Trying to Keep Up With AI? You’re Not Alone.
Something has changed in the conversations I’m having with businesses about AI.
A new AI tool appears and within weeks there’s a conversation about using it.
Generate content. Automate lead generation. Improve customer service. Analyse data. Replace parts of a marketing workflow. Or even give an AI system responsibility for work that people were doing themselves.
They’re moving quickly. Sometimes very quickly. And I understand why.
The technology itself is evolving at a ridiculous pace. What was difficult, expensive or simply impossible a year ago can now be done with a subscription, a bit of experimentation, and a few simple prompts.
But I’ve also found myself thinking about something that feels slightly at odds with how we normally approach good marketing.
We plan.
We start with the business objective. We look at the market, the customer, the problem we’re trying to solve and the options available to us. We consider the risks. We decide what makes sense. Then we implement, measure and adjust.
With AI, I’m seeing a different sequence emerge.
Look what this can do.
Could we automate this?
Could we generate that?
Could AI handle this?
And perhaps the question we should be asking first is:
Should we?
That’s the bit that interests me.
I’m not suggesting businesses should slow down and wait for AI to settle down. They shouldn’t.
There are genuine opportunities to make marketing more efficient, effective and scalable, and businesses that ignore them will have their own problems to deal with.
But there’s a difference between experimenting with a new technology and building your marketing operation around it.
Businesses can now have ChatGPT, Gemini, Copilot, AI-enabled CRMs, automation platforms and AI search tools all sitting somewhere inside their marketing operation.
But having access to all of those things doesn’t necessarily mean they have an AI marketing strategy.
In fact, I wonder whether we’re entering a period where businesses don’t need more AI education or more AI tools.
They need someone to help them make sense of it all.
Because the more AI can do, the more important the decisions around it become.
Start With the Problem, Not the AI
The obvious response is to start looking for the right AI tools.
And there are plenty of them: tools for content, research, analytics, sales, customer service, CRM, automation and almost every other part of marketing.
But there’s a danger in starting with the technology rather than the problem.
If you start with a tool, you’ll naturally look for things that tool can do. Start with the business problem, and you might reach a very different conclusion.
Take lead generation.
A business might decide it needs an AI agent to find prospects, write personalised emails and follow up automatically.
But if the sales pipeline is low on leads in the first place, there could be an underlying problem. Perhaps it’s poor targeting. Or weak positioning. Or an offer that isn’t converting. Or a sales process that isn’t working.
AI might help, but first you need to make sure the process is robust before you automate it.
That starts with having a clear marketing plan and understanding what the business is actually trying to achieve before deciding where technology fits.
And why I think businesses need to start asking a different question:
Where could AI create the most useful change in our marketing?
That might mean automating repetitive work. It might mean making better use of customer data. It might mean improving research or content production. It might mean changing how the business approaches search.
Or it might mean deciding that something shouldn’t be automated at all.
Because capability isn’t strategy.
That’s why digital strategy needs to start with the business problem and the value you’re trying to create, rather than the technology available.
Knowing that AI can do something doesn’t mean it should do it, that it will create commercial value, or that the risks are worth taking.
And that distinction becomes even more important as AI moves from helping people do things to doing things itself.
When AI Shifts From Assistance to Action
We’ve moved quickly from using AI to help us do things to asking it to do things for us.
ChatGPT can write an email or analyse a report. An AI agent can potentially access information, make decisions and take action across multiple systems.
Once AI is connected to your CRM, marketing platforms and customer data, it stops being just another tool. It becomes part of how the business operates.
I’ve written before about the risk of outsourcing too much of our thinking to AI in The Skills We’re Losing to AI Without Realising It. The same issue becomes more significant when we start giving AI the ability to act on those decisions.
If we start asking AI what to do before we’ve decided what we should be doing, we risk simply making bad decisions faster.
And when those decisions are automated, they can happen at a scale that wasn’t possible before.
Once AI starts taking action, the questions become more important:
What should it be allowed to do? Where does human judgement need to remain?
AI Is Already Part of the Customer Journey
People are increasingly using AI to research products, compare options, ask for recommendations and answer questions that they might previously have put into a search engine.
That changes the relationship between a business and its potential customers.
The customer may no longer be searching for a list of websites and deciding what to click. They may be asking an AI system to research the market, compare options and help them decide.
That creates a different marketing question:
How do you make sure your business is understood and considered when an AI system is helping your potential customer make a decision?
We’re starting to hear terms like GEO, AEO and AI search optimisation. The terminology will probably keep changing. But the underlying challenge stays the same.
Businesses need to make sure the information AI systems can find about them is accurate, useful and representative of what they actually offer.
That means thinking beyond keywords.
What does the business stand for? What does it know? What questions does it answer? What evidence supports its claims? What information exists online that helps an AI system understand the business, its expertise and its relevance to a particular customer?
And perhaps most importantly, what makes the business meaningfully different from the alternatives?
Because this isn’t simply about optimising for AI.
It’s about making sure the business is well understood, supported by evidence and sufficiently differentiated for an AI system to represent it accurately when helping someone make a decision.
AI search doesn’t remove the need for marketing strategy. It makes the fundamentals of marketing strategy even more important.
When Everyone Can Create Content, What Makes Yours Matter?
For years, businesses have been told they need to produce more: more blog posts, more social content, more emails, more video.
AI makes that increasingly easy.
But if everyone can produce an endless stream of competent content, the ability to produce it isn’t much of a differentiator anymore.
What becomes valuable is the information behind it.
Your experience. Your customer knowledge. Your data. Your research. Your opinions. The things you’ve learned by actually doing the work.
That’s also what can make a business more useful and more distinctive when customers are using AI to research their options.
AI can turn those things into content. It can’t create the underlying expertise simply because you asked it to.
So perhaps the opportunity isn’t to use AI to produce more content simply because we can.
It’s to use AI to extract more value from what the business already knows.
That requires a different approach to content strategy.
Instead of asking, “What can we get AI to write?”
Ask:
“What can we share that would be genuinely valuable to our customers?”
That’s a much more interesting question.
AI Doesn’t Fix Bad Processes. It Scales Them.
One risk of rushing into AI is that we might automate a process that wasn’t working particularly well in the first place.
AI can generate thousands of emails, process customer data, publish content or make decisions far faster than a person can.
If you’ve designed the process badly, AI doesn’t fix that.
It scales it.
There are obvious risks around privacy, accuracy and security. There are also less obvious ones.
What happens when an automated system makes a decision that doesn’t reflect your brand?
What happens when inaccurate information reaches thousands of customers?
What happens when a team stops checking the output because the system has been reliable enough in the past?
And what happens when nobody is quite sure who is responsible when something goes wrong?
These aren’t arguments against using AI.
They’re reasons to think about the process before you automate it.
The more autonomy you give an AI system, the more important it becomes to decide what it can access, what it can change, what requires human approval and what happens when it gets something wrong.
That’s not just an AI problem.
It’s a marketing management problem.
So What Does an AI Marketing Consultant Actually Do?
If businesses don’t need someone to simply show them another AI tool, what do they actually need?
They need someone who can connect the technology to the marketing strategy.
That might mean auditing what they’re already using, identifying where AI could create genuine value, redesigning workflows, connecting systems or deciding where AI shouldn’t be used at all.
It also means understanding the risks.
What data is involved? What decisions can AI make? Where does a human need to remain involved? What happens when the system gets something wrong?
An AI marketing consultant isn’t simply someone who knows lots of AI tools. They’re someone who understands marketing well enough to know where AI belongs — and where it doesn’t.
The value isn’t in having more AI.
It’s in using it deliberately.
AI Is a Tool. Strategy Is Still the Advantage.
AI isn’t going to slow down.
There will be more tools, more automation, more capable AI systems and more ways for businesses to use them.
I don’t think the answer is to step back and wait for things to settle. They won’t.
The challenge is making sure we’re not allowing the technology to decide what the business should be doing simply because it can.
Good marketing still starts with the same questions it always has.
What are we trying to achieve? Who are we trying to reach? What problem are we solving? What makes us different? What should we automate, and what still requires human judgement?
AI changes the answers to some of those questions. It doesn’t make the questions less important.
The value isn’t in having access to more AI. It’s in knowing what to do with it.
And perhaps that’s the real shift we’re seeing in marketing.
Businesses don’t necessarily need someone to keep introducing them to the next AI tool.
They need someone who can help them decide which tools matter, where they belong, and where they don’t.
That’s strategy.
And AI hasn’t replaced it.
The challenge isn’t finding another AI tool. It’s working out where AI fits, what to automate, and how to integrate new tools without creating unnecessary complexity or risk. As a marketing consultant, I help businesses work out where AI can create genuine value, how it can be integrated into their marketing, and where human judgement still matters.