Become Strategic with Customer Segmentation and Lift KPIs

The idea that different groups of people want different things within a product, category and vertical is as old as commerce itself. According to the Association of Consumer Research, customer segmentation started in ancient Babylonia. By the 19th century Germany book publishers would only publish new children’s books in late Autumn to be purchased as Christmas gifts. Scholars, lawyers, doctors, clergy, women, Freemasons, and teachers were each important segments for the book trade. In fact, the only book published intended for the general market was the Bible.

By the 20th century the idea that product development and marketing should be tailored to different groups of customers became a foundation of business strategy, marketing and success. From Ford to Walmart to McDonalds every retailer, product and brand conceived products and executed marketing based on differences, similarities and affinities of customers. Many of the largest brands in the world started by serving specific segments within a vertical. Nike for running. Apple for home computers, Books for Amazon. Coca-Cola for brain health!

It is safe to say, customer segmentation should be a top-level strategic objective of every company and every product in-market. 

Segmentation is about knowing your customers but it is also ultimately about two things. Insights and matching. As different needs and use cases of consumers are understood, the more successful a brand will be meeting them and the more they will sell. The end result of segmentation should always be incremental outcomes. 

It’s here where data and marketing technology has mostly failed us as marketers. Our tools have tabs for “segments” but segments are not the product. Segmentation is a feature of many different solutions for data collection, organization, collaboration and activation. Segmentation has somehow become a second class citizen. And tied to a tool segments are strictly tactical. This is a huge problem for marketers that needs solving.

Customer segmentation should be foundational for everything. Products, pricing, promotions, markets. And while segmentation is absolutely helpful for tactical purposes, but it needs to be the starting point for strategy (obligatory Sun Tzu quote) and a first-order solution for your marketing.

Let’s get back to insights and matching – the core of segmentation strategy and tactics that make the strategy successful. Segmentation is where relevance is understood, delivered and measured. Segmentation is where you optimize performance. Segments are where winning and losing customers occurs. Segmentation is your ROI. It is your margins. It is your LTV.

Treating customer segmentation as a first-order solution does a few very important things for marketers.

1. Looking at results through the lens of segments immediately surfaces opportunities to improve your KPIs. Segmentation = insights.

2. Looking at results through the lens of segments immediately informs who’s who. Segmentation = matching.

Why are your merchandising segments different from your product segments? Why are your product segments different from your marketing segments? Why are your social segments different from your loyalty segments? Why are email segments different from site targeting? 

The answer to those questions is that point-solutions capture data and measure. They capture data and measure because that is how they lock-in as a vendor. It’s not about what’s best for you as a marketer. It’s about what’s best for them as a software company. 

The other issue is that sadly, tools like Google Analytics and SQL have trained a generation of marketers to be completely tactical. This means operators focus on measurement as the starting point of customer strategy rather than the outcome of it. How many times have you heard someone say “we have to improve our conversion rate.” You almost never hear someone say “we need to acquire different customers.”

Outcomes, good and bad, are the very last data point you should collect on a customer. Trying to optimize bottom-up is incredibly difficult. You are starting with the smallest amount of data and the smallest group of customers. It makes no sense to begin customer segmentation here. The data and measurements you collect should be seen through the lens of affinity based customer segments, not the opposite.

What you want to know is the context, the “why” and “where” that outcome occurred (and didn’t occur) and to whom. You need the affinity of the outcome, not a list of everyone that had it. We tend to gravitate toward “what happened” because it is a lot easier than trying to figure out “why it happened” but that is a product failure.

MarTech has failed to deliver on its promise because “added to cart” + “not purchased in the last 30 days” is a shitty segment. Much better would be to overlay that group of customers on top of affinity based customer segments. Then you can understand who these customers that “added to cart” + “not purchased in the last 30 days” actually are. Who in this group has bought from us before? Who has not? Who that bought before is in our loyalty program? Who is not? Who in this group came from social media last visit? Who came direct?

This is the context marketers need to be successful.

Knowing the answers to these questions with your customer segmentation means sending multiple relevant marketing offers rather than one irrelevant offer. This is the whole point of segmentation – relevance! Why are some experiences performing better than others? Are we incorporating learning back into tactics? Does it make sense to generate 100 experiences for our customers or is 10 the optimal number? 

When segmentation is unbundled and is a product unto itself the answers to these questions become obvious. Opportunities and places to start improving KPIs move from a blank page to being obvious. The optimizations you do start a flywheel. Maybe most importantly you move with more velocity and have higher impact with your optimization. Faster campaign ideation. Faster tactical determinations. Faster performance lifts. Personalization at scale is really segmentation at scale. 

Because we can now measure everything we have carved up our customers into segments that have become meaningless. Both to them and to us. When every part of the business has their own segments, segmentation is strictly tactical. We are not operating a segment based strategy for the business. This means you do not really know your customers.

At Neuralift AI we’ve built an application that begins and ends with segmentation because clearly segmentation deserves its own product. An app that is first-order for strategic insights and tactical relevance. An application for outcomes not measurement.

Reach out to see for yourself how transformational really knowing who your customers are can be.