predictive analytics crm

3 Predictive Analytics for CRM

Predictive Analytics are all about finding the insights that will help you understand what might happen in the future. They help you recognize patterns in historical data, repeated transactions, and relationship cues. If you use predictive analytics effectively, you can facilitate a more proactive business approach.

There is a gigantic and horrifying sea of predictive analytic research & theory out there – so much so that it could make your head spin. So, for the sake of sanity, let’s hone-in on three of the major predictive analytics types and how they can make you a smarter, savvier, sales or marketing professional.

The three predictive analytics we’re going to discuss are:

  1. Sequencing
  2. Cross-Selling
  3. Lack of Action

1. Sequencing

Simply put, sequencing has to do with analyzing probability. If a target performs action A (downloads a whitepaper), followed by action B (clicks on a pricing page), what is the probability they’ll perform action C (buy the product)? This concept stems from the work of a Russian mathematician named Markov. Markov researched probability theory and determined that action C is equally as likely to occur if actions A and B occur first.

Seems easy enough: all you do is look at the historical patterns of behavior for A and B to determine your C outcome.

But, what if you don’t have a lot of historical data?

markov
Andrey Markov

If you don’t have enough historical data to analyze how two or more events affect the desired outcome, there’s another way. Part of Markov’s model includes an assumption about the data. Not surprisingly, it’s called the Markov Assumption. If you consider only the last successful event to predict future behavior, you can assume the probability between that single event and the desired outcome is about the same as it would be otherwise. For example, ideally, we’d like to see Jane download a whitepaper, click a pricing page, and then buy a product. But Markov’s Assumption says if someone who visited the site before Jane downloaded the whitepaper and then bought the product, it’s likely that Jane will do the same. She doesn’t have to download the whitepaper AND click the pricing page for us to assume she will buy the product.

What does that tell us? We don’t need gobs and gobs of historical data to start a decent sequencing pattern. We can begin to discover sequences based on one action that leads to a purchase. That means you can initiate data sequencing today based on your most recent successful conversion, and then continue to tweak as you gather more and more historical data to cross-reference.

So what are some ways we can use sequencing analytics for better sales and marketing?

Marketing: One way marketing can use sequencing is with behavior-triggered email campaigns. If a person takes actions one or two, they are put into a specific track of the campaign. As you monitor the actions taken throughout each campaign, you’ll begin to recognize the patterns to customize your messages going forward.  Two tips to keep in mind for your behavior-triggered campaign:

  • Knowledge first, branding later. Create a campaign filled with rich, engaging content that will help the recipient. As they move through the cycle, sprinkle in your branding. The point is to establish trust and position your company as an insightful resource. Once the trust is earned, everything else will follow.
  • Use mobile to help you sequence. More people than ever are reading emails on their mobile devices. Optimize your campaign for both desktop and mobile to help get a better grasp on your audience. Are you seeing more conversions from mobile? Weave that into your sequencing analysis to refine the approach. For example, make it easier for them to take action via mobile by replacing hyperlinks with buttons. Buttons are bigger and easier to touch on smaller screens.

Sales: Sequencing fosters proactive approaches to the buying cycle. For example, if someone visits your website, downloads a whitepaper, and requests a free trial of your product, you can assume they’re pretty interested. Using the sequencing patterns, you can then predict that the next time someone downloads that whitepaper, it’s probable that they’re interested in the same thing. So, the proactive approach would be to reach out to that prospect. By reaching out you open dialogue that may be able to speak to any hesitations or concerns they have. IMPORTANT NOTE: There’s a right way and a wrong way to reach out.

Wrong way – “I noticed you downloaded our whitepaper on CRM integration this afternoon. Let’s talk about it.”

Right Way – “Hi, I’m reaching out to local IT professionals about CRM integration. Is that something you’re pursuing at this time?”

Don’t act like a stalker (even if your sales process inadvertently makes you one). The right approach can make the sale without seeming creepy or overwhelming to the prospect.

2. Cross-Selling

What do you see when you make a purchase on a website like Amazon? As you go to check out, it shows you suggestions such as: People who bought this TV also bought this HD cable and these DVDs. They’re using analytic data to figure out which companion products you’re most likely to need/want based on your purchase — it’s cross-selling. You can use predictive analytics in the same way.

Just as people who buy a TV often end up needing cables, people who buy one of your products or services may end up needing a companion product. Look back through transactional data from the last two years to see which products were most often purchased together. Then, use your predictive analytics strategy to plan the cross-sell strategy.

Pro Tip: Focus on your current customers first when implementing a cross-sell strategy. First of all, you have more data available to you about their buying habits. Second, research has shown that cross-sells to existing customers are 60-70% more likely to be successful than those aimed at new customers.

Your CRM system contains much of your key customer data – so use it! If your social media feeds are integrated into your CRM data, that’s even better! Use pattern analysis of Tweets, Shares, Likes, etc. for deeper insights. Multichannel analysis!

frequently bought
Amazon uses cross-selling to suggest products

You can start implementing your cross-selling data by:

  • Showcasing companion products together on your website
  • Creating a bundled offer for two of your most popular products or services
  • Automating cross-selling in a confirmation email (Think Amazon’s “Frequently bought together” tactic)

3. Lack of Action

The first two predictive analytics we covered have to do with interpreting the data around a customers’ actions. But, have you thought about the uses of predictive analytics for the opposite? If a customer you have frequent interactions with suddenly wanes in their communication, there may be a problem. Perhaps they’re unsatisfied with something, or worse, they are thinking of severing business ties. Best case scenario, you realize their lack of action early enough to reach out to them and save the relationship, but you can’t know to do that without first realizing there’s a problem.

So, how do you use lack of action to your advantage?

Look back on the customers who have left you. Can you pull out any patterns or trends? Use those patterns to create a fall-off model. For example, if you notice customers tend to respond slower (or not at all) to your emails prior to dropping your business, you can factor that in to your fall-off model. Think of your fall-off model as the ying to your analytics yang. When you combine the patterns from fall-off with your sequencing data, you can start to predict whether someone is likely or unlikely to buy. If they’re more likely to buy, the marketing and sales departments can reach out to encourage the final push.

As the trend towards predictive analytics continue, the activity management and sales forecasting of organizations will evolve to match. As humans, we are creatures of habit. Our behavior patterns tend to remain steady, and if you learn to read those patterns, you can vastly improve the effects of your sales and marketing efforts. Many CRM systems are starting to offer predictive analytics capabilities to keep up with this trend. Infor CRM, Salesforce, and Microsoft have all introduced predictive analytics in their latest releases, and another major CRM player, SugarCRM has one in the works. When it comes to predictive analytics, there’s no time like the present! Use the power of your information for smarter sales and marketing, starting now.

Looking to improve your sales and marketing efficiencies? Technology Advisors offers a wide variety of business software solutions to help you achieve your goals. Contact us for more information.

Danine Pontarelli's picture
Danine Pontarelli
Director of Marketing

Danine is the Director of Marketing for Technology Advisors Inc. She spearheads TAI events, marketing campaigns, and social media efforts. Prior to her work at TAI, Danine was a copywriter in the B2B publishing industry. Her interests include blockbuster disaster movies, tank tops in an array of colors, used book stores, Clint Eastwood, and being surrounded by trees. 

Related Articles

July 22, 2019

Moreover, social networks with algorithmic feeds, such as Facebook and Twitter, have been prioritizing visual content over text-only content, which in turn earns image-based posts higher engagement. If you’re a customer-focused brand, it’s worth investing into a social media monitoring system with sophisticated image recognition capabilities to avoid missing out on that context. This can help you assess a potentially problematic situation quickly and efficiently, and even catch wind of customer pain points before they reach out with their concerns. Awareness of visual mentions can help you collect customer feedback, address service gaps, and improve existing products or services based on their real-life use cases.

June 6, 2019

Before coming to work for my current employer, I held a writing position at another company. I stayed there for five years, and when I finally decided to move on, it was tough. My coworkers were fantastic people. They worked hard; they were silly and supportive; I’ll never forget them. But, as dedicated as they were, and as much as we all enjoyed working together, there was a general exasperation about the conditions of the workplace.

The pay wasn’t great, but that was hardly the issue.