AI has created a genuine opportunity for growing businesses. Used properly, it can reduce the time it takes to work through a problem and give a smaller company access to capabilities that previously required considerably more time or resource.
Tools such as ChatGPT and Claude can help with research, analysis, planning, content, documentation and a growing list of everyday tasks. I use AI regularly and see real value in it.
But it is also very easy to mistake a fast answer for a good decision.
A polished answer is not necessarily a good decision
An AI model can only work with the information and context it has been given. If the question is slightly wrong, the answer can take you in the wrong direction. If important information is missing, the output may look convincing while failing to address what is really happening inside the business.
That can lead to a lot of time disappearing down rabbit holes. You refine the prompt, add more information and ask the question in different ways. The output becomes increasingly detailed, but it may still not deal with the actual commercial problem.
The difficulty is not always getting an answer. It is knowing whether you are asking the right question in the first place.
Businesses rarely operate as neatly as the information suggests
A spreadsheet may show that revenue is growing. It may not show that the growth is coming from lower-margin work.
A customer contract may appear profitable. It may not show the additional support, rework and informal promises being absorbed by the delivery team.
A business may want to create more recurring revenue. The answer is not simply to produce a list of possible subscription services. You also need to understand what customers will value, what the business can deliver, how it should be priced and whether the sales team can explain it properly.
None of that means the AI output is useless. It means the output needs to be put back into the reality of how the business operates.
This is where experience earns its keep
Someone who has built, managed and advised businesses will often recognise what is missing from the information. They can challenge the assumptions, ask the uncomfortable questions and understand how a decision in one part of the business will affect another.
AI can produce twenty ideas for increasing recurring revenue. Experience asks why customers are not already buying it, whether the delivery team can support it, what margin it should produce and how it will be sold without creating another unprofitable obligation.
The useful contribution is not simply knowing how to prompt the technology. It is knowing what to challenge when the response comes back.
The real opportunity is the combination
I do not see this as a choice between AI and experienced advice. The strongest approach combines the two.
Experience helps establish what the real problem is, what information matters and which assumptions need to be challenged. AI can then accelerate parts of the work. It can organise information, identify patterns, compare options, test ideas and produce clear working material far more efficiently.
Experience comes back into the process to challenge the output, apply commercial judgement and decide what is practical.
A practical working model
How the combination should work
- Understand what is actually happening in the business.
- Identify the question that really needs answering.
- Use AI to speed up the research, organisation and analysis.
- Test the output against commercial reality.
- Turn the useful parts into a clear course of action.
AI provides the speed. Experience provides the context and judgement.
Better advice should not mean more days
Traditional advisory work can involve a considerable amount of time gathering information, collating data, preparing documentation and turning it into something management can review.
Some of that work still needs to happen, but AI can make parts of it considerably more efficient. That should mean less time being charged for producing and formatting material, and more time being spent understanding the problem, challenging the thinking and helping management decide what to do.
For a growing business that may not need, or want, another full-time senior person, this creates a useful middle ground. The business gets access to experienced commercial thinking without taking on permanent headcount. It also benefits from the speed of AI, but with someone taking responsibility for interpreting the output and applying it to the real situation.
The aim is not to sell an AI project. It is to use the technology to make experienced commercial support more focused, efficient and useful.
AI is a tool, not the person responsible for the result
AI will continue to improve and become part of how most businesses operate. Ignoring it would make little sense. But expecting it to understand every aspect of a business, identify every missing piece of information and make a commercially sound decision without experienced input would be equally unwise.
The value comes from knowing where AI can help, where it needs to be challenged and where human judgement still needs to lead.
Used together, AI and real-world experience can help a growing business reach better decisions more quickly, without turning every question into a lengthy and expensive consulting project.
For related reading, see why profit does not always follow revenue growth and seven signs the commercial model has not kept pace .