How AI can measure and lift your KPIs for customer success


Share on LinkedIn

We all want customer success. More sales. Better financial results. Happy customers buying our products at good margins. No, make that great margins. We would all like customers like Apple has. Customers that are highly engaged by the most important measure of engagement there is – customers buying their goods and services in volume, at what everyone believes are fabulous margins. Amazing. But… the ‘but’ is that engagement leading to great customer success is harder said than done. Much, much harder.

So where to begin? First and foremost, we need a good measure of engagement. Currently, the most commonly used measure of engagement is a net promoter score (NPS), but NPS is a spot measure of the customer, at a moment in time; not a comprehensive view of their commitment to the brand and products they buy. Ideally that commitment is measured in terms of their possible future purchases in local currency terms.

The latest object-oriented AI and reinforcement learning modelling offers marketers a way to build a comprehensive measure of commitment to a brand and products, derived from all of the digitally recorded data a vendor has accumulated about a customer journey – purchase transactions, marketing events, warranties, recalls, service events, external events from the internet – such as social media, even macroeconomic events. All compiled in one customer journey per customer and processed by the latest, patented AI technology.

The magic is that this measurement of brand commitment and engagement is only a first step. Marketers can use this as a key variable in the most important aspect of customer experience – the customer success we outlined above – selling to engaged customers at great margins – but what products and when? What can they do to move that process forward? Can we make the right offer at the right time? And how will that offer be received?

Smartphones and the Internet has made everyone speed up. More and more vendors are having to work at the speed of the customer and the internet. ‘Boom’. Too late. Your customer just bought from a competitive vendor, the offer received and researched on their phone, while they were commuting.

If we cannot move at the speed of our customers, we will be left behind. Unfortunately, NPS, which measures customer experience at a given point in time, must wait for another interaction, which may take place months or years later. At the same time, we cannot ask a customer every day or every week to fill out an NPS scorecard, they will just stop answering, or reading our emails.

Dollars and cents

Let’s come back to that measurement in dollars and cents, measured dynamically, and keyed to outcomes that vendors want to track. If a customer buys a vehicle for $50,000, we want to know about every customer interaction, from marketing to warranty claims, to service events. In other words, all the events that would cause that customer to increase or decrease their propensity to purchase another vehicle from us; and at what price point, higher or lower, than their original starting point or value.

The data available today makes it possible to analyze every touchpoint, not just rely on survey results. A customer may have hundreds of touchpoints with a vendor, but never fill in any surveys. The more variety of touchpoints, from marketing to sales to support; the better the results. Understanding and deriving customer commitment using sophisticated AI and machine learning models is revolutionizing how businesses are now, or should be, interacting with their customers.

Not only can these models measure, they can now go one step further and recommend the next best actions (“NBAs”) that a marketing team should take with a customer. Each customer, one at a time, not in super segments, that lump in thousands of customers together based on typical demographics. Instead we can get to know individual customers, who are treated and valued as individuals. NBAs which can be incredibly helpful and can move at the speed of data, thus giving customers a more targeted offer, deal or action.

Again, the goal posts have moved. Basing NBAs on systems that use traditional “rules-based” or “Ai-Lite” technology is now obsolete. Three years ago, is a long time ago. Now marketers can use state-of-the-art reinforcement learning to produce NBAs, sequenced events timed and valued. Even allowing you to include margins for the various events, to get both gross and net margins.

Example: send an email and follow-up with a call center interaction. Emails are relatively “free” from a cost point of view, but call center interactions have serious variable cost implications. Email works, call center interactions work, but using both together may work best of all.

Jean Belanger
Currently CEO of After graduating (LSE), I joined Wood Gundy. I left to start a VC fund, investing in start-ups. 3 went public. After 15 years in finance, I decided to run companies rather than finance them. The first, programming tools vendor, Metrowerks CodeWarrior, built most of the software used on the Mac in the 1990s. When Metrowerks was acquired by Motorola, I was named VP Biz Dev for their semiconductor business, where I invested over $450M in M&A in 14 months. After Motorola, I started data science supply chain software and IOT pioneer, Reddwerks.


Please use comments to add value to the discussion. Maximum one link to an educational blog post or article. We will NOT PUBLISH brief comments like "good post," comments that mainly promote links, or comments with links to companies, products, or services.

Please enter your comment!
Please enter your name here