With the power of big data and AI — paired with the fragmentation of the media landscape — marketers have rapidly shifted to highly personalized digital marketing tactics. This move has been hugely successful, and in fact has become the norm.
Today’s consumers now expect personalized ads and offers. Compared to generic ads, custom messaging is typically more effective in generating higher click-thru and conversion rates as it resonates with a consumer’s specific needs, values, and priorities.
Of course, in order to build trust and meaningful engagement with consumers, you need to take a Goldilocks approach to executing personalized ads that are “just right” for your audience. Ads with too little personal relevance can fall flat, while those that are too spot-on can feel like an invasion of privacy.
This is where your segmentation strategy comes to the rescue: with the power of big data and machine learning, segmentation is the engine that can drive effective personalized marketing.
If you understand the wants and needs of your brand’s priority segments, you can craft more relevant messaging without being so specific that you raise “Big Brother” concerns around their privacy. And if you rely on first- or third-party data streams, you can target that messaging to people most likely to belong to each segment.
Combining segmentation and third-party data to inform programmatic ad buying
There are millions of individual data points available from hundreds of third-party data providers; not all will be available from all sample sources. Knowing who to call and what to ask for can be really overwhelming. It’s helpful to work with a partner who has experience connecting marketers to the right data streams specific to your business situation.
For broad communications geared toward acquiring new customers, you need to capture third-party targeting information about known segment members. To do this, your research partner needs to deploy your segmentation short-form typing tool among a larger sample of consumers contacted through specialty panels who then append the relevant third-party data.
Alternatively, you can use your Data Management Platform (DMP) or partner with a data technology firm to build a targetable custom audience (i.e., millions of devices that look like segment members from your survey) that can be deployed across marketing platforms and exchanges.
Database scoring to improve segmentation personalization
If you have a robust first-party database, you also have the opportunity to use lookalike modeling to predict segment membership across the customers in your database and use that information to personalize your sales, service, and other touchpoints with current and lapsed customers. This is a practice known as database scoring.
By applying this tactic, you can arm your call center with different up-sell or win-back scripts based on segment membership. The emails you send can be triggered or customized based on segment. Logged-in users can enjoy different website or app experiences without taking any proactive customization steps.
If you need a segmentation that will inform your database deployment, then you need to involve the sales and service teams early. If they’re not consulted at the start of the project (and at key milestones throughout), they’re unlikely to buy into the segmentation or adjust their behaviors in segment-specific ways.
This is especially crucial in business-to-business (B2B) settings where you may need to more simply define the segmentation subject to the constraints of a sales rep being able to classify someone in real-time using simple “if-then” rules.
Engaging stakeholders and incorporating existing market segmentation research
These extra steps cost money and your research budget is tight. You might be wondering what you’re going to have to give up in order to get these extra pieces approved — but it’s not a zero-sum game.
A cost that looms large relative to your research budget is just a rounding error on a media buy. Since these activities increase the efficiency of the advertising expenditures, you may be able to get those teams to contribute.
Of course, whether or not they pitch in funds, they should still be active participants at the beginning. After all, if you plan to activate your insights in CRM or advertising, then there are specific research steps that you need to plan for and execute. These types of activations should not be afterthoughts.
That said, while you’ll get the most efficient and effective outcome if these steps are baked into your process from the start, you may already have a segmentation you love.
Don’t despair — it’s still possible to do these types of activation steps and extend those strategic wins into tactical wins, too.
Is segmentation still creating effective ad targeting?
Your pre-existing segmentation scheme should be critically assessed in light of new market conditions and consumer behavior shifts. If it has become obsolete, return to step one and start to design a new strategic framework to drive your re-entry strategies and win in a post-COVID world.
If you find evidence that your segmentations are still spot-on and are hitting all of the key wedge issues that differentiate consumers in your category, then congratulations! But keep in mind that consumer media habits have shifted markedly since the pandemic.
With millions having adopted (or abandoned) news, and millions more having cut the cord as they flocked to adopt more streaming services, the digital targeting models you have already been using may not work as well anymore. It’s time to rebuild (or update) your targeting metrics, custom audiences, and database scoring models for success.