Insights

Five questions to ask before AI starts recommending your competitors

If your customers are starting to ask AI for recommendations, these are the questions worth answering now, while you still have time to act on the answers.

Here is an uncomfortable thought experiment.

A customer with exactly the need your business serves opens an AI assistant and asks it to recommend an option.

What does it say?

Does it name you? Does it describe you correctly? Does it recommend you, or one of your competitors?

Most business leaders cannot answer that with confidence, which is itself the problem. The questions below are a way to find out where you stand before the answers start costing you customers you never knew you lost.

1. Does AI know we exist?

The first question is the bluntest.

When someone asks an AI assistant about your category, does your business come up at all?

Not whether it can find you if it already knows your name, but whether it surfaces you unprompted when a customer describes the kind of thing you offer.

If the answer is no, that is an important external signal: your business is missing from this tested conversation. It does not make the other readiness questions irrelevant. Discovery, understanding and recommendation can reveal different gaps at the same time.

2. Can AI explain what we do, accurately?

Being mentioned is not the same as being understood.

The second question is whether AI describes your business correctly. Does it get your products right, your customers right, the problems you solve right? Or does it blur you into a generic category, miss what makes you different, or describe an older version of the business that no longer exists?

Inaccurate description is quietly damaging. A customer acting on a wrong explanation of what you do may rule you out for the wrong reason, and you will never know it happened.

3. Does AI recommend us for the right needs?

The third question is the commercial heart of it.

When a customer describes a specific need that you serve well, does AI actually recommend you for it?

Not just mention you in passing, but put you forward as a good fit, with a reason.

This is where many businesses discover an uncomfortable gap. They are findable and roughly understood, but when it comes to the moment of recommendation, AI reaches for a competitor instead. Often for reasons that have nothing to do with which business is actually better.

4. Can AI represent us reliably?

The fourth question is about reliability.

Does AI describe you accurately, support its claims and stay consistent? Does it distinguish what is known from what is uncertain, rather than inventing details to fill a gap?

Appropriate uncertainty can be a positive sign when evidence is limited. Unsupported certainty and contradictory claims are the issues to investigate.

This is about the reliability of AI representation, not a judgement of whether the business itself is trustworthy. Look for accuracy, factual support, consistency and an honest acknowledgement of limits.

5. Can AI point a customer to the next step?

The final question looks ahead.

When a customer needs to act, can AI identify the appropriate next step for your business?

Can it point them to the right action, whether that is getting a quote, making contact or starting a purchase?

This matters more every month. AI systems are moving from recommending to assisting, and eventually to acting on a customer's behalf. The businesses with a clear, well-described next step will be ready for that shift. The ones without may be recommended and then lose the customer at the handover.

What to do with the answers

These five questions are not academic.

They map directly onto whether AI-mediated discovery works for you or against you. The awkward truth is that most businesses cannot answer them today.

The good news is that they are answerable.

You can find out what AI systems actually say about your category, your business and your customers' needs, and treat the results as evidence rather than guesswork.

Once you can see the answers, you can act on them.

The point of asking now is simple: it is far better to find the gaps yourself than to have your customers find them first, in a conversation you never saw.