Customer data onboarding has become a gold standard practice for today’s marketers working to achieve people-based marketing at scale. Most companies understand what it is, and what the benefits are, but often overlooked are the intricacies and rigor involved in the partner selection process. Here, I’ll dive into some of the fundamentals of preparing your data for effective onboarding, the key factors to measure and test before selecting a partner, and the steps for evaluating the most effective activation partners for your company’s data set.
Making the most of your data
The most important ingredient for the successful onboarding is a foundation of quality customer data. Onboarding is a classic garbage-in/garbage-out scenario. This requires companies to develop a data management program to manage, update, and remove customer information and maintain data hygiene via procedures such as the US Postal Service National Change of Address (NCOA) processing. This is important because the onboarding matching process is dictated off matching personally identifiable information (PII) elements such as name, address, email, phone, IP address, and others; so, maintaining accurate customer profiles is the first step to effective onboarding.
Measuring and testing before selecting a partner
Measuring and testing are less important with direct onboarding than with third-party onboarding. Direct onboarding is commonly used for a media or social targeting use case where email is the most popular mechanism for matching to the provider. Third-party onboarding is a bit trickier; let’s first define each:
Direct onboarding matches an anonymized customer data identifier directly with major media and social networks, such as Facebook, Google, and Viant. Uploaded data is matched to the associated platform’s persistent ID for each customer record from the network.
Third-party onboarders, such as LiveRamp, DataLogix, and Neustar have relationships with data providers and publishers, amassing a network of cookies and device IDs associated to registered/logged in customers. This allows marketers to upload anonymized customer data to be matched to the network of digital identifiers.
In the third-party onboarding data scenario, matching is predicated on the match waterfall determined by the marketer. Matching can be done at the individual level or the household level. You can match on one identifier, such as an email address, or you can match on a sequence of identifiers in the order in which you specify: name, address, ZIP 1-4, email, phone number, etc.
Depending on the provider, your match rates could vary, as the providers have different data partnerships and agreements which, in turn, develops differentiated networks of known customers to match against.
It is best practice to send the same customer file with the same matching logic to each of the onboarding partners you are evaluating. This will give you a good idea of the differentiation of matched records. In addition, marketers can request match-back files to determine what records were not matched and compare and contrast what universe was found with each provider. This may lead you to contract with one onboarding partner or develop a process to onboard the un-matched records with a second onboarding provider, should the matched universe be discrepant.
Once customer records are matched, the onboarding provider will match records to cookies and devices. This is another point of evaluation. Depending upon your use case and where you are looking to utilize this data, device IDs or cookies may be more valuable; so, evaluating the scale of each of these identifiers is critical.
Evaluate the most effective activation partners
Once the file has been matched to cookies and devices, the data is now in a usable format. This means data can now be pushed to data management platforms (DMPs), demand-side platforms (DSPs), or directly to social networks or other platforms with which the onboarding partner has a data connection. Each of these activation points will have its own set of match rates. This all depends on the cookies/device IDs being tracked by each of the onboarding partners and how they overlap with those being tracked by the activation providers.
This is yet another evaluation point, as an onboarding partner may have more overlap with one DSP than another. Or it may be an opportunity to evaluate activation partners for your data, based on which can provide the greatest scale.
If you are operating with a DMP, all of your data should flow to your DMP to better facilitate orchestrating your audiences for targeting, measurement, and personalization use cases.
Lastly, it is important to note that the activation of your onboarded data does not have to be solely used for media activation use cases. Marketers should also look to leverage this data for measurement and personalization. Marketers can deploy multi-channel attribution, look-alike modeling, audience measurement, and insights from the onboarded data set. For personalization, this data can inform site personalization, dynamic content, and site search recommendations.
Summary
While onboarding has become a gold standard practice, the partner evaluation, optimization of the onboarding process, and the utilization of the data for additional use cases has lagged behind. As marketers, we must consider these intricacies when selecting an onboarder and ensure use cases are established before contracting. To start making the most of your data, read “The Marketers Guide to Identity Resolution” to take your first step.
