Where does FullContact get its data?
FullContact defines data in two important ways: private and public. Public data is sourced from numerous locations and we openly share this data, while abiding by relevant terms and conditions. Private data comes from 3rd party Trusted Sources and consumers themselves, we do not disclose these sources. API data customers are only ever allowed to license the public portion of our identity graph; we do not sell or release data sourced from our users.
Data Source Types FullContact Leverages:
- Public Websites: We capture information from over 15M public websites and then tie that into data from other sources, including 3rd-party data providers ("Trusted Partners").
- Trusted Partners: We partner with key providers of data including demographics, firmographics, affinities, and social data. Depending on the model of partnership, they provide a combination of public and private data.
- Social Networks: we have partnerships with select social media networks enabling us one-on-one lookups to their database. Furthermore, we use public APIs on some platforms to extract data, while complying with each of their respective terms of service.
- In-house Data Research Team: FullContact employs hundreds of data research analysts that use a combination of technology and human intelligence to create new connections and strengthen existing connections.
- Crowdsourcing Network: Many of our data customers embed our data within their own SaaS offerings. We provides APIs which allow the extended user base of those applications to contribute updates and corrections to our data.
- Service Users: These are users of our FullContact consumer applications for contact management. All of the data they provide is considered private and used only for internal identity resolution/verification.
- Identity Resolution Cooperative: Select data customers may optionally contribute their contact data to our system. With few exceptions, all data provided in this way is considered private and used only for internal identity resolution/verification.
How do the FullContact algorithms connect the dots on social data?
FullContact provides identity resolution for all of the disparate pieces of public contact information out there on the web. We do this by aggregating billions of contact records, all with numerous attributes, including quality, freshness. and frequency. Our patent pending algorithms process all of this data and automatically produce clean, accurate full contact records. As a final step, we then check each data element to make sure that it’s publicly available before providing it to our customers.
For example, imagine that five people are sitting around a table and each pulled out a contact record for our CEO, Bart Lorang. One person might have his Twitter handle and email address; a second might have just an email address and an image; and the others might have various other pieces of data. If all of that information were mashed together, as humans, the group could put together a "FullContact" record for Bart with a high degree of accuracy. This is basically how FullContact works. We take a variety of data points and utilize algorithms to logically build clean, accurate FullContact records.
Where can I find a full list of the social network types that FullContact can return?
A full list can be found here.
Can I re-license FullContact’s data?
No. We do not allow relicensing or reselling of our data. For existing customers, please see our Terms of Service for more details on relicensing and reselling.
How accurate is FullContact's data?
Our accuracy rates are over 90% and match/fill rates are usually between 20-60%. Match rates refer to whether we can return at least one social URL for a given email or other query parameter. Our fill rates are often dependent on the nature of the queries we are given. For example, a list of old or out-of-date email addresses would typically result in a lower fill-rate than a list of current, accurate email addresses.
Geographically, we see matches that fall within the following percentages:
United States: 40-60%
United Kingdom: 20-40%
Australia / South Africa: 20-40%
Other International Countries: 10-30%
To note: When querying MD5 Email Addresses, we typically see match results of 10-40%
The data may change on people profiles, why is that?
Our algorithm is constantly at work to improve, update, and enhance our data set across our 1 billion people profiles. Not only are we attributing new data to people, we also may be cleaning up details like affinity or social data (for example). Partners of ours may also be recalculating data and coming up with new results. Keep in mind the data you (as a customer) may have access to, may be different if it's not the latest refresh from our data. This is also where our new Enrich API comes in where you can get pushed the latest and greatest data in real time, without having to re-query us.