sign-off.
- 4
Agreement
Scope, rights, exclusivity and acceptance criteria go in writing. Your lawyer reads it, not ours.
- 5
Transfer and payment
Money follows acceptance. micro1 pays in full inside 30 days of it; Labelbox Prism adds a later revenue share to a signing fee.
What is a ticket archive worth?
Not one buyer prices per ticket. These are the tops of what each advertises per company, and reported deals have been far smaller.
- $10K–$1M+Scale AI, an illustration it says is no guarantee
- $100K–$3M+micro1, across its three published tiers
- ~$50K–$1M+Labelbox Prism, small targeted set to large enterprise deal
- $100K+Nyne, a typical partnership
Ranges as the buyers print them; none of them is an offer. Check where your archive sits by headcount, volume and history.
Questions support leads ask
Short answers here. The buyer pages and the blog go further on each one.
Can I sell my support ticket data to AI companies?
Yes, by licensing it rather than selling it outright. micro1, Scale AI, Labelbox Prism and Nyne each name support records or tickets on their own pages and pay under a license agreement, once personal details are out.
How much are support tickets worth to AI companies?
Buyers publish ranges per company, not per ticket: from $10K at Scale AI, which calls its figures illustrative, up to $3M+ in micro1's top tier. Those are ceilings, and reported deals have been far smaller. See where your archive sits; it is not an offer.
Who buys customer support data for AI training?
Training-data suppliers to the AI labs. Four name support data on their own sites: micro1, Scale AI, Labelbox Prism and Nyne. Each buyer's terms in full.
Is it legal to sell customer support conversations?
It can be, under a written license, when you own the tickets and your privacy policy and customer terms allow it. Support you ran for a client under contract is usually the client's data. Customers' personal data stays out unless your contract and the law allow it. Your own lawyer should read every draft.
Do I need my customers' permission to sell support tickets?
Your privacy policy and customer terms decide what you can license. If they do not cover licensing for AI training, settle that with your lawyer before talking price.
Can I sell Zendesk tickets, or does Zendesk own them?
Think of Zendesk as the filing cabinet, not the owner of the files. Your agents wrote the replies and your company holds the account. What limits a deal is usually your privacy policy and customer contracts, though your Zendesk subscription agreement can set rules on export and data handling, so read the version you signed.
Does the ticket data leave our systems?
Yes, eventually, stripped of identities and only under a signed agreement. Scale AI describes anonymizing on your side, within your own systems, before extraction; the other pages are vaguer about where scrubbing happens. Do your own removal first regardless.
Get a quote
Start with the head office ZIP, an inbox for our reply, then a handful of taps describing the helpdesk history. A person here reads it and writes back naming any buyer whose published entry rules your company clears, or telling you plainly that none does.
Describe the archive; never paste it. No tickets, exports, screenshots or sample files belong in this form, and this site has nowhere to upload them.
Said once, plainly: Sell Support Tickets makes introductions, and the buyer on the other end may pay us for one. More on the how this works page.
Received. Nothing has gone to any buyer.
- We set your answers beside every buyer's published entry rules (headcount, years, language, location).
- An email follows with any names that fit, each with a link to that buyer's terms so you can check our reading.
- Your company stays private unless you answer that email agreeing to an introduction.