Inside Strategic Segmentation Blog Series – Part One
Aligning Your Segmentation Approach with Strategic Priorities
Written by Chris Diener, Senior Vice President, Analytics
As marketers, we’ve all been there. Under pressure to drive long-term sales growth, marketing has been asked to completely rethink its customer segmentation strategy and build a brand that can serve its most valuable customers more effectively and efficiently. You may even need to restructure your entire organization based on it. In fact, one of my past clients, a large telecommunications firm, did exactly that. They formed divisions around the segments.
Large-scale segmentation projects come with a host of challenges. Success depends on your careful guidance and ability to manage expectations and create alignment with key stakeholders across the company.
Based on my experience as an advanced marketing analytics expert with more than 100 segmentation projects across various industries and virtually all usage contexts, this is what I’ve seen: the power that strategic market segmentation can have to align teams and drive rapid customer growth and the mistakes that can make it infamous.
In this blog series, we’ll discuss the key elements to design and implement a successful segmentation project, key issues to watch out for, as well as some effective remedies.
Blog series at a glance:
- Part 1 – Aligning your segmentation approach with strategic priorities (this post)
- Part 2 – The Importance of making segmentation immediately actionable
- Part 3 – Strawman planning and hypothesis generation to maximize segmentation impact
- Part 4 – The ability to reach the segments differentially or accurately
- Part 5 – Prioritizing action steps within each segment: TURF or Next Best Move
Before you can build a successful segmentation model, you first need to agree on the goals of the project and ensure commitment across the organization to put your most valuable (future) customers at the center. Ultimately your target segments should define your entire brand strategy from the products and services you bring to market to how you approach marketing to maximize CLV.
In addition to focusing on the right segments, you also need to identify the right information on each segment to inform decisions across the organization. This is why the prioritization of strategic needs is so important.
Market segmentation is more than just a cluster analysis of like-minded respondents. It’s about uncovering insights on each target segment and then applying that data to predict and improve outcomes across the business. Which products, pricing, and distribution model will resonate most with each of your segments? What is the right messaging and communications strategy?
Without a clear understanding of the strategic imperatives, and the underlying questions, it won’t be until the end that you realize that your boat has just entered the wrong destination harbor – and you can’t go back. What does this mean? Let’s go up to 20,000 feet for this one.
Figure 1 below shows two strategic dimensions and how they align with the segmentation approach.
Figure 1 – How strategic imperative guides segmentation approach
The Axes – Strategic Needs
Let’s first understand the dimensions or axes. The vertical axis is “Subjective Focus”. If the primary goal of your segmentation is to inform product development decisions, then your focus will be near the top of the axes. If your product or service offering is mostly set, then your primary focus will be communications or positioning – i.e., reaching the right people with the right message.
As an example, we had a client in the automotive space do a segmentation study that focused purely on informing product decisions (i.e. which features or upgrades should be included in each future model). In contrast to this, a project for another auto OEM focused on how to best communicate around their existing product line and brand to appeal to different segments.
The horizontal axis is all about what I term “Implementation Priority”, which refers to types of targeting. Usually, the mandate with segmentation is to identify differences between segments to increase effectiveness with specific products or communications. This is what we call “Insight” or “Subjective Targetability”. The other implementation priority is “Reach” or “Objective Targetability.” Given that we have different sets of products, pricing, and communications, how can we reach each target segment differentially with the right combination of those products, prices, or messages?
Depending on the underlying needs of your target segments, the strategic areas to focus on might differ. For example, if we are looking at a line of home improvement products and the objective reach accomplished through retail distribution is similar or the same for all of the products in the line, then the priority is determining the subjective reach. In other words, how can the products be positioned and priced differently based on particular consumer issues, triggers, and beliefs?
By contrast, in a project for a large specialty retailer, our goal was to maximize reachability with mass marketing. This required clear differentiation between segments across both demographic and geo-psychographic elements used by their marketing partners.
The Space - Methods
Depending on your strategic goals, your segmentation approach will differ. By using the chart to identify the method, you can better map how you will implement your segmentation based on which quadrant you’re in.
Most organizations focus only on attitudinal (bottom left) or database (top right) segmentation, the most common approach to segmentation. Attitudinal segmentation uses survey research and is probably the most versatile because you can decide what attitudinal data to collect and how to best quantify those measures. The tradeoff with this approach, however, is constructing a method to reach consumers in each segment can be more difficult with limited attitudinal data available.
Database segmentation relates observed characteristics (demographic/firmographic, behavioral/transactional), to underlying customer needs. The obvious benefit is that these segments can be easily understood and reached, but with limited information, strong hypotheses are critically important. Many organizations begin with database-based segmentation and then feel the need to embark on an attitudinal segmentation.
Depending on your strategic priority, however, a more specialized approach may be needed. For example, if informing product decisions is the core focus of your segmentation, you are in the top-left quadrant of the chart and likely need to apply a series of trade-off models and simulations. This would allow you to understand how to segment the market based on product or service feature differences and configurations to maximize demand across the market.
In comparing segmentation by optimization (top left) with attitudinal segmentation (bottom left), you don’t know how your segments differ by product or service, nor if they would differ in a way that is the best to minimize cannibalization between segments. Trade-off modeling and simulator-based optimization allow you to define which segments are most likely to buy each of the products or services in the mix that result from the optimization.
By moving across the horizontal axis to the bottom right, making Objective Targeting or Reach a priority, we can “stretch” an attitudinal segmentation to greatly increase its reach, while not abandoning the fundamental need to have subjectively targetable segments. There are numerous methodologies that accomplish this goal.
They include methods such as “pre-analysis” which selects attitudinal measures that already have a propensity to differentiate on the reach elements. Other approaches include “Reverse Segmentation,”, Latent Class Segmentation with Covariates, Canonical Segmentation, and Lookalike Segmentation. These are all “data fusion” approaches.
A particular data fusion approach we use is based on the NLM (Nascent Linkage Maximization) algorithm (Diener and Uldry, Fusing Data from Surveys to Databases: Segmenting For Effective Linkages, 2002, AMA Advanced Research Techniques Forum). This usually provides a 100% improvement in reach effectiveness while sacrificing perhaps 20%-30% of its subjective targetability.
The question is, what kind of segmentation has your organization done, and for what purposes? Are your applications aligned with your approach? Have you tried to make one segmentation fit all strategic uses? Have you extracted the most value from your efforts so far?
In the next post of this series, we will dive into best practices for each quadrant to maximize the strategic value of your segmentation project - informing key decisions and applications over time.