Top 5 things you need to know about customer segmentation
1) It's not the same as targeting
A customer segmentation should work independent of time, place or channel, for every team in your business. You may already have a targeting tool such as RFM - a view to help drive maximum value from a single campaign. Where there's undeniably a place to use a model like this and enhance with metrics such as loyalty, this is quite different to segmentation.
Segmentation creates a view of your core customer types and enables you to understand the best way to drive value, what channels to use, how to retain, even who to ignore - creating a long term strategy for each customer type, knowing the right way to connect with them. This of course will feed into your campaign targeting.
Employee engagement is just one of the by-products – creating 'personas' so that every employee can understand core customers and what makes them tick - you won’t get that from an RFM.
2) Start with the data
Too often customer segments are created (at worst) on a creative team’s intuition, with data shoe-horned to fit. Always start with the data – use a clustering tool to look at all possible variables and let it tell you what’s driving value.
For a segmentation to work across all teams, it needs all relevant data considered (an example might be calls into your call centre) – this is obviously dependent on a robust data structure.
Consider incorporating an industry tool such as Experian’s MOSAIC as part of your segmentation build. This is an established, comprehensive tool and when used alongside your own data can immediately give you some of the behavioural/attitudinal insight, which you may otherwise invest in research to understand.
3) Don’t get carried away!
At the recent Customer Analytics in Financial Services conference Mike De Halpert from Ebay Analytics talked about the first segmentation he ever delivered, and his delight at discovering that by combining every possible methodology (RFM, Loyalty, Behavioural, Lifestage being just a few) he could create a whopping 224 MILLION segments!
Of course he quickly realised that this demonstrates a major challenge with segmentation –simplicity is key. With data now so big that an estimated 0.5% is actually being analysed, the real challenge is identifying relevant data that can be powerfully used to create segments. Manage the complexity brutally. The skill is coming up with something clean, that works – which you'll be measuring through continued, robust ROI analysis.
4) ...and end with data
It’s only half the challenge to create the segments.
Enhance them with everything you know about a customer – their spend patterns, attitudinal, geographic are just a few examples. This is where the role of research comes in – actually going out and completing in-depth interviews with customers to understand how you can connect with them.
Lifetime Value is an important segment metric; have you got a lower spending segment that sticks around for years? If you have the capabilities then measure social interactions and analyse the value of this. How should you treat a segment that’s particularly talkative about your brand?
Visualise data wherever possible, create personas that excite people, and can be easily digested not just internally by every member of the business, but also your stakeholders, for example your marketing agencies. If budget permits then create a bible of customer information so that every one of your front line staff knows their customer types inside out.
5) Once you've got it - use it!
Creating and setting out your core customer segments will drive minimal value for your bottom line. The real value comes from using them. Here's just a few of examples;
• Maximise your ROI on any acquisition activity by tightly targeting lookalikes of highest potential segments
• Personalise. We recently recommended changing window displays for a fashion retailer where the representation of core customer types in store did not match the local area potential
• Thinking Customer Segments in every business decision – products, creatives, promotions, targeting are just the start…
Find out more about Ikano Insight here.
Thank you for your input
Thank you for your feedback
DataIQ is a trading name of IQ Data Group Limited
10 York Road, London, SE1 7ND
Phone: +44 020 3821 5665
Registered in England: 9900834
Copyright © IQ Data Group Limited 2024