Artificial intelligence has arrived in automotive retail with a level of pace and substance that makes it hard to ignore. For many dealer principals and CEOs, the conversation has shifted quickly from curiosity to pressure. Competitors are talking about it, vendors are selling it, and the industry narrative suggests that standing still is not an option. That much is true. What is less often said, and far more important, is that most businesses are approaching it the wrong way.
The instinct is to start with systems. New platforms, integrations, and dashboards. It feels like progress, it looks decisive, and it gives the impression that something meaningful is underway. In reality, it often leads to the same outcome: technology that sits underused while teams carry on as they always have.
If AI is going to make a difference inside a dealership, it has to begin with the people who will actually use it. That sounds obvious, but it is rarely where the investment goes first. There is still a tendency to assume that capability will follow once the software is in place, but experience suggests the opposite. Without the right level of understanding, confidence, and support, even the best systems fail to deliver.
The scale of the opportunity is not in question. Across industries, AI is already improving productivity, sharpening decision-making, and removing hours of repetitive work from people’s days. The long-term potential is widely accepted to be significant. At the same time, very few organisations would claim to have fully embedded it into their operations. There is a gap between investment and impact, and it is not small.
In a dealership environment, that gap tends to show up quickly. This is a business built on pace, pressure, and people. Sales teams are managing leads, customers, and targets in real time. After-sales is balancing efficiency with service quality, while managers are expected to have a clear view of performance without always having the time to step back and analyse it properly. Introducing AI into that mix without bringing the team along is unlikely to land well without some planning.
What tends to work better is a quieter, more grounded approach. Start by helping people understand what AI actually is in practical terms. Not the theory or the headlines, but what it can do for them in their role. That means being honest about its strengths and its limitations. It is very good at handling routine tasks, summarising information, and spotting patterns. It is far less reliable when judgment, nuance, or accountability are required. Once people see that clearly, the conversation becomes more sensible.
Training plays a bigger role here than many expect, but only if it is done properly. Generic sessions tend to wash over people. What makes a difference is when a sales executive can see how AI might help them follow up leads more effectively, or when a service manager can use it to get a clearer view of workshop performance without spending half a day building a report. When it connects directly to the job, it stops feeling like an abstract initiative and becomes useful.
There is also a human side to this that cannot be ignored. Not everyone is comfortable with the idea of AI being introduced into their work. Some will worry about whether it makes parts of their role redundant. Others will simply lack the confidence to engage with something new. There will always be a group that has seen enough game-changing tools come and go to be sceptical from the outset.
Those reactions are predictable and often justified. What does make a difference is taking the time to explain why change matters, how the technology will be used, and what it means for them. When leaders are clear that the aim is to remove low-value tasks and make jobs more manageable rather than replace them, it tends to shift the tone of the conversation.
It also helps when leadership is visible in the process. When dealer principals and senior managers use these tools themselves, even in simple ways, it carries more weight than any formal communication. It shows that this is not being pushed down from the outside, but adopted from within.
In terms of implementation, the businesses that make progress tend not to overcomplicate it. They start with a small number of clear use cases that solve problems people recognise. That might be something as straightforward as summarising customer interactions, generating performance reports, or helping teams communicate more consistently. None of these is transformational on its own, but they are visible, practical, and easy to measure.
Once those early examples begin to show results, the dynamic changes. People start to see the benefit for themselves. Conversations become more constructive. There is less resistance and more curiosity. From there, it becomes easier to expand into other areas of the business.
One of the more effective ways to support that is to identify individuals within the dealership who take to it quickly and give them a role in helping others. These are not necessarily technical people. More often, they are operators who understand how the business works and can translate AI into something their colleagues can relate to. When that support comes from peers rather than external consultants or central teams, it tends to land better.
Alongside this, there does need to be some structure. Clear boundaries around data use are essential, particularly given the nature of customer information in automotive retail. People need to know what is acceptable and what is not. Without that clarity, either nothing happens because teams are unsure, or the wrong things happen because they are not.
Equally, it is worth being disciplined about measuring what is actually changing. Time saved is a good place to start. Improvements in response rates, reporting accuracy, or stock management decisions are others. These are the kinds of outcomes that matter in a dealership context and also justify further investment.
AI allows sales teams to spend more time with customers and less time on admin. It gives managers quicker access to information that would otherwise take hours to compile. It creates a level of consistency that is difficult to achieve manually. In that sense, it acts as an amplifier. Good teams become more effective. Weak processes become more visible.
That is why this sits squarely with leadership rather than technology. The decisions being made are not just about tools; they are about how the business operates and how people are supported to perform.
For those running dealership groups, there is a familiar pattern here. The industry has been through enough waves of change to recognise it. Early adopters gain an edge, others follow, and over time, the baseline shifts. The difference with AI is the speed at which that shift is happening.
There is no need to rush into large-scale transformation programmes. In fact, that is often where things go wrong. A more measured approach, starting with the team, focusing on practical use, and building from there, tends to deliver better results.
The businesses that get this right will not be the ones with the most advanced systems on paper. They will be the ones where the team understands how to use them, trusts them, and sees the benefit in their day-to-day work.
Adrian Favill is a director of Mad Devs
