NicoElNino - Fotolia
The robots are coming – and they’re going to take your jobs. That is a popular mantra, but I really want to let some of the (hot) air out of the inflated robotics and artificial intelligence (AI) hype.
My starting point isn’t that AI and robotics are not going to be important – they are. But AI is hugely overhyped, hugely misunderstood, and already hugely impactful. We are building a mirage of a robotics revolution on the backs of some interesting startups that haven’t delivered much at all in self-thinking software or hardware robotics that’s going to have mainstream impact.
This confused situation is the creation of my industry. Enterprise IT suppliers talk too much about parts of AI that deliver the least potential value. This is perverse when there is AI that has proven to deliver tremendous value and is underused today.
A great example of this is deep learning. Fascinating stuff, but, I would argue, not yet a game-changer for a commercial organisation such as a bank or insurer. These businesses need AI to do the hard work of growing revenues and driving down costs. The challenge here is how AI could transform their interaction with customers – how can AI make a customer more loyal, more ready to buy more services and products?
In the final analysis, software robotics is a new way to describe what many businesses have been doing for years – automating more repetitive, sometimes complex, tasks.
Today, AI is at the heart of some fundamental processes that affect profit and loss in meaningful ways. For an energy utility, machine learning in its marketing systems continually improves how it presents the best offers by learning about customers in real time. This has delivered a 20% increase in conversion rates. For an insurer, robotic process automation sequences the activation of apps and data to cut out dull, mundane tasks and speed up the claims process.
Perhaps not as exciting as human androids, but these are examples of AI and software robotics embedded in everyday business processes. The challenge is how you get AI to grow with your business and not become an assortment of technologies that are all fantastic, but don’t integrate or scale well.
So, my checklist for the CIO or their close colleagues in marketing and customer service might be the following:
Unify from the get-go
When chatbots fail in their human interactions, it is a result of how AI deployments are too often disconnected. The quiet success of today’s AI deployments is in the effort to unify the intelligence behind the technology – an always-on central “brain” that functions across all different lines of business, channels, systems and data. This ensures that customers get consistent treatment, and the best treatment, no matter how they interact with your organisation.
Think human first
The goal should be to humanise rather than mechanise interactions, because that’s what delivers the best value – applying AI to give intelligent guidance to human customer service agents, so they can provide customers with next-best actions personalised to them. Or empowering customers who want to look after their own affairs with intelligent guidance and support when needed. Or use robots to take away rote tasks, so employees can focus on interacting with customers.
Continuously optimise the customer experience
Connecting to big data can identify opportunities for better customer service and experience. Detect, predict, test and learn in real time when customers need help, and act immediately, improving sales results and customer satisfaction. Act proactively and pre-emptively, connecting to event streams and finding issues or patterns of behaviour before they become a problem for the customer.
Govern your AI with rules
Develop a library of machine learning algorithms that allow you to automatically anticipate customer needs and trigger next-best actions as well as open fulfilment or service cases. By using business logic, the business has the ability to change to ensure that machine learning is governed by rules that keep it within regulatory bounds and aligned with corporate values.
Develop strategies powered by machine learning
It is so important to democratise access to AI tools. Choose drag-and-drop visual tools that can be used by business people rather than data scientists, so they can easily connect streaming big data and machine learning algorithms into their decision strategies. Filtering ensures that unnecessary data is not evaluated, which safeguards performance, especially for data streams that can entail millions and even billions of data records.