Using AI and Predictive Analytics to Increase Customer Value by 34%

I recently presented at the Technology for Marketing event, showcasing a series of case studies on how RedEye has worked with clients to use AI and Predictive Analytics to increase Customer Lifetime Value. It’s such a huge topic that in the 25 minutes allotted I don’t know whether I was able to do it justice! I was really rushing the key elements, which were the actual results, and I certainly did not leave enough time for questions! So, I thought I would write a series of blogs expanding on the key points I wanted to get across!

Below is a summary of the five key points I was trying to make that I will expand upon in the next couple of weeks:

1. Predictive, AI and Machine Learning are not a fad, they are here to stay. Things come and go, some faster than others. But with AI and Predictive Analytics for multi-channel marketing, the results RedEye is able to achieve for our clients is enough evidence for me to believe that this one is definitely here to stay.
   

2. When we are trying to understand and predict a prospect or customer’s ‘next best action’ the most powerful data is always recent data. Recency is king. Today recency is behavioural or engagement data, or to be a bit more explicit, how a prospect or customer is engaging with your brand. If you can map patterns of behaviour at an individual customer level, then these patterns give clues to likely outcomes and then marketing activities can be developed to influence the final outcome you want to see.
   

3. Data is absolutely key. There is no firm agreement about the outlook for the Customer Data Platform tech space, but I would argue that the requirement is definitely there. For too long organisations have ‘made do’ with limited or siloed databases that do not give a full view across online and offline data and therefore do not give a complete Single Customer View. To ascertain the patterns in the data that I describe above, businesses need to be able to tie together customers not just by household and email address but by website activity and app engagement (to name but two), as well as across the various devices they might use. CDPs and other capabilities now resolve this problem and it is this data that opens up real opportunities for marketers using Predictive Analytics and AI.
   

4. Integration is crucial. Integration is, of course, at the heart of marketing automation and AI is not going to function if the key elements (data, analytics and channels) are manually driven or not tightly integrated. So many organisations are still reliant on customer databases dumping data into an ESP on a weekly basis and this is not an architecture that is going to open up opportunities in AI and Predictive Analytics.
   

5. I asked the audience for a show of hands on the question of whether their organisation used Customer Lifetime Value as a business KPI. I would estimate that about 10% of the audience said that they did, and assuming that brands were in the minority in the audience that would fit with Econsultancy’s estimate that 42% of organisations track the measurement. But for me it is a perfect fit with AI and Predictive - it is measurable and it resonates throughout an organisation, from platform users to Boards.

If you attended the presentation (thank you!) you will have seen me reference back to the Where’s Wally cartoon, which gave me the chance to reiterate my overarching point. As marketers, given the knowledge of where a prospect or customer is likely to be on their individual journey with your brand, do you feel as though you could influence that prospect or customer towards a favourable outcome for your brand?! That is, of course, a rhetorical question… I hope…

Keep an eye on my LinkedIn page for future blogs!