AI may be creating new sales channels and removing friction from customer journeys, but businesses cannot assume that every customer will adopt it in the same way or at the same speed, says Tim Lawless, digital AI enablement lead at E.ON Next ahead of CustomerX
TL;DR
- Ahead of speaking at CustomerX in London, E.ON Next’s Tim Lawless believes that AI adoption will move at the speed of customer trust, not technological capability. Customers may embrace AI for one task while remaining deeply cautious about using it for another.
- He sees AI both as a way to remove friction from existing customer journeys and as an emerging sales channel through platforms such as ChatGPT.
- Businesses must support customers across traditional, AI-assisted and increasingly agentic channels rather than assuming everyone will make the same transition at once.
- Internally, customer-facing AI and workforce transformation require different approaches. Employees are more likely to engage when AI helps remove frustrating tasks rather than being presented principally as a means of replacing jobs.
- Companies should fund outcomes – such as increasing sales or improving retention – not isolated AI projects or another chatbot.
- No organisation yet has the complete AI playbook. As Lawless will explore at CustomerX, cross-sector collaboration offers businesses a chance to compare approaches, borrow intelligently and learn where they are ahead – or falling behind.
It took Tim Lawless’s mother 15 years to trust the internet sufficiently to enter her credit card details online – an anecdote worth bearing in mind amid predictions that consumers are about to hand purchasing decisions, payment credentials and control of their household accounts to autonomous AI agents in a tenth of that timeframe. AI technology may be advancing rapidly, but customer confidence does not necessarily move at the same speed.
“Trust will stop your customers from adopting AI solutions in the way that the frontier labs think is going to happen,” says Lawless, digital AI enablement lead at E.ON Next ahead of his panel session on AI at CustomerX next month.
That does not, however, mean customers will reject AI, he says. Instead, adoption will be uneven across demographic groups, between sectors and even within the behaviour of an individual customer. Someone may happily ask an AI tool to suggest a restaurant or plan a holiday while remaining unwilling to let it switch their energy supplier, access their bank account or complete a transaction. The same customer can be an enthusiastic adopter in one context and deeply cautious in another.
For businesses, this makes AI much more than a technology project. It becomes a customer, product and channel strategy.
AI as an emerging sales channel
As a result, Lawless sees two significant roles for AI. The first is to improve the customer journeys that already exist. “AI acts as an accelerator on what was already there,” he says. “The way we’re approaching this is very much: what does AI unlock that you couldn’t do before?”
At E.ON Next, one answer is using AI to remove some of the work involved in obtaining an energy quote. Previously, customers could drop out when asked to enter all their information manually. They can now upload a bill from another supplier and receive an instant quote.
“It speeds up what was already there, rather than just becoming a bolt-on feature,” says Lawless.
This distinction matters. Customers do not particularly want an AI feature and few wake up wishing that one of their suppliers would launch another chatbot. They want something done more quickly, easily or effectively.
The second role is more disruptive. AI is beginning to sit between the customer and the businesses trying to reach them.
“You’ve got to see OpenAI as an emerging sales channel,” he argues. For now, many consumers are using generative AI as what he describes as “a fancier Google”, reducing the effort involved in scrolling through pages of search results. The next stage will require considerably more trust. An AI may not simply identify the cheapest energy supplier; it could offer to carry out the switch as well.
And businesses must be ready for that possibility without assuming that every customer will immediately embrace it. “You’ve got to optimise the new channels, but you’ve also got to use AI to improve your channels,” says Lawless.
This is an important CustomerX lesson. The arrival of a new channel rarely eliminates all the others. It adds another behaviour to an already fragmented customer journey. Businesses will have to serve customers at the leading edge of agentic commerce, those using AI principally for research and those who still prefer established digital or human interactions.
“How are you performing across all those channels? How do your customers like to interact with you? Can you give them a halfway house?” Lawless asks. “This really does become not just implementing AI solutions, it also becomes more of a channel strategy and a product strategy – and not a one-off AI transformation.”
Two transformations, not one
As a result, Lawless separates AI enablement into two pillars inside the organisation. The first is customer-facing AI: using the technology to develop better products, improve journeys and produce better outcomes for both the customer and the company. The second is an organisation-wide change in how employees work, make decisions and produce output.
The two are connected, but they are not the same transformation, he says. “One is almost like a product transformation and the other is a company-wide transformation of the ways of working,” he explains.
Customer-facing improvements are relatively easy for employees to understand. Internal transformation can generate much more anxiety, particularly when discussion of AI begins with jobs, automation and task reduction.
Lawless believes that perception changes when teams are allowed to experiment with the technology in the context of their real work. AI can be framed as “almost a junior analyst” that takes on the annoying, repetitive task that nobody wants to do, freeing people to concentrate on work that “actually moves the needle”, he says.
That requires participation rather than instruction. “Until people have gone through that journey and actually played around with it, I think there’s an awful lot of anxiety that can cause people to become a bit resistant,” Lawless says. “You need people to become a bit more hands-on and build and try and really experiment with things.”
Stop funding chatbots
The difficulty is that established businesses are used to transformation programmes with defined destinations. An organisation implementing a CRM or ERP platform broadly knows what the finished project is supposed to look like. AI offers no such certainty.
“I don’t think the place that all companies want to get to is certain at the moment,” says Lawless. “No one really knows the true organisational benefits for it and what I’m seeing across the industry is no one therefore knows where to start.”
His advice is simple: create the two pillars, begin delivering useful things and learn from what happens. But experimentation must be connected to the organisation’s wider objectives, rather than trapped indefinitely in a collection of pilots.
That also means changing how work is funded. Lawless is trying to move E.ON Next from being use-case-led to outcome-led. Instead of commissioning a project to deliver a chatbot, a business could fund a product team to generate more sales, reduce customer effort or improve retention. The team can then test where AI contributes to that outcome and change course when customer behaviour or technology develops.
“Really understand your end-to-end and where AI accelerates rather than just launching a chatbot,” he says, “because a chatbot’s not really going to move the dial for most companies.”
Nobody has the whole answer
So, who is doing this well? “I don’t think there’s any one organisation which has got this whole thing right just yet,” he says, but he points to IKEA as an organisation linking customer-facing AI with changes in how its people work, including redeploying employees to create more value rather than concentrating solely on task reduction. But Lawless does not believe there is a finished model that others can simply reproduce.
That is also where cross-sector collaboration becomes valuable. Energy companies can learn from retailers, retailers from financial services and all of them from businesses experimenting with different combinations of AI, data, customer experience and organisational change.
For Lawless, the value of taking part in CustomerX is this opportunity to look beyond his “own little bubble”, compare approaches and understand where E.ON Next may be ahead – and where it may not.
The purpose is not to copy another company’s AI strategy wholesale. It is to learn from what others are building and then make it work inside your own organisation. As Lawless puts it, “[AI is] a new capability and an accelerator rather than a cookie-cutter approach”. The winners will not necessarily be the companies with the grandest AI vision, they will be those that start solving genuine customer problems, learn faster and remain flexible enough to follow customers wherever – and however quickly – they choose to go.
To hear Tim Lawless, digital AI enablement lead at E.ON Next, share his expertise on AI in commercial teams, register for CustomerX.
This article first appeared on the CustomerX Substack. Follow CustomerX on Linkedin for more updates.
