AI, startup hacks, and engineering miracles from your friends at Faraday

Predicting customer lifecycle outcomes with data analysis and machine learning

Perry McDermott on

Predictive customer lifecycle marketing


This post is part 3 of our 4-part Customer Lifecycle Optimization series

Expanding on your customer lifecycle map

In our last article, Lifecycle mapping: uncovering rich, predictive data sources, we discussed the importance of mapping out your customer lifecycle to better understand where your most predictive customer data is hiding.

Lifecycle mapping is the first step to using artificial intelligence (AI) to optimize your customer lifecycle marketing initiatives.

Now, we'll pose some questions to help identify your predictive customer attributes and lifecycle events, pinpoint where that data is located, and recognize patterns to predict outcomes for future prospects, leads, and customers.

Step 2: Data discovery

Data discovery is the second stage in the customer lifecycle optimization (CLO) process. The primary task of this stage is to expand on your lifecycle map to identify authoritative data sources that establish progress.

As a reminder, your customer lifecycle map should look something like this:

<a href="
Customer lifecycle mapping tool

At each stage, prospects, leads, and customers will complete certain events that will individually or collectively trigger a transition in or out of that stage. Individuals will also have different attributes, which help determine whether or not they belong in any given stage.

Events typically emerge from data that represents behaviors and change. For example, your e-commerce system records purchases, your CRM records wins, and your ESP records email engagement. When individuals click a link in an email, sign up for a demo, or make a purchase, they're completing transition-triggering events.

Attributes exist in the data that describes the customer themselves. For example, your CRM will contain contact information, lead status, and possibly physical addresses. If you service a specific geographical market, a determinant attribute will likely be the individual's household address. If they live outside of your market, they will fail the litmus test.

Go through each cell of your customer lifecycle map and think about which events or attributes in your data could be used to trigger a transition or pass the litmus test. Think about where you can find that data.

Step 3: Predictive groundwork

Now that you've formalized your lifecycle and its representation in data, you can start recognizing patterns and eventually predict outcomes.

Imagine loading up your converted leads (customers in "retention and expansion") alongside your stale leads that never converted. What differences can you find?

This comparison analysis is especially effective when you've added depth and breadth to your existing data (see our previous article for more information on data depth and breadth).

As you experiment with this type of pattern recognition, you'll quickly realize that it's the kind of thing that computers do very well. That's where machine learning comes in handy.

Machine learning refers to a class of artificial intelligence techniques in which previous outcomes are algorithmically analyzed to uncover patterns, a process known as "training" a predictive model. The model can then be used to predict likely outcomes given new data by looking for indicative patterns.

What's next?

Once you've mapped out your customer lifecycle, identified where to find your most predictive data, and developed models to facilitate future predictions, you're ready to use those predictions to guide your outreach.

For a deeper dive into the four step customer lifecycle optimization process, download our free whitepaper, AI for the customer lifecycle: Making the most of your data.

Whitepaper — AI for the customer lifecycle