What's the point of segmentation?

Whether it's broad reach or tightly targeted, you still need it.

Segmentation is a simple term that most people understand.... but...

The issue lies in how it’s used. This simple concept is based on the idea that you should break up your total market into groups, to better tailor your marketing to what each group wants. But the problem lies with the assumed implication that you have to break it up. That isn’t always the case.

Segmentation is only useful if there are noticeable and distinct differences within a market that can be capitalised on.

What is good segmentation?

A good example of segmentation is one that focuses on dividing a market based on its commercial appeal. This could be a segmentation that is as simple as ‘Small businesses struggling with financial admin that outsource their accounting within a 25-mile radius of our office’.

This segmentation states the potential customer, the issues their business is facing and the constraints in which the marketer must plan based on internal resources, in this case determined by geography.

This is a far cry from the common examples of differentiation and personas, where markets are subdivided into countless smaller groups with hyper-specific criteria around the individual traits of people. Generally loaded with information but little about the financial value or quantity, just vague ideas of who the customer is.

This information is neither useful, nor practical because the resources required to meet the exact criteria cannot realistically scale with a growing business.

The second nail is the mathematical reality of over-segmentation.

Here is an example:
1,000,000 customers make up the total market.
500,000 are female.
150,000 have blonde hair.
60,000 of these blonde-haired women have children.
40,000 of these blonde-haired women that have children work full time.
7,000 of them are classed as ‘passionate’ about homeware.

The further you segment, the smaller the total market gets, inevitably reducing the potential value of the market, in terms of both its cash value and the total quantity of customers.

A good example of how this should look is:
1,000,000 customers make up the total market.
400,000 are women within 20 miles of stores (70% of sales are in-store, 30% online).
250,000 have an income level within the range of our typical customers.

This is a far simpler segmentation that gives a far greater market to address. With this information you can then perform further work to understand the media they use as well as research into category entry points to understand their wider behaviour. This example does not seek to superficially group people based on character traits which have absolutely no influence on purchase behaviour.

Keep things simple.

The basics of segmentation in the classical sense are often all that are needed to define the parameters in advertising and media choices, rather than an over-reliance on behavioural signals from platforms such as Facebook.

Thanks for reading.

This piece was an excerpt from the much larger Anti-personalization: The best ad for one, is the best ad for all.

The people have spoken.

Thanks to the voters on the recent poll, the majority would clearly prefer ocassional and well thought out emails. Which is how I’ll keep it from now, sometimes you might see more - sometimes less.

Either way, I want it to be useful to you.

Better marketing, better business.

I’ve enabled replies to these emails now so if you'd like to chat, just hit reply.