Third-Party Data
By the AdFlint research team · Last reviewed November 2026
Audience information bought or licensed from a broker that has no direct relationship with the people it describes.
Brokers assemble demographic, intent, and purchase signals from many sources and sell segments for targeting or enrichment. Advertisers use it to reach categories their own data cannot describe, particularly for cold prospecting. Its two structural problems are accuracy, since you cannot inspect how a segment was built, and provenance, since consent was collected somewhere else, which is why platforms have steadily narrowed the third-party segments they accept.
Key takeaways
- Third-party data comes from brokers who have no direct relationship with the people in the segment.
- You cannot inspect how a segment was built, so accuracy and consent are hard to verify.
- Platforms have narrowed what they accept, so check current availability before you plan around it.
- Test every paid segment against a free audience and include the data fee in cost per result.
In practice.
Third-party data is audience information that comes from a company with no direct relationship to the people it describes. Data brokers and aggregators build segments such as home buyers, frequent travelers, or small business owners from many sources, including purchase panels, public records, app activity, surveys, and modeled guesses. They license those segments to advertisers, ad platforms, and data management tools. You buy access to the segment, not to the people behind it. The pricing is usually a data fee added per thousand impressions, a flat licence for a list, or a subscription, so ask for the total cost per usable record or per thousand targeted impressions.
The segments reach you in a few ways. A demand-side platform may offer a catalog of them for programmatic buying. A broker may sell you a list or an enrichment file that adds attributes to your own records. Some ad platforms used to offer broker-supplied targeting categories directly, but platforms have narrowed those options over time, so check what your platform lists today rather than assuming. The advertiser picks a segment, pays a per-thousand-impression fee on top of media cost, and the platform matches the segment to browsers or devices it can identify. Match rates are another hidden number. If a vendor sells 100,000 records and 30 percent match to people the platform can reach, you pay for the full file and use a third of it.
Two weaknesses explain why it has lost ground. The first is accuracy: you cannot see how a segment was built, and labels like in-market for a mortgage can rest on thin evidence. The second is provenance. The consent given when the data was first collected may not cover your use, and privacy rules in many places make advertisers responsible for that. Third-party data also depends on identifiers, so the browser limits on cookies reduce how many people a segment can reach. Compared with your own customer data, it is less precise and more exposed to policy changes. Platforms also apply their own policies. Several have removed or restricted broker categories over time, which means a segment that worked last year may be gone, so confirm availability before building a plan on it.
Where it still helps is cold prospecting for products that your own data cannot describe, such as reaching high-income households for a premium service in a market where you have few customers. The common errors are paying a premium without a test, buying a broad segment and assuming it is targeted, and failing to ask the seller about sourcing, update frequency, and consent. Another error is judging it on last-click results alone, since a prospecting segment mostly introduces people to you and does not close sales. Compare with contextual targeting, which uses the page content instead of data about the person, and with your own list. Both avoid the provenance problem, and contextual needs no identifier at all.
Measure it against a control. Run the segment in one ad set and a plain interest or broad audience in another, with the same creative and budget, and compare cost per qualified lead after a few weeks, including the data fee in the cost. Ask the vendor for the segment's source, refresh cycle, and size. This week, list every third-party segment you pay for, find out what each one costs per thousand impressions, and pause any that you cannot show beating a free alternative. Keep contracts short so that you can leave if quality drops. Finally, keep a record of what you bought, from whom, and on what terms. If a regulator or platform asks, being able to show the source and the consent basis protects you.
A financial planner tests a paid segment
Suppose a financial planner buys a high-net-worth household segment through a programmatic platform. Media cost is $8 per thousand impressions and the data fee adds $3, so the planner pays $11 per thousand. A $2,200 test buys 200,000 impressions and produces 18 consultation requests, or about $122 each.
As a control, the planner spends $2,200 on a broad audience with the same creative at $9 per thousand, which buys about 244,000 impressions and produces 21 requests, or about $105 each. The paid segment is more expensive and no better. The planner drops it and puts the saved fee toward a customer-list lookalike audience. Small differences like this are common, which is why the control group is the whole point of the test.
Third-Party Data compared with
The settings this gets confused with, and how to tell them apart.
Common questions.
What is an example of third-party data?
A broker's segment of households that recently moved, or a list of people labeled as in-market for a car. The broker compiled it from sources unrelated to your business.
Is third-party data legal to use?
It can be, but the legal responsibility often falls on the buyer to confirm the data was collected with appropriate consent and disclosures. Ask the seller for documentation and speak with a privacy professional for your markets.
Why is third-party data going away?
Browsers limit the cookies it relies on, privacy laws have tightened, and ad platforms have restricted broker segments. It is shrinking rather than disappearing, and reach varies by platform and browser.
Is third-party data worth the cost for small businesses?
Rarely at small budgets. Your own customer lists, site visitors, and platform interest targeting are free or cheap and easier to verify. Test any paid segment against those first.
You should not need to know this to advertise.
AdFlint handles the settings for you, inside the Google and Meta accounts you already own.
Next step: How AdFlint tracks and optimizes these numbers
Try AdFlint free