Marketing maturity from the operational to the strategic grows in different ways within organizations. My perspective often helps accelerate that maturity via data driven approaches, letting the CMOs drive the organization aspects.

Marketing maturity can be viewed as a continuum. Just be careful not to expect that continuum to be absolute. No two organizations will follow identical maturity paths. Facebook’s maturity will differ from Pepsi’s, but along the way they will each benefit from growing capabilities like we discuss.

Most accept today’s maturity necessitates use of data, both internal data and that harvested externally either through collection or procurement. Our discussion then agrees data driven marketing is the scope we are covering.

Testing: Marketing’s Use of the Scientific Method

A/B Testing – If you send one customer advertisement A and another customer advertisement B, then compare the efficacy of those advertisements, you are conducting A/B Testing. You are comparing one marketing activity side-by-side with another.
Multi-variate Testing – Testing can go well beyond simply testing one thing versus another. Assessing multiple variables simultaneously, possibly customer segmentation, advertisement configuration, engagement cadence, and on requires maturity. What is important for this approach? Declaring the intent, tracking the data, evaluating the success in attaining the intent.

Consider an example. A car dealership makes odometer readings for customers when they arrive for service. Marketing can occur based on something as simple as a customer’s miles driven - no external data needed.

A/B Testing could include focusing only on the timing of sending a coupon for an oil change. Predicting within two weeks when a customer will start getting “change oil” lights on their dashboard could mean the difference between the customer going to the dealership versus the customer patronizing a discount oil change retailer. Testing sending an oil change coupon once a month versus sending specific to the forecast timing is the A/B Testing. The testing could include offer timing, offer differences (10% discount vs. 20%), etc. Likewise, moving along to multi-variate enriches the options to test.

Churn and Uplift Modeling: Moving from Predicting to Predicting by Treatment

Churn - Predicting churn makes sense. You want to know when you expect a customer to disengage.
Uplift - Uplift goes beyond simple Churn models. It predicts churn, including the variables of the marketing activity used. Uplift also gives feedback about the optimal treatment to use by situation. Moving from simple churn analysis to uplift models uses different data science techniques to build the understanding of the customer.

Marketing here shows growth in maturity, moving from testing to predicting. Because the testing processes are foundational, prediction models can be built incorporating testing. How can the dealership keep the customer coming in for oil changes once the warranty (read free oil changes) runs out? Uplift prediction viewed through different marketing treatments will help bridge the gap retaining customers longer along their journey.

Offer Optimization: The Refinements Begin

Offer Variation - Sticking with our oil change example, should you provide the same offers across all customer segments? The science would tell us no – different offers will have different efficacy.
Optimization - Offers align with the correct set of variables, including segment, timing, marketing medium, customer journey point, etc.
Maturity - Offer optimization builds on uplift modeling and testing.

One customer might select a “Save $20 on your next oil change” coupon, while another might select “Save $50 on your 90,000-mile checkup.”

Next Best Action: Understanding Your Customer’s Journey

Beyond Offers - It is one thing to change your offers (one coupon versus another). However, marketing is much richer than just coupons. There are so many marketing activities in which to engage with your customer. One next best action might be “Will you provide a review for us?” Some customers will do just that for you. Whether a customer does write a review for you or not is also valuable information. It can give you feedback on customer engagement, customer segmentation, customer journey, and other insights.
One Level Up - Next Best Action sits above your offer optimization infrastructure seeing across all marketing activities. It could differentiate based on medium used. A next best touchpoint could change from an email to a text message. It could mean a customer’s Instagram feed includes an ad for your product. It could mean so many different next best actions.
Cross-functional - One customer might get an oil change coupon, another customer might get “Ready to trade in for a new car?” offer. The action might change functions completely – moving from service to sales.
Customer Journey - The better you understand the customer journey, the better you can be in all your marketing, but Next Best Action benefits from knowing that journey.

Marketing Mix Modeling: Optimizing Marketing Expenditures

Spend - Yes, there are unlimited ways to market, which are the most effective?
Testing - Should your budget go to radio ads, emails, social media ads? Yes, you can test these areas, too. Marketing mix models should give you good insight, but you can test those predictions, too.

Marketing Automation: Algorithmic Marketing

Automation - Marketing maturity necessitates automation.
Multi-Variate - Automation should be just as multi-variate as testing. Marketing should be automated by schedule, by customer segment, by platform, by marketing activity and medium, etc.
Wide Application - Automating Next Best Action should include automating actions taken, data collected, testing, tracking, model updates – iterating the entire optimizing of Next Best Actions.
Customer Journey Alignment - Automation should drive and align with the customer experience.

Automation should be viewed as a means of building maturity. Some industries have customer base sizes of a critical mass that necessitate automation in its segmentation approaches. But even if you could do the segmentation by hand, automating it can still provide benefits.

AI: Artificial Intelligence Enters the Chat

Adding AI - Automation maturity will integrate AI agents.
Maturity - Applying AI to all aspects of marketing indicates a very mature marketing organization.
Applying AI - AI applications in marketing can include automated predictions, automating operations in providing marketing activities, automating understanding the customer journey, automating the testing processes, automating sentiment analysis, etc. The only thing currently limiting AI application to marketing is imagination.

Marketing will never lose its artistic touches. Beautiful designs will never go out of style. But adding science to marketing has helped it improve in ways never imagined just a few short decades ago. Marketing maturity includes moving along an organizational continuum towards more automation, better data analytics, and clear understanding of the customer journey. All the above is not an exhaustive list of marketing activities. We hit a lot of high notes, but remember, your organization’s marketing journey is unique.

Chris Ford is a data scientist who loves working with organizations. Whether he is working in healthcare, finance, higher education, or any other industry, his passion is finding the correct levers to pull in the correct situations. He loves working along the reporting continuum of operational to tactical to strategic with stakeholders and navigating all the implications those differing views present.