Customer segmentation

We are not all the same and neither are our customers! The scatter gun / one size fits all approach to marketing does not work. Just like us, people want to receive interesting, relevant and engaging communications. Successful marketing requires that you understand who your customers are. What do they like to buy, how often do the make a purchase. From a business perspective you also need to know the value of your customers. This requires customer segmentation. So what's a customer segment? One of the easiest definitions is "a group of customers with shared needs" (Nigel Temple - Marketing Consultant). From this definition, it's clear what you need to do; identify customers with shared needs. Where do you start? A few simple reports run across your customer data may be quite revealing: How many customers do you have? When did you acquire those customers? If consumer, which age groups or gender are most prevalent If businesses, which market sector do they operate? Where are they located? All over the country or close to you?

Armed with this general view of your customers, the next stage is usually to see how they perform. Don't worry if you don't have all the information above , external lifestyle or business demographics can be sourced to supplement your in-house data.

The classic method for measuring customer performance is to build a recency, frequency and value model. Within the model you score customers based on their purchasing habits (RFV - also referred to as RFM, where 'M' is for monetary value).

The RFV model comprises scores base on:

• How recently they each customer placed an order
• How many times a customer ordered on average within the last ‘n’ months
• Average customer order value e.g. gross margin

Recency, Frequency and Value

If your technically minded, you might like to have a go using the following approach. If you'd like to see the results, but need help with the segmentation, then call David Willis directly on: 01494 871342 or send him an email: david.willis@information-drivers.com to discuss how this could be achieved.

To build an RFV model, run a query on your database to identify the following information for each customer:

• Date of first and last order
• Number of orders
• Total gross margin (or other value such as sales - take cost into consideration)

Using this information, calculate the RFV quotients for each customer:

• RECENCY: Time since last order is current date minus date of last order (Days)
• FREQUENCY: Duration of custom: current date minus date of first order (Days) divided by number of orders
• VALUE: Total gross margin divided by number of orders

These give you absolute values for each customer. The next stage is to categorise each of these quotients. You will need to determine value ranges for each category.

Taking each element in turn, the objective is to achieve categories that have broadly speaking similar numbers of customers in each. The number of categories will vary depending on your business.

Having identified the categories, each customer record is scored base on the raw quotients e.g.:

RECENCY:

• 1 to 90 days - score 3
• 91 to 180 days - score 2
• 181 to 365 days - score 1
• Greater than 365 days - score 0

FREQUENCY:

• At least once per month - score 3
• At least once in two to three months - score 2
• At least once in four to six months - score 3
• Once time only purchase - score 0

VALUE :

Clearly the scoring depends on the business in question. If Gross margin is not readily identifiable, then the sales value could be used, but consideration should be given to the distribution of margin across the products and services your organisation sells.

Assuming a similar ranges of scores are applied for value then your best customers would be those with scores of 333 (highest worth) down to 000 (lowest worth).

Customer segmentation results

Your results could look something like these. Note the observations made next to the results.

Using this information you can formulate interesting, relevant and engaging propositions.

Managing direct marketing performance

Not only are customer different from each other, they also change their buying habits. It is best to develop an on-going scoring process to manage the effect of direct marketing.

Campaign examples:

• Are all high value customers regular clients? What was the effect of the campaign to encourage more orders?
• Are frequent low value purchasers harming your margin? What was the effect of encouraging less but bigger orders?
• Running a campaign to encourage "One time only" customers to shop again? Did it have the desired effect?
• Did the customer discounting policy encourage more frequent shopping? How did this effect the overall value of customers?

Once you have become familiar with the RFV model, it can be extended to include a time element. The time element will provide for comparison between different periods.

Applying customer segmentation

This customer segmentation approach, will help you gain insight into your customers shared needs e.g.:

• Identify opportunities to do further business
• Improve marketing communications and promotions
• Address potential churn

The results can be applied to:

• Customer Relationship Management, to direct customer service representatives
• Websites, to manage on-line propositions
• Campaign management, to direct targeted propositions

Following this segmentation will help you understand your best customers, sell more to the right people and retain more profitable customers. Would that be of interest? If that's a yes, call David Willis directly on: 01494 871342 or send him an email: david.willis@information-drivers.com to discuss how customer segmentation could be applied to your business.