All marketers know that 1 to 1 personalization, today’s holy grail of marketing, starts with data. But actually achieving personalization is another story. It’s common to start out thinking something like, let’s track everything and figure out how to use it later, but that approach rarely works. It’s messy, disorganized, and if you aren’t sure what you’re looking to track or don’t have a mapped plan for how you’re going to use the data you collect, you’ll end up creating piecemeal experiences. And no customer wants that. There’s way too much data in the world for humans to scale it, and the reality is we’re never going to catch up without help. Top that off with the fact that what customers say they want and what they really want are different things, and utilizing data to its full potential – what we call data activation – becomes essential.
Activating data means every customer interaction is influenced by all available dataーincluding all prior transactions, on-site and offline behaviors, campaign engagement, product interactions, real-time behaviors, and other customer attributes gathered across channels and devices. Truly activating data can cut through the noise, the confusion, the deluge, and deliver the optimal customer experience
Like anything worthy of investing in, the steps to improving customer engagement with data activation take time. If you’re just starting out on your journey, here are four areas I’d recommend covering first:
Surface the MarTech Struggle
Data activation requires technology, which in-turn requires buy-in, so a good place to start is with the why. Begin building out a business case by outlining your current issues and what you think you might need to resolve them. Keep a lookout for anything that results in fighting with your technology, like breaking under the weight of more complex use cases, or the inability to make decisions easily (e.g. without the help of a data scientist or an engineer). These are sure signs that you need to modify your stack.
Think about struggling with MarTech like a traffic jam. Drivers know where they want to go, but getting stuck in traffic prevents them from actually getting there. Legacy, archaic MarTech creates a similar challenge. Many marketers know what they want to achieve, but their stack prevents them making it a reality. Really dig into those pain points and don’t be afraid to think outside the box when it comes to what might solve them — flying cars included.
Identify Your Needs
There’s a lot of technology out there that claims to have what marketers need to fill the gaps, but be careful. It’s not true data activation unless it provides a few key things:
- A single customer view: The customer view powers everything, so make sure what you’re being promised is up to par. Your single customer view should stitch together anonymous and identifiable sessions from every single touchpoint and adapt to your definition of a customer.
- Predictive intelligence: Once you have a single view of the customer, you need algorithms that you can run against these profiles to intelligently predict future behaviors. Marketers aren’t mind readers, but with the right technology we can get pretty close.
- Scalable architecture: A lot of systems will break under the weight heavy computation across large data sets, which is a pretty big problem when experiences are at their best when they’re informed by the fullness of data. Make sure the platform you choose can handle the data complexities you have today, as well as ten years down the line.
- Powerful recommendations: It’s great to have a single view of your customer and solid predictions about their future behavior, but activating data requires one more key step. Make sure you choose a platform that can turn all that insight into actions with personalized recommendations. Make sure that the recommendation engine is built with pre-computed technology so the recommendations are lighting fast to retrieve even at the largest scale. Lastly, recommendation engines need to be customizable and change with your business needs.
- Robust templating solution: You need a powerful, but flexible template language that allows you to present your recommendations to the end user in a variety of outputs (HTML, JSON, etc). Your templating language needs to handle basic key-value replacements, complex iterations, control flows etc. You also want a language that’s open-sourced and has an active community behind it. (My recommendation is to look for solutions that use the Liquid templating language.)
Demo with Intention
It’s absolutely crucial to make the most out of your demo, but it can be tricky. A lot of providers focus on the business, spending most of their time addressing broad needs and use cases. Instead, suss out whether or not the technology will actually allow you, the person in the demo environment, to accomplish what you outlined wanting to do in the beginning of this process.
Don’t be afraid to get into the nitty gritty details. In fact, insist on it. Ask for examples of how you can do things like complex iteration. Ask how they handle each use case or capability at scale (as in millions) in a short time frame. Ask for customer references so you can confirm the claims are legitimate. Ultimately, a demo you can trust is one that shows you more than tells you. If you aren’t at least 80% confident in the solution by the end, it’s either not the solution for you or it wasn’t a good demo.
Look for a Partner, Not a Tool
The Marketing industry is changing at a faster rate than anyone could have predicted, so you need to make sure that you’re working with third parties that won’t leave you high and dry during a sudden shift. After all, a good partnership will take you a lot further than a good price. Look for a team that’s made up of the right people: those who really believe in what it is they’re providing, not just selling it. Make sure they want to help you get the most out of their offerings, and that they can brainstorm new use cases to help propel your business forward. Solutions that solve for now are great – but in a world where things are moving at lightning speed, you need tools that can scale and solve for now and later. At the end of the day, their solution helps your business and your business helps their solution. You’re in this together.