Marketing has come a long way from the days of focus groups and customer surveys. One of the most important developments has been the rise of predictive analytics, which has become accessible to a wide range of marketers.
So many new technologies are available to marketers today. Keeping track of all the latest developments can seem overwhelming.
These four important steps can help traditional marketers become experts in predictive marketing and reap the rewards for early adopters of a powerful new technology.
1. Prioritize business sense above mathematical skills
If you're considering signing up for a data science course, wait just a minute! Just like you don't need to know how an internal combustion engine works to drive a car, you don't need to understand the underlying math that powers predictive analytics to be a predictive marketer.
For predictive marketers, it's much more important to understand your business, your target market, and your customers than to be able to crunch raw data.
Even if you have the ability to uncover trends in huge piles of unsorted data, you need an entirely different skill set to draw meaning from these trends and turn them into market strategies, product ideas, and successful campaigns.
Moreover, new technologies can increasingly do a lot of the number crunching that once was the job of data scientists. These technologies keep the complicated math "under the hood" and present insights in a way that savvy marketers can easily understand and use.
2. Focus on the questions, not the solutions
The best way to begin with predictive analytics is to determine which questions will help you better understand your business and your customers.
Start with a hypothesis— e.g., "new competitors are eating into our market share" or "customers have left because they're don't like the latest product lineup"—which you can test with the right predictive analytics tools.
A director of customer relationship management at a large discount retailer recently uncovered a huge marketing opportunity by simply asking how many of its buyers were repeat customers. Customer data analysis showed that most of the company's customers only bought once. This insight laid the groundwork for a company-wide effort, led by the now-promoted director, to boost customer engagement and find customers with higher predicted lifetime value.
The key lesson here is that predictive analytics algorithms that can forecast things like customer lifetime value or likelihood to engage function as tools to solve a problem, not solutions in themselves. Many failed attempts to adopt predictive analytics stem from marketers searching for the magic algorithm that will reveal the perfect campaign instead of starting with an idea and testing it.
3. Predictive marketing should blend art and science
Data science will not replace the need for creative thinkers.
Dan Pingree, CMO at Moosejaw, an outdoor gear and apparel retailer, describes predictive analytics as "a way to inspire and validate the creative process."
Insights from predictive analytics tools—which can uncover unexpected buyer clusters or surprising purchasing trends—show you customer personas and marketing opportunities that will inform creative campaigns. Remember that while predictive marketing data will provide important insights, you should never shut off your creative instinct.
4. Learn from people with traditional marketing experience
Although technology has deeply transformed marketing, you can still learn a lot from traditional marketers as you adopt new techniques. Database marketers focusing on direct mail campaigns tend to have the most experience regarding predictive analytics. Because sending a postcard or catalog is much more expensive than sending an email, database marketers have relied on predictive models, such as likelihood to buy and clustering, to target their messages to segments most likely to respond.
* * *
Don't reinvent the wheel. The principles used in database marketing for many years apply directly to predictive marketing. If you have a current or former database marketer on your team, take him or her out for lunch to learn about advanced segmentation. If you don't have a database marketer on your team, look for somebody in your professional network.
Though the technology that powers predictive analytics may be complex, it should be something that every marketer can add to their toolbox.
Start small, focus on measurable campaigns, and you'll be an expert predictive marketer in no time!
