For years it has been something of a cliché in the loyalty world to describe attempts to understand customer behaviour through data as “drinking from the firehose” (or hydrant depending on your geography). Given the advent of social media, other high-volume, unstructured data sources and the "always on" customer, the imagery probably needs an update; loyalty marketers and analysts are these days faced with a tsunami of data. For example:
- Typical customers of a mobile telecom company generate daily data records in the billions per day
- 500 million tweets per day are made on Twitter
- By 2020, 80 zettabytes of data will be created per annum, 20 times that of today.
This explosion across the “3 V’s of data” (volume, velocity, variety) has taken many marketers as well as loyalty marketers by surprise. Relationship marketing has always been a very data-centric activity, and for a long time the data that drove loyalty programmes was structured transactional and customer profile data, held in relational databases. Today, progressive loyalty managers are harnessing social media and web data to enhance their understanding of customer behaviour and their sentiment. If an organisation’s highest value customers are referring to them in blogs or on social media sites, then loyalty practitioners have an interest in that data footprint both from a customer service perspective but also from an analytical sense – what is the trend over time of this sentiment amongst the most valuable customers?
The challenge for loyalty marketing is how to incorporate these new data sources into the traditional Single Customer View (SCV) approach to organising and centralising all that is known about each individual member.
In theory, a true SCV model for the modern day should encompass some if not all of the new universe of available data and integrate them seamlessly; alongside traditional transactional and profile data. This in turn should be able to drive marketing across traditional direct channels but also through digital and mobile personalisation. This is no small task and in order to deliver on this vision, a number of issues need to be fully understood and addressed:
Scale - The sheer volume of data and the number of different sources is potentially overwhelming. Even within the sphere of social media, different channels (e.g. Facebook vs. Twitter) may require different approaches. Added to this there is web data and mobile data each requiring understanding and a bespoke approach. How is the approach prioritised?
Matching - How to match the ‘new’ data sources with the ‘old’ is fundamental. The existing SCV approach requires the ability to uniquely identify a customer such as email/physical address. It is unlikely that these will always be present amongst mobile, social media or web data. However, cookies, URL’s, tagging, social media ‘handles’ and other identifiers may provide a solution here
Technology - What, if any, technology is required to support a ‘New SCV’ proposition. Do existing IT and database platforms provide some or even all of what is required or are there IT solutions in the market that can help support the achievement of objectives? The existing traditional relational database/SQL server based platforms and BI tools of many organisations will not offer sufficient capacity or speed for today’s data volumes. Cloud-based services and tools created to scale exponentially are likely to play some part here
ROI - Most crucial of all, how do organisations investing in new data ventures make a return on this investment? Actionability is fundamental. Data needs to be delivered to marketers in a commercially-useful form to drive more targeted, personalised and real-time customer experiences and communications, as well as adding value across other disciplines such as product development.
Big data represents a huge opportunity to data-driven marketers, but without the right tools or knowledge there is also a risk of becoming overwhelmed and even paralysed by the sheer scale of the datastreams available. In order to harness the potential of Big Data, even organisations with hitherto robust and advance analytical IT infrastructure and capabilities need to take a fresh look at how they support their data activities and how the volumes associated with Big Data can be compressed and transformed into actionable insight that will make a positive difference to both the customer experience and the organisations commercial performance.