Imagine that you have a friend who’s a woman in her thirties. She’s married with one child around two years old. And on Pinterest you suddenly see her pinning things like this:

You might deduce that she is pregnant again. Given her demographic profile and social media behavior, it’s a reasonable guess. And if you did guess that, you’d be engaging in an informal form of predictive analytics.

Now imagine that you’re a retailer and a female who’s aged around 16 suddenly starts looking at items on your ecommerce site that are typically bought by women who are pregnant. And she buys some of them on your site and in your stores.

So, to encourage her to buy more, you send her a few coupons for similar items and before you know it, her angry father is in your store accusing you of encouraging his teenage daughter to get pregnant.

This happened to Target. But what the father soon found out was that his daughter actually was pregnant. Target knew it before the father did.

Target realized that pregnant women in the second trimester often bought unscented lotions as well as supplements like calcium, magnesium, zinc, and other items.  That behavior identified likely customers, but then Target had to be careful about how they marketed to them because a pregnant woman doesn’t want to get a flyer with the headline “Congratulations!” if she hasn’t registered on a Target baby wish list. The incident with the teenage girl and her father was very instructional.

Finally, when Netflix presents you with these suggestions? Yup, predictive analytics:

A spokesperson for Netflix described the breadth of the company’s data and signals: “We monitor what you watch, how often you watch things. Does a movie have a happy ending, what’s the level of romance, what’s the level of violence, is it a cerebral kind of movie or is it light and funny?”

30% of Amazon’s revenue is produced by its recommendation engine. In 2012, the Obama campaign hired more than 50 analytics experts to prioritize voters and determine which messages and outreach techniques were likely to be most effective with each individual voter.

Predictive analytics in B2B

Large B2B companies have been using predictive analytics for years, too, to better prioritize sales leads, determine which products a prospect would be most likely to buy, nurture contacts who aren’t yet ready to buy, and develop more reliable sales forecasting. It used to be that only the largest companies, like IBM, SAP, Target and Amazon, had the data and data scientists to do it, but that’s now changing.

Predictive analytics is being democratized. Lattice and Mintigo are two of the companies that are providing cloud-based B2B predictive analytics services that eliminate the need to hire increasingly-pricey data scientists internally. Their SaaS services start with the company’s internal CRM and marketing automation data, and then they add in data from thousands of public sources such as company revenue and income, number of employees, number and location of offices, executive management changes, credit history, social media activity, press releases, news articles, job openings, patents, etc. 

From this they use data science to identify common characteristics of the accounts that were won by sales, and predict the likelihood of closing each prospect. For example, a good signal for an office supply company to contact a prospect may be when they sign a lease for a new building, or put out a press release about expanding to more cities or hiring many new people.

Sometimes the signals are far more obscure than that, though; for a company selling CAD-CAM software a key signal was the number of design engineers prospects were hiring and the number of workstations in use.

Sales has prioritized leads and sales people have important new information about the accounts, which cuts down on their research time. Marketing has segments of lower-priority prospects to nurture. And predictive analytics can be equally useful in growing existing accounts and closing new ones.

This goes way beyond the lead scoring of a marketing automation system, as valuable as that is. Marketing automation typically just uses the information from the CRM and the ‘digital body language‘ of a prospect’s online interactions with the company’s website, emails, and other digital communications. Predictive analytics companies are adding a huge amount of data to that, and then sifting through all of it to find the most useful buying signals.

Actually, this image doesn’t fully capture the advantage of predictive analytics, which can be basing its recommendations on 50 to 100 times more data than lead scoring that’s based on internal data.

With new sales and marketing technologies like marketing automation and predictive analytics there’s a huge advantage to early adopters who get it right. Marketing automation is only being used by about 12% of tech companies, and 5% of all companies, and many of them are not using it fully, or well.

B2B predictive analytics is providing double digit increases in leads, opportunities and sales – sometimes high double digits. Early adopters of sales and marketing technologies can reap huge sales increases while their competitors are wondering what hit them. But several years from now predictive analytics is likely to be table stakes – everyone will be doing it, or being left behind. To the early adopters go the spoils!