Boobook's guide to customer segmentation: How to define, understand and activate your audience
In one of our previous articles, we explained customer segmentation and its importance in optimizing your marketing strategy.
As this topic is quite extensive, we decided to dig deeper into it and tackle "how" to do a customer segmentation project. Over the last two decades, we honed our expertise by working with different international companies. Each project is different, but most clients ask us a critical question: "Which product to offer to which customer, why, how, and when?" Considering customer segmentation is one of our main services, we designed a workflow that we apply consistently with minor alterations depending on the client’s vision and the project size.
Before we take a closer look at our guide to customer segmentation, we’d like to share our checklist that needs to be revised throughout the process. These ingredients are mandatory if you want your project to succeed:
✓ Get all stakeholders involved & aligned
✓ Spend sufficient time on defining the purpose of segmentation
✓ Ensure actionability by integrating outcomes in strategic plans and linking to CRM
✓ Embed segments into company thinking by sharing insights in many different ways
✓ Make sure the segments differ on the who, what and why
The segmentation process we apply at boobook has ten stages plus one extra step that is recommended and mandatory if there are significant shifts in the industry and market:
1. Stakeholder workshop: Define the purpose and align
2. Qualitative research: Determine the right questions
3. Database deep-dive: Revise content, quality, availability
4. Collect and locate data: Conduct an effective research
5. Explore segment solutions: Finding the correct methodology through an iterative process
6. Decide on the final segment model: Choose the most suitable approach for your client (and your team)
7. Integration of model segments into the existing database: Setting up a prediction algorithm
8. Build personas: Fine-tune the segment groups
9. Validate personas: Put your results to test
10. Communicate and act
+
11. Revise and adapt when market dynamics significantly changed
How to set up, launch and deliver a customer segmentation project
1. Stakeholder workshop: Define the purpose and align
Most segmentation projects aim to create a custom framework used and applied by many departments, ensuring a specific guarantee for a consistent customer approach. As consultants, it's essential to get buy-in from all stakeholders - from the beginning of the project.
How
Ideally, we organize a workshop (or an online meeting) with a key stakeholder team (market intelligence, marketing, sales, product development, CRM stakeholders).
What
Typical topics we discuss at this early stage are:
- Expectations: Determine what successful segmentation means to the client
- Experience: Discuss previous experience with segmentation or research and examine the sector and similar products (who, what, why, how)
- Usage: Define who will use the outcome of the segmentation and how. Agree on deliverables to embed segmentation into strategic and everyday operations.
- Hypothesis: Specify use cases, key characteristics they would segment customers on, level of complexity, approach, etc.
- Challenges: Analyze any hurdles you might encounter (communication, resources, transparency, etc.) and agree on how to tackle them.
Things to consider:
- Define the purpose of the segmentation
- Develop the right products: What are the different product requirements in the market, and what needs do they fulfil
- Optimize marketing campaigns: Which marketing messages are needed to reach the customers of most interest
- Optimize marketing campaigns through individual targeting: Who to target, when, how, which product and which message
2. Qualitative research: Exploring the market
If there's insufficient knowledge of customer differentiators or the terminology is too complex or new, qualitative research should be the next step. Considering the qualitative research method focuses on obtaining data through open-ended and conversational communication, we can discover "what" people think and "why" they think so.
How
Organize focus groups, one-on-one interviews, in-home visits, join online communities, etc.
3. Database deep-dive: Revise content, quality, availability
At this stage, the main goal is to understand all the details of the internal database to make sure the segmentation is actionable.
How
Online meeting with Customer Insights / Market Intelligence team and database owners
Checklist:
Some typical questions to ask:
- Background on the database
- Explanation and links of the data fields currently available
- Who is in the database, and who isn’t
- How has it been used so far, by whom and for what purposes
- How has the data been captured, and the impact on the interpretation
- Can the data fields be interpreted in the broader sense
This discussion will also look forward to the current internal data to understand if any definitions/data fields/composites will likely change over time and if fields might be dropped.
A further key topic is a discussion about the data quality of some fields to highlight any that should be avoided when working towards actionable value segments.
4. Collect and locate data: Conduct an effective research
This is where all the nitty-gritty work gets done. To collect the data, you need to develop a plan detailing where each variable will be found and which resource and method will be used to find it. Ideally, several sources/customer views are available to create a 360° view of each segment.
How
To answer questions "who, what & when", we need to look at the behavioural and demographics data (sourced from transactional data, sales data, mobile data, and textual data).
To answer the question "why", we need to collect and analyze the motivational data (answers to attitudinal, motivational questions, self-reported).
Checklist
- WHO (who are they, demographically)
- WHAT (how do/will they behave)
- WHY (what drives their behaviour)
5. Explore segment solutions: Finding the correct methodology through an iterative process
There is no one way of grouping people into segments. Depending on the goal of the segmentation and the chosen segment differentiators (age, gender, attitudes, behaviour, etc.), customers can be segmented in many different ways. However, it's essential to review if the segmentation method will deliver sensible, relevant and valuable results that answer the who, what and why.
How
Methods to develop segmentation model: K-means, Latent Class, Ensemble, Decision tree (e.g. CHAID)
Methods to predict segment membership: Discriminant analysis, Multinomial logit (MNL), Random Forest, Deep learning
Once we select the methodology, we proceed with a traditional segmentation approach (when segments differ on a few dimensions) or cluster ensemble segmentation (when segments differ on many dimensions).
6. Decide on the final segment model: Choose the most suitable approach for your client (and your team)
In this phase, we organize a workshop with the client to present the approach and methodology. Several reasonable solutions are presented and discussed together with the stakeholder project team.
Checklist:
Typically, we judge them based on the following criteria:
- Apparent differences in key customer differentiators and profiling data: Segments can be easily visualized and remembered.
- Actionable: Clear actions for each segment and a consistent profile within a segment, sufficient differentiation between segments.
- Predictable: Segment membership can be quickly and accurately predicted, ideally using a small set of customer characteristics.
- The right size: Not very large, nor very small segments.
7. Integration of model segments into the existing database: Setting up a prediction algorithm
Once we analyzed the data, it's time to link the new segments to the client's CRM database. Modelling the segments into the database is an integral part of the segmentation project because we develop an algorithm to predict the new segments using the database variables.
This is an iterative process to increase the overall prediction level. To achieve that, the model, approaches and accuracies are discussed regularly.
How
There are a few critical steps in this phase:
- The prediction power of a segmentation model is checked during the segmentation modelling
- Once a solution is acceptable from a "making sense" perspective plus sufficient prediction power, we develop the algorithm
- The models will be built using a training set
8. Build personas: Fine-tune the segments
A customer persona (or a buyer persona) is a fictional character derived from customer research and segmentation. In this stage, we are ready to identify and develop personas that will help us understand customer needs and pain points. By creating personas, we help the client to imagine their target audience.
How
With basic information like age, gender, and occupation, we believe it's even more important to break down their habits, motivations, and desires. Why do they do certain things in a certain way, and how can we get their attention? Working with personas enables each department to become more empathetic and effective when working directly with customers and prospects.
9. Validate personas: Put your results to test
We can allocate a customer to the right segment through an algorithm and a shortlist of golden questions (self-selection). This is often used for further research with customers of specific segments (recruited using the algorithm) to validate them and get additional insights into them.
10. Communicate and act
At boobook, we believe creating a final presentation and delivering insights is just one part of the handover. The other, more crucial and more valuable phase is actually helping our clients integrate and apply these insights by sharing our recommendations and strategic ideas. Through extensive consultations, we ensure the client doesn’t only receive the results but also understands what to do next, meaning how to interpret the insights into actionable points. This last step is where we connect all the puzzle pieces so the company can implement the insights with confidence and relevant knowledge.
The analysis results should always be combined with insights, preferably presented to the stakeholders and encouraging them to follow our advice. The presentation is also a reference document to share the segmentation insights with other teams/departments and enable a more straightforward implementation of the customer segment strategy throughout the entire organization.
How
Share segment insights through workshops, visual materials, such as posters, showcards, summary decks and a designated online portal.
Checklist
- Developing a product
- Developing communication approach for each key customer segment, built into the business strategy/marketing plans
- Embedding segments into CRM
11. Revise and adapt when market dynamics significantly changed
Customer segmentation projects evolve and transform over time. Shifts in customer behaviour, mindset, and needs can change due to demographic trends, technological innovations, and economic fluctuation. We always advise our clients and partners to look at customer segmentation as an ongoing process. Segmentation should follow any changes in the market, and it should help you predict the changes.
Why choose boobook to help you with customer segmentation?
There are various customers within each business, with unique attitudes, buying patterns, pain points, and expectations. Your customers identify the value of your business through the way you communicate with them, offer support, and provide solutions to their needs. The key to creating and maintaining successful customer relationships is understanding them and predicting their behaviour. By conveying customer segmentation analysis, companies can meet and exceed the expectations of their customers.
We hope our guide will help you and your team understand customer segmentation better. As we continue the series on segmentation, we’ll present tools and the most used methods for market research.
If you need help or guidance with customer segmentation, feel free to reach out! Our team of experts will help you expand your knowledge about your customers and provide tailored solutions based on in-depth analysis and insights.
Never miss a story. Subscribe to our newsletter!
Related posts
Generational marketing: How to connect with Gen Z
We’re continuing the series on generational segmentation with Generation Y or Millennials, individuals born between 1977-1995.