I’m not a Justin Bieber fan. But there are a number of people in my family who are. You could even call them ‘Beliebers’. There is a regular stream of his, let’s call it “music”, that can be heard around my home, as well as his lifesize cardboard cutout presence in my eldest daughter’s room. Marvellous.
On one occasion, I took to Twitter to serve my cathartic need and express my opinions. I felt better, for at least a few seconds. A couple of days afterwards, I then receive an email from a major credit card company, one I have been a member of for several years. Also a credit card that I also haven’t used for a similar period.
The email was offering me the opportunity to purchase Justin Bieber concert tickets prior to going out on general sale. Spooky and a quandry. Do I buy the tickets and make my family very happy? Or do I pretend the email was a figment of my imagination.
So, of course, I took advantage of this ‘deal’ and bought the tickets, with the caveat that I had to renew my annual membership.
Of course, this could have been a complete coincidence, but having an idea how these things work, I think not. Some pretty clever data correlation had taken place, which knew exactly who I was, matched my social profile with my email address, knew my family make up, and that at this particular moment in time Justin Bieber was on the agenda.
It was a very relevant piece of content, and served to me via email (not via a tweet reply, which may not have had such an impact). Timing was key.
If you were to look at Maslow’s Hierarchy of Needs, my motivation was to please my family, gain a limited amount of status and recognition and become a hero in my daughters’ eyes for at least a fleeting moment. That’s why I did it.
So, is there a model that could be applied to match motivation? A set of data-layers that serve as a means to ensure relevance can be delivered in-line with what people want? Maybe there is, and one that can be used as a framework to help organisations with their consumer and customer communications, to bring together the “who” and the “when”. I’m putting forward this one.
The more ‘traditional’ segmentation models don’t necessarily include real-time dynamics, they call upon the less frequently updated data, mainly based around the ‘Who’; geographic, demographic and ideally psychographic. These make up the three base layers of the model.
- Geographic: The physical and more permanent location of home, urban, rural, density, place of work, etc.
- Demographic: The attributes of gender, age, family make-up, social class, affluence, etc.
- Psychographic: The values, beliefs, attitudes and behaviours of people, what they like, what they don’t like, their interests and affinities.
All of the above provide a very rich picture. People offer their information across various sources, which organisations need to be bring together to start to build the ‘single customer view’ and deliver relevance.
Information across all these three layers does change, but the frequency associated is not perceived to be that dynamic, although some psychographic information can be quite faddy and ideally needs to be kept as up to date as possible. e.g. some social data such as which brands people ‘Like’ on Facebook can actually become quite ‘dusty’, dependent upon why they liked the brand, when they liked it and whether or not people keep their ‘unlikes’ up to date. Most do not.
To increase relevance, these three layers can be enhanced by adding a further layer on top. They underpin the fourth layer, or the apex, which is all about people’s actions. The ‘when’.
- Actions: The activities people undertake in real-time, which includes transaction and purchase activity, web behaviour, social intelligence, mobile activity, TV viewing, etc.
Actions turns fairly static ‘segmentation’ models into ‘audiences’, who ebb and flow around the things of interest at any given time. However, actions on their own aren’t enough. You may get lucky by reaching the right person at the right time, with the right content, but without the three underpinning layers, achieving relevance will be a game of chance.
There’s no escaping the need of Big Data, the combination of structured and unstructured, which is then finely tuned and filtered.
It is appreciated that the practical aspects of pulling these four data-layers together and combining the who and when is no mean feat.
However, the benefits of applying the framework can better inform a number of things:
- ‘Targeting’: I don’t like this term, but it will enable organisations to ensure they reach the right people at the right time.
- Content: From a content strategy perspective, it will enable marketers to work out which content should be served to who and at what time in order to be compelling and relevant.
- Multichannel: It will enable the ability to serve people with content via the medium they prefer or will take notice of. i.e. a piece of real-time social intelligence can be used to inform a piece of personalised DM, which can be delivered the next day.
- Attribution: It will also provide a means to construct more accurate and continual ‘test and learn’ approaches and provide a means to build robust attribution models.
- ROI: Increased relevance around audiences will reduce wastage (if serving content via a physical medium), reduce spam, enhance interest and affinity and increase the more tangible metrics of enhanced loyalty and sales.
I’m hoping this basic set of layers can help you as marketers and that it serves a purpose when you’re constructing a framework to deliver relevant content, across multiple channels, to achieve enhanced loyalty and increase acquisition and sales. It would be good to know your thoughts. Thanks for reading.