LinkedIn is committed to protecting member privacy. As a result, we've moved towards privacy-enhancing technologies like Group Identity. Group Identity uses our first-party data to group member accounts based on shared professional traits, like job title, company, and industry, to help deliver ads and measure campaign conversions on a portion of the LinkedIn Audience Network.
Important to know
• Private Testing cannot measure lift for a portion of traffic on the LinkedIn Audience Network.
• Some members might appear in both the test and control groups, but our data and platform checks have shown that the effect is generally insignificant.
• Some members might appear in both the test and control groups, but our data and platform checks have shown that the effect is generally insignificant.
As part of our commitment to member privacy, we evolved our core testing products, A/B Testing and Brand Lift Testing, which historically relied on individual identifiers. With Private Testing, marketers can conduct A/B tests and Brand Lift tests in conjunction with Group Identity.
How Private Testing works
Private Testing works behind the scenes, so marketers continue receiving A/B test and Brand Lift test results. Private Testing stores critical test information on a virtual device, which is an isolated, server-side environment that has its own computing and storage resources. This helps achieve security isolation. This allows LinkedIn to confirm whether a member account is in the treatment or control group without identifying the member.
Private Testing provides the following enhancements:
- Members' accounts receive the right treatment for their assigned group. For example, group A or group B, or the test or control group.
- For Brand Lift Testing - If the member is in the control group, they will not be shown the tested ad.
- For A/B Testing - If the member is in treatment group A, they will not be shown an ad for Group B.
- Attribution of performance to the ads, including Brand Lift test survey responses and A/B test key results.
- Enhanced member privacy by stripping member identifiable data from the request.
With these improvements, LinkedIn can measure a campaign’s impact while enhancing members’ privacy.