Data Science and Personalization: Learnings From the Summit Floor

While we can all agree that personalization is important, it’s becoming more and more apparent that different organizations — and even different marketers — can have distinct views of what it actually means, and beyond that, what it takes to be successful. At the center of the back and forth is, unsurprisingly, the role of data and data science in personalization best practices — specifically, do you NEED a data scientist to be successful with heavy-duty, data-driven personalization?

We brought this topic to life during an extended breakout session at Adobe Summit — AKA our Personalization & Optimization Workshop 2016. This highly engaged, highly hands-on session brought together marketers, analysts, data scientists, and thought leaders from all corners of the industry and with a variety of personalization-experience levels. The goal? To have a deep, diverse conversation to break down some of the barriers around data science, and at the same time, solve some of the challenges digital marketers come against in this arena day after day.

So, how did it go? In a word: illuminatingly. Attendees piled into the room, and each selected a chip based on the topic he or she wanted to focus on. Participants included a senior-level data analyst from a high-tech firm, a marketer from a leading car manufacturer, marketers from the home-security space, nonprofit leaders, data scientists, and many more. And, of course, we had members of the Adobe Target team on hand — including data scientists and analysts of our own — to help round out the conversation.

A Common Data-Science Challenge
With this broad spectrum of people and expertise in the room, we dug in. One conversation that really stood out started with a marketer who worked for a nonprofit. She wasn’t a data scientist, nor did she have access to that level of analysis and interpretation within her organization. She was really a one-woman show who wasn’t leveraging any level of personalization in her work right then — and that’s why she was at Summit and in this session in particular.

Without the people, resources, or background, she wanted to understand her options, considering her end goal was to drive donations. Maybe some basic email marketing? Something else? While many organizations start with manual approaches to personalization — because that’s what they’re accustomed to — she recognized that she didn’t have the support to tackle that much heavy lifting. In her mind, data, algorithms, and automated personalization seemed like the way to go.

At first, the room was quiet; no one had any real, concrete answers for her, at least not in the beginning. One attendee mentioned that, from a Web perspective, she should tap a minimum of four people: a tech person, a creative person, an analyst, and a project manager — the latter, most likely, being her. But, is that too much for an organization like this, especially when it’s simply not in the resource- or budget-allocation plans?

Data-Science Experience Weighs in
In that very same room — with a marketer so new to data, data science, and personalization in general — sat a sleek, sophisticated team from a premier automaker, and they, of course, understood everything there was to understand about this conversation. They understand the buyer journey as it relates to their industry’s unique challenges, and they employ data science in a meaningful and incredibly elevated way to enhance the customer experience — at every point of the journey. They pull in audience data that signals “I’m in the market for a new car,” and then use some data-science magic to target potential buyers in a very granular and highly effective fashion. They use products like Adobe Target to help with the data-science heavy lifting and to optimize real-time decisions. This tight team of marketers and data scientists are like royalty — they’re the dot connectors, relevance drivers, and optimization linchpins of the organization.

We had an equally impressive team in the room from an online security and software business. To them, incorporating data science is practically second nature — they just do it. It’s woven into their models, so they can best figure out the buying propensities and marketing opportunities associated. But, unlike the nonprofit, they have marketers, analysts, and data scientists on hand — the trifecta of the data/personalization process. In their minds, they need all of the bells and whistles, especially their savvy data scientists and analytics pros. They couldn’t do their jobs or run their business — at least not with the same level of success and relevance delivery — without them.

Do You Need a Data Scientist?
So, back to the core question: does everyone need a data scientist? One attendee said yes, resoundingly — but, full disclosure, he’s a data scientist. From his perspective, companies that are serious about Big Data and the power it wields should have an expert on staff who can dig through it all and come out the other side with something truly meaningful. That said, he also acknowledged that many companies — the nonprofit, for example — simply don’t have the luxury of pulling someone with this skillset onto the team. Or, maybe, it’s just more than a small business needs — it’s overkill. So, then what?

Ultimately, the rooms landed in the same place: just because you don’t have a data scientist doesn’t mean you can’t employ data science. Adobe Target is extremely user-friendly and brings data science to all kinds of businesses — big and small, experienced and inexperienced, marketing-led and data-scientist driven. You don’t have to be a data scientist — or even work with a data scientist — to gain some serious value from data-driven processes such as machine learning. You can readily integrate it into your marketing right now — no heavy, scientific lifting required from you.

That said, I’m not advocating throwing the baby out with the bathwater! Just because marketers now have platforms and solutions that support marketing-driven data science doesn’t mean data scientists are obsolete — far, far, far from it, in fact. It simply means that organizations need to take good, hard looks at their structures, goals, data, and data-collection processes and decide what makes sense now as well as in the future. What are you doing now that’s working? What could you do based on the foundational pieces you have? What resources can you allocate to data-driven personalization — and what’s not in the cards, at least not right now?

Start by organizing these initial jumping-off points and go from there. Countless options exist at every level — from sophisticated carmakers and tech teams to growing charities and every business in between. All it takes is knowing where you are and where you want to go — and of course, recognizing that, with a little elbow grease and some smart solutions, data-driven personalization can get you there. And, if you need a little help? Check the chart above, which maps out the data science and personalization universe.