Guides · Published
How Much Money Do Companies Make Selling Your Data?
Estimates of what companies earn from personal data range from about $263 to over $700 a year per person. Here is what those numbers mean and how business data differs.
Short answer
There is no single figure. Estimates of what companies earn from personal data run from about $263 a year per person for advertisers to at least $700 a year for a US user, depending on who is counting and what they include. Those numbers are what platforms earn from ads and data use, not payments made to the people whose data it is.
If you run a company, the question is different. Your business can license records it already holds, such as support tickets, CRM notes, project files or code, to AI developers. Payment depends on the buyer and the data, and it usually comes only after a buyer reviews a sample.
What the personal data estimates measure
Most published numbers describe how much value a consumer generates for an advertising business. They are estimates, and they rest on different assumptions.
- Advertiser value: Datapods estimates that the data one person generates in a year is worth about $263 to advertisers, with medical uses potentially around $110. The site notes the figure could be much higher (Datapods).
- Platform revenue per user: Proton points to Facebook's disclosed 2023 figures, where average annual revenue per user was $217.26 in the United States and Canada and about $70 in Europe. Proton's own analysis puts the annual value of a US user's data at at least $700 and says that number has risen since 2022 (Proton).
- What people say they would accept: A Security.org survey found the average American would sell their social media data for just over $2,000 (Security.org). That is a stated price, not a sale, and it is not what companies pay.
- Where the money goes: Privacy Bee describes targeted advertising as the core of billion-dollar industries built on personal data, and covers data brokers that most people have never heard of (Privacy Bee).
The takeaway: a single person's data is worth little on its own. Value comes from scale, from how specific the data is, and from who buys it.
How business data is different
Company records are not the same as personal browsing or purchase history. Business records can include technical detail, workflows and expert decisions that AI developers need to train agents. That is why buyers may pay for them, and why the price is set by sample review rather than by a per-user formula.
| Personal data | Company records | |
|---|---|---|
| Who has the data | Each consumer | The business |
| Typical measure | Average revenue or value per user | Negotiated license for a copy |
| Who gets paid | Usually the platform or broker | The company that owns the records |
| What sets the price | Scale and targeting value | Fit with a buyer's needs and a sample review |
How a company licenses its records
Smelted is a free matching service for this case. It licenses anonymized internal business records to AI data companies. The company keeps ownership, which means it grants a license to a copy. The company chooses what is included and which buyers it meets. Names, contact details and IDs are removed before anything is shared. The buyer pays Smelted's fee, so the company pays nothing.
The process runs roughly like this:
- A short intake. Tell us which tools hold your records and how many years they cover. A 1-2 minute form at smelted.ai/get-started gives a ballpark range.
- A 15-minute call. Book one with Chris Irving at calendly.com/smelted/intro-call, or call 954-682-7560.
- Matching. Smelted finds which buyers fit your company and introduces you to 2 or 3 of them.
- Sample review. A buyer reviews a sample before it names a real number. Smelted does not quote a figure before that.
- Signing and payment. Plan on 1 to 3 months from the first call to payment.
Smelted works best for companies with about 50 to 500 employees in staffing and recruiting, engineering and architecture, B2B software, and science labs or biotech R&D. Companies under 30 employees should expect buyers to pay less; Chris can talk through what is realistic.
What affects the price
- Volume and history: More years of records, across more tools, usually means more to review.
- Type of data: Support tickets, specifications, code and CRM notes differ in value. Audio, voice and video datasets are handled case by case.
- Clean data: Records that are easy to anonymize and well organized take less work to prepare.
- Buyer interest: The price depends on what a specific buyer needs.
Things to check before you sell
Before sharing any data, decide what you are willing to license and confirm you have the right to do so. If the records belong to clients, the answer may be different. Smelted does not give legal or compliance approval, so check contract terms and any privacy obligations with your own advisers. Physical books and published works are not something Smelted can handle, because the copyright is not the company's to license.
FAQ
How much money do companies get for selling your data?
For personal data, estimates vary widely. Datapods puts advertiser value at about $263 a year per person, while Proton reports Facebook's 2023 revenue at $217.26 per US and Canadian user and estimates a US user's data is worth at least $700 a year. None of these numbers is a payment to the individual.
Can you sell data to AI companies?
Yes, if you are a company with records you own and have the right to license. Smelted matches businesses with AI data buyers and removes names, contact details and IDs before anything is shared. The company keeps ownership and chooses what is included.
How to sell data for AI?
Start by listing the systems where your records live, such as Zendesk, Jira or GitHub, Salesforce or HubSpot, and how many years they cover. Then book a 15-minute call or use the short form to get a ballpark range. A buyer reviews a sample before a real price is set, and payment usually follows 1 to 3 months after the first call.
How much is one person's data worth?
It depends on how the value is measured. Security.org found the average American would sell social media data for just over $2,000, but that is a stated price, not what companies pay. Advertising-based estimates like Datapods' $263 a year and Proton's $700 or more reflect what platforms earn.