Tech4T - Segmentation, profiling, scoring and statistics.  +44 (0)1733 890790

   

 

 

Want to understand what differentiates one group of customers from another?

Or responders from non-responders?

Or buyers of product 'Y' from buyers of product 'X'?

Or active customers from those likely to lapse (churn) ...

The answer can be found by profiling, segmenting and scoring your customers and prospects

What are popular statistics measures?  Click here

What is segmentation?

People or organisations with one or more characteristics that cause them to demand similar product and/or services - based on the qualities of those products such as price or function, can be grouped into what is termed a segment. A true market segment is distinct from other segments (different segments have different needs), is homogeneous within the segment (exhibits common needs), responds similarly to a market stimulus, and can be reached by a market intervention. The term is also used when consumers with identical product and/or service needs are divided up into groups so they can be charged different amounts.

Segmentation chartPeople in a given segment are supposed to be similar in terms of criteria by which they are segmented, and different from other segments in terms of these criteria. These can broadly be viewed as 'positive' and 'negative' applications of the same idea, splitting up the market into smaller groups.   

Segmentation is a form of critical evaluation rather than a prescribed process or system, and hence no two markets are defined and segmented in the same way. However there are a number of foundation criteria that assist segmentation:

  • Is the segment viable and can you make a profit from it?

  • Is the segment accessible?

  • How easy is it to get into the segment?

  • Is the segment measurable?

  • Is realistic data obtainable to consider its potential?

The are many ways that a segment can be considered, and some of the basics include:

  • by geography - such as where in the world was the product or service bought.

  • by psychographics - such as lifestyle or beliefs.

  • by socio-cultural factors - such as class.

  • by demography - such as age, sex..., and for business, size, turnover, industry type..., and so on.

A company should evaluate each segment based upon potential business success. Opportunities will depend upon factors such as the potential growth of the segment, the state of competition within the segment, how much profit the segment will deliver, how large the segment is, and how the segment fits with the current direction of the company and its vision.  

Customer profilingCustomer Profile chart

Profiling (and segmentation) are needed to target marketing campaigns, evaluate market potential, develop appropriately balanced sales and franchise territory areas, identify where to locate new outlets, identify hot spots to target sales effort, and to help you tailor your communications in terms of improved relevancy to increase sales. 

Profiling your customers in essence refers to identifying the characteristics that differentiate your customers from people who are unlikely to ever buy your product or have need of your service. It goes deeper though to differentiate your best and most loyal customers – those that buy more frequently, place higher value orders and are open to other offers – from the rest. Not all customers are ideal - some will purchase infrequently, won't spend a great deal of money or will have little loyalty, yet will take a lot of your selling time.

One approach to profiling is to first segment - or group - customers by their value to the business, then append external profile data – census, lifestyle, market data or business/consumer geo-demographics – to enrich segmentation and customer understanding. To do this we first statistically analyse customer trading data to determine how much people spend, how often they buy, what they buy, where they live, their life stage, what attracted them to your company, etc., then compute a score to identify the ‘best’ type of customer. 

Using the Postcode (or ZIP code), we add external ‘profile’ data which allows us to accurately describe what your best customers look like.

Then, using geographic analysis techniques, we can position your existing customers on a map - classifying by value - and develop what is termed a ‘heat map’ to identify where geographically to find more of the same.

So, what does this information now allow you to do? 

For starters, if you are planning a product launch, moving into franchising or thinking of expanding into a new geographic area, by applying some mathematics and a ‘weighting’ to take account of known competition and the brand life stage, you can estimate what a specific geographic area may deliver in terms of sales potential.  Add to this your knowledge of the effort needed to sell your product or service and a realistic estimate of how many prospects (or customers) a sales team could cope with, and this information can be used as the basis to compute a set of ‘right-sized’ sales or franchise territories, balanced by opportunity and drive time.

Customer Profile MapTerritories created in this way will deliver economic benefits, prevent territory owners/sales teams being located too close to one another or located in unsuitable areas, and ensure each has sufficient territory to generate a revenue stream without undue competition. If you operate a location based business model, then choosing the right site to open your outlet has as much to do with the demographic profile of your customers as it does the physical location of the outlet.

With accurate customer profiles and people or businesses grouped by segment, the right analytical resource, tools and data - including competitor locations, traffic patterns, parking, high target market density - you will be able to determine how many outlets or branches you can open in an existing market, where you should expand your network, which outlets should be closed or consolidated.

You will also be able to select the outlet, branch or other business locations that are profitable and contribute strongly to your overall network. Perhaps most importantly, customer profiles enable you to target the right people using the right media and channel to get across your marketing and sales messages.

What is scoring?

For those of you new to the concept of using scores for database segmentation, profiling and to then select individuals for targeting, the following example should help.

Most people have at some time applied for an insurance policy or credit of some kind - a loan, mortgage, finance for a new car, etc. - and during the application process have been asked questions such as age, length of residency, how many credit cards, health history, etc.

For each of the answers given, data will have been entered into a computer program that assigns a number (or score) to each piece of information. This is usually termed a score card. The way the numbers are then added together will depend on the formulae used, but will greatly influence the decision to give you what you are requesting.   

The important thing to remember is that whilst each separate piece of information may in its own right not be seen as important and might otherwise be ignored, only when these items are combined does the resulting information and its score become significant and predictive.  

A predictive model would usually be created by a statistician and used to compare the characteristics (or profile) of individuals who are known to represent, for example, a good vs. bad risk. This produces a formulae and computes a value (or score) for each individual. Depending on how high or low the final score is, the lender or insurer can determine the most appropriate offer for that individual and be fairly confident in the degree of risk.  

The picture to the right shows an example of how important various pieces of information are in predicting who might respond to a specific type of mailing or email campaign.  

Direct marketing - using scores to target prospects

In direct marketing, to find the most likely responders for your offer could well involve making hundreds of separate or combined counts and selections, and then only using only the information you believe is important - i.e. maybe missing key predictors!

However, by using a scoring process, the selection task can be greatly simplified. Each individual can be allocated a single score based on their propensity to take the kind of action you desire. 

There will never be a single model or formulae that meets all your targeting or profiling needs however.

A separate model and scoring formulae may need to be created to optimise the targeting and response to each different offer you make, or to achieve the results you desire. And remember, customer needs' change over time as do the activities of your competitors. What works well today may not work so well in the future, so the art is to continually test, learn and refine to get the most from your marketing spend!

Profiling - using tools such as FastStats to simplify the process

Profiling in market targeting is used to compare the distribution of a group of  individuals - based on their purchasing, behaviour, etc., characteristics - relative to other individuals on your customer or prospect database.

Once a profile has been created, a marketer can then build predictive models based on the results of that profile.

Predictive models are used to score and segment your marketing database to, for example, find prospects that ‘look like’ your existing high value customers. This process helps you test and purchase only the most responsive lists.

The model reports to the right illustrate potential gains, revenues and profits and has been created using FastStats

Applications

Typical applications for scoring and profiling include:
  • Selection of look-alike prospects (i.e. prospects who match the profile of selected customers)

  • Using scoring to grade and select the best prospects from third party lists. See also profiling with TRAC geo-demographic data

  • Using scoring to grade prospects to prioritise contact strategies by mail, e-mail, telemarketing or face to face

  • Using different analysis selections, base marketing priorities on: financial risk or likelihood to churn (lapse or defect), respond, purchase, upgrade, etc.

  • Profiling responders against the base of all those marketed to in a test campaign, and score prospects in order to prioritise who to select for the roll-out campaign

  • Developing sales and franchise territories, and identifying where best to position outlets

The Statistics

When it comes to profiling and segmentation, many of the ways this information is described in the first instance is in statistical terms, and later these are translated in a more human speak.  To provide guidance as to some of the more popular statistics, and what each means, definitions are as follows:

Data set - A data set (or dataset) is a collection of data, usually presented in a table format. Each column represents a particular variable (field). Each row corresponds to a given piece of information - a number, piece of text, etc.

Mean - this is the average, or central tendency of a data set and is a measure of the "middle" value of the data set. All you have to do is add up (sum) all the values in a set of data and then divide that sum by the number of values in the dataset.  Mean

Mode - This is the value in a set that occurs the most often. Mode

Range - The range is the spread of numbers in a data set, the minimum to the maximum.  Range

Median - This is the number that is in the middle of a range. Median

Outliers - Extreme values that will skew the analysis

More information about descriptive statistics on Wikipedia here

How Tech4T can help you

Developing models, profiles and scores typically requires extensive analytical or consultancy projects, but Tech4T have the technology and skills to greatly simplify the task.

One of the age old problems of customer analysis, segmentation and predictive (or response) modeling is being able to apply the results back to your marketing database to improve targeting and hence grow customer value. Another challenge is to make sure that all relevant information is used in the analysis process - even where data is scattered across different systems and locations.

Fortunately we have an ideal solution.

We can merge and consolidate your data which can be supplied to us in almost any format, undertake the customer value, market potential, segmentation and profiling analysis, present the results, and importantly append the outcome to your database as a set of 'easy to use' indicators which can then be used to make targeting selections.  We can also provide an on-line 'Targeting Workbench' where you can conduct your own analysis and then turn the results into new strategic and tactical direct mail and e-mail campaigns.


  Targeting Workbench        Step-by-Step Analysis Guide         Analysis Services        Geographic Mapping