With the power of big data and AI, and the fragmentation of the media landscape, marketers have rapidly shifted to highly personalized digital marketing tactics. This shift has been hugely successful, and it has become the norm— consumers have come to expect personalized ads and offers, and they are typically more effective (generating higher click-thru and conversion rates) since they resonate more with the consumer’s needs, values, or priorities than generic ads.
But this is a Goldilocks problem that you need to execute “just right” in order to build trust and meaningful engagement with consumers. Ads with too little personal relevance can fall flat, while those that are too spot-on can feel like an invasion of privacy. And with COVID-19 impacting different consumers in different ways, it’s critical that you personalize your messaging in a way that resonates with your target audience.
This is where your segmentation 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 trigger the “creepy” factor. 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.
There are literally millions of individual data points available from hundreds of third party data providers, and they are not all available from all sample sources. So, 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 like you to the right data streams, cherry-picking the right partners and data points for your specific 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 large 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.
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 armcall 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. And 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, 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 settings where you may even need to more simply define the segmentation, subject to the constraints of a sales rep being able to classify someone into a segment in real time using simple “if…then…” rules.
These extra steps cost money, and your research budget is tight, so you’re 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. Whether or not they pitch in funds, they should be active participants from the start. After all, if you plan to activate your insights inCRM 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 to help extend the strategic wins into tactical wins, too.
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, hitting all the key wedge issues that differentiate consumers in your category, then congratulations! But keep in mind that consumer media habits have shifted markedly in recent months. With millions having adopted (or abandoned) news, and millions more having cut the cord as they flocked to adopt more streaming services, any digital targeting models you have already been using may not work well any more. It’s time to rebuild (or update) your targeting metrics, custom audiences and database scoring models.
You must be logged in to post a comment.
