The rise and rise of Big Data.


“Big Data”, “data mining”, and “segmentation strategy” – all words we hear a lot of in our industry these days. But what are they really? And how do they relate? And how can we ultimately use them to increase ROI?

Big Data
According to IBM, 2.5 quintillion bytes of data are created every day and 90% of all data in the world has been created in the last two years. Data is no longer just customer demographics and transactions – it has many more layers like social media data, locational data and online search data to name a few. This increasing complexity means greater insights about your customers are now within reach.

Data mining
Data mining is the process of uncovering these insights from these large and complex data sets. These insights lead to the production of business and communication strategies.

Segmentation strategy
Your communications strategy may require a segmentation strategy that allows you to target sub-sets (segments) of your customer database in different ways in order to satisfy your larger business strategies. The segmentation strategy recognises that your customer base is comprised of different persons, with different needs, behaviours and potentials who should be communicated with very differently. It groups customers with similar characteritics/behavioural patterns to efficeintly communicate a more personalised/relevant message to the particular segment of your customer database. It also allows for more personalised and relevant incentives to be communicated to your customers – ultimately increasing your sales. For example, instead of sending blanket advertising to your customer base as a whole, you might send your VIPs more regular communications as they are more likely to purchase, and you might target your customers who haven’t been in-store for some time with a great offer to bring them back.

Segmentation has greatly benefited from the era of Big Data and advances in data mining.  These advances enable customer bases to be segmented more effectively and into even smaller fragments – enabling still more personalised messages direct to customers.

The good news is that this progress is only going to continue and, as the cost of the mediums that deliver personalised messsages decreases, we will get closer to the utopia where customers only receive messages they are interested in and marketers increase their ROI (by reducing wastage associated with sending generic messages to a mass market).  But, as Hywel Evans, Regional Director of Decision Science Asia Pacific, said at the Marketing Association’s recent Smarter Data conference, the true 1-1 utopia does not render segmentation strategy irrelevant. In order to grow a business, companies need to uncover insights from their data and create business strategies around these insights. It would be too expensive to create a business strategy for every single customer, therefore strategies can only hope to change the behaviour of groups of customers within the database, to grow the business overall.

justONE developed a cutting edge segmentation strategy for Ziera to use in their new season communications to their customer base. With help from Datamine, insights were uncovered within Ziera’s customer database that enabled distinct groups of customers to emerge (based on certain behavioural aspects). Ziera now have the ability to communicate distinctly with these groups – and provide them with more relevant messaging. This forward-thinking strategic vision saw justONE and Ziera recognised at this year’s New Zealand Direct Marketing Awards, winning the “Best Data Strategy” award.

Where do you start to employ a segmentation strategy in your business?
First, you need to ensure you have a great process in place for collecting the right data (if possible discuss what kind of data you should be collecting with a data mining consultant). Once you have your data, data mining is used to analyse and uncover insights about specific groups within your database. Some examples of useful insights from data mining are:

  • Flagging those customers who are more likely to stop purchasing: Within your data you can attempt to identify common characteristics among your customers who have stopped purchasing. You can then search for similar customer profiles and highlight them as a group who are at risk of also stopping purchasing. This is essential for customer retention. An example could be a mortgage customer that is coming up to the end of their fixed-term mortgage. The data might show a significant number of customers leave at this stage and you may be able to reduce the churn in this group by targeting them one month out from the end of the term with a great offer, ensuring they re-fix their mortgage with you.
  • Grouping customers who are interested in similar products: Your data may show that female customers between the ages of 20-30 are more likely to purchase beauty products than all other age/gender groups, therefore you want your catalogue to these customers to showcase lots of beauty products, whereas your catalogue to your older, male customers won’t.
  • Finding a customer’s potential:  Looking at a customer’s entire transactional history rather than their most recent behaviour helps to cross/upsell as it predicts a customer’s potential spend – not just their current spend levels. If recently a customer has only been purchasing items on sale, but has in the past spent a significant amount of money on full price items, they have the potential to repeat this behaviour and therefore communications to them shouldn’t just be focused on the promotion of sale items.
  • Finding common characteristics of your current customers in order to better target prospects: By targeting prospects who are more likely to become customers than the general population, you greatly increase ROI from prospecting. Your customer base may be female, predominantly urban, between the ages of 20-30 and spend in excess of $10,000 on credit card transactions annually. You can target prospects who fit this description specifically to increase your chances of your communications resulting in new customers.

Once you have uncovered distinct groups of customers within your database with similar behaviour, you can create strategies to change the behaviour of these customers.  By providing more personalised and relevant messages to your customer base, you will receive higher response rates and ultimately higher revenue. The benefits don’t just stop there – the customer also benefits as they receive personalised messages that are more relevant and useful to them. It’s a win-win.

Author: Virginia Bashford