The optimism of #Purpose & science of #design in forging #digital customer experiences! Opinions my own!

“Give a man a fish, and you’ll feed him for a day; give him a religion, and he’ll starve to death while praying for a fish

Event based marketing or EBM for short ain’t new – it’s been around since the 1990s. Tracking, detecting customer behavior using data for smarter interactions isn’t supposed to be rocket science – it might be tedious, but not conceptually beyond reach for most. Personally, it’s one of my favourite marketing disciplines as it embodies the “do more with less” belief – similar to a form of martial arts called Wing Chun or it’s modern adaptation by Bruce Lee, jeet-kune-do. Simply put, attack when attacked (the shortest distance between you and the opponent) and with the most damage inflicted in mind, using the least effort – this is almost like Sun Tzu’s #ArtofWar in marketing!

The opponent in this case are our customers, and the idea is we orchestrate interactions or “attack” using they display significant behavior change “shortest distance”.

Many a vendor touted their event libraries, a collection of “ready made” experiences from previous clients. How the prospect  t hen responds to this says alot about their sophistication.

For advanced practitioners, they would question the degree of suitability of  these templates given they’re implemented in different geographies and/or for different business objectives. Prospects behind the curve  – the laggards – they will be duly impressed the thicker the event library is. Somehow they think these collective experiences are portable and saves them time and having to think for themselves. Risk mitigation I suppose.

Technologies supporting event-driven marketing are dime a dozen. Writing SQL codes, querying databases is but the norm for most – and often times, prospects confuse this for the larger EBM concept. It’s very much like playing golf – we all play, but do we play it well…? Hence the rhetorical question: would increasing the complexity and temporal vectors of detection yield significant improvements?

A classical definition of EBM is the ability to “detect” changes in customer behaviour, and as such writing a piece of code to query a relational customer database technically complies. Having to industrialize this across millions of customer records and adding complexity in patterns detected – this is where manual work falls short.

Imagine a customer being detected for placing a large sum of cash in a deposit account. Now  the definition of “large” (sum of cash) would be subjective from customer A and customer B – this implies that there’s a moving average, a snapshot for every customer in consideration – not so easy if we have to scale this by segments. Here we also hint at having to develop an offer matrix, an interaction map if you will.

What’s more the time horizon (of detection) matters – a large sum of cash should be detected on a daily, weekly of monthly basis? In actual fact, it should be all of the above – but computational power required would also increase linearly and complexity can be a hindrance – we could face diminishing returns when “good enough” actually suffices.

This is when the plethora of available detection mechanisms come into play: from literal and explicit to state-based and temporal.