How to create a customer profile (find the best prospects)

15 Sep 2016  |  by Ray Kirk

Customer profiling can help you find your best prospects - find out how.

Understanding the characteristics or profile, of your current customers and your strengths and weaknesses across market sectors is a prerequisite for any data-driven marketing activity.

Regardless of whether the goal is to expand to new sectors or up-sell in existing sectors it is necessary to consider your current sector performance and the remaining market potential before embarking on list acquisition and expensive marketing activities.

What is customer profiling?

Customer profiling compares your current customer base against your prospect marketing database, or a data “universe” provided by a supplier of business or consumer information.

A profile report highlights market sectors in which your business is over, or underperforming in relation to the business or consumer population. This information can be used directly to guide marketing strategy and, in particular, prospect list selection.

Until recently profile reports have required bespoke consultancy and hours of processing. It is now possible to produce profiles on millions of records in just a few seconds.   Most importantly this opportunity is accessible to marketers without a background in statistics. Customer profiling is not limited to detecting simple customer characteristics. But rather, it enables any segment of the data to be analysed against any background. For example the customers for a particular product can be analysed against the applicable industry sector or the set of prospects that have been offered the product.

Building a profile

There are three basic steps to building  a profile report:

1. Identify your target customers as the analysis selection.

These are the particular customers or businesses whose distinguishing characteristics you are interested in identifying, and the backdrop population that you are profiling against. This is known as the Base Selection and could potentially be the whole database.

How to build a customer profile

2. Chose the variables to profile.

These are the customer characteristics that you wish to explore and which will enable you to identify your target customers.

Chosing the variables for building a customer profile

3. Build the profile and interpret the profile report.

This profile report (created using the FastStats Discoverer Modelling module) has a collapsible section for each included variable with a row for each category.

For each category we can read the number and percentage of people out of the Analysis Selection (the “customers”) and the Base Selection (the “universe”).

The remaining columns show the market penetration: an index of over 100 shows overrepresentation in the category; under 100 shows underrepresentation and the z-score indicates the statistical significance.

Building and interpreting the customer profile

Market Penetration

The Index determines the size and direction of the Penetration graphic and the z-score; the colour from yellow (low significance) to red (high significance).

Market penetration

In this example you can see that your customers are overrepresented in the 30-40k and 40-50k income brackets but under-represented in higher income bands relative to the base population.

The information given by a profile report may confirm or add to your knowledge of your market sector – often both!

By studying the market sectors in which you are under and overrepresented you can develop marketing strategies to either concentrate in your key markets and/or improve the weak areas.

Market Potential

It may be the case that market sectors in which you are overrepresented offer little further scope for opportunity. This would be the case if the market is oversaturated, or is a small niche market. The Market Potential graphic helps identify the market sectors offering the largest opportunities.

Overview of market potential

Specifying Business Questions

Remember that the Analysis Selection specifies the target segment to be profiled and the Base Selection defines the underlying base population. The Analysis and Base Selections can be set to any selection of records to give you the flexibility to investigate real business questions.

Often the Base Selection for your profile will be a reduced subset of the universe database. The Base Selection might be defined to cover the potential market sector e.g. your product or offer is only applicable to limited companies, consumers with a driving licence, current account holders etc.

Similarly the Analysis Selection may be people who contacted the call centre, customers who have made a second purchase, customers who have not upgraded and so on. In many cases the Analysis Selection will be defined by a set of unique reference numbers (e.g. account numbers) from operational systems.

Now that you are able to build an accurate profile of your customers your marketing activities can be honed to target them more effectively. But this is just the beginning, an effective data strategy across the board can really boost the results of your marketing.

Discover how data can improve your marketing, download Why You Cannot Have an Effective Marketing Strategy, Without a Data Strategy now.

Ray Kirk

Consultant Developer

Ray is leading the development of AI techniques within Apteco. He is extending and automating the statistical techniques within FastStats® which he developed in his early days at Apteco. This “Apteco Intelligence” will then drive campaign optimisation in PeopleStage and deliver insight in the Orbit online platform.

Subscribe to our blog and get all the latest data analysis and campaign automation news.