Most of us have been there. Something has stopped working, or is getting close to it, and you need to find a replacement. The question is always the same: what exactly do I need, and how much should I expect to pay?
For anything with a model number, a part code or a specification — and that covers an enormous range of things — AI is very good at answering this. Take a photo of what you have, describe your situation and ask what the right replacement is. You will get a clear answer and a typical price range. From there, if you cannot find what was recommended at the right price, you can ask for alternatives. Sometimes those alternatives turn out to be the better choice.
That is exactly what happened to me recently, with a car battery.
I drive a Land Rover Defender 90 TD5 and the battery was getting tired. I did not know what type I needed, so I asked ChatGPT. I told it the make, model and year of the vehicle and asked what battery would be the best fit.
It recommended a UK644 battery, explained why it suited that vehicle and gave me a price range to expect. Armed with that, I rang around a few suppliers.
None of them could supply a UK644 anywhere close to the suggested price. Every quote came back £40 to £70 over what ChatGPT had said was reasonable. Some could not source it at all.
So I went back to ChatGPT. I explained the situation — that the UK644 was proving difficult to find at a fair price — and asked if there were any alternatives that would work just as well in my vehicle.
It came back with the UK633. This is the standard battery for a Land Rover 110 TD5, a slightly different model from mine. What ChatGPT explained was that the UK633 is actually a heavy-duty battery, built for the extra demands of the larger 110. It would be better for longevity, not just a like-for-like swap.
When I rang the same suppliers and asked for a UK633, everything changed. It was widely available, came in £40 cheaper than the UK644 had been quoted, and turned out to be one of the most commonly produced batteries across a wide range of vehicles — which is exactly why it costs less. Higher production volumes bring the price down.
It fitted my Defender 90 perfectly. I came away with a better battery than I had originally been looking for, at a price well below what I had been quoted for the one I thought I needed.
The part that made the real difference was not the first question — it was the follow-up. Most people would have accepted the expensive quotes and assumed that was simply the going rate. Asking AI to find a different route to the same destination opened up an option I would never have found on my own.
This is one of the most useful things AI can do in any buying situation. It knows that different part numbers, different specifications and different brand names can often solve the same problem. It can also explain why a particular alternative might actually be the better choice — not just a cheaper substitute.
The battery example is a good one because it shows how the market works. A component that fits a common, high-volume vehicle will almost always cost less than one for a niche or older model, even if the two are physically interchangeable. AI knows this. Asking it to think laterally can put that knowledge to work for you.
The same approach applies to almost anything you need to replace.
If your washing machine or dishwasher needs a spare part — a door seal, a pump, a belt — photograph the old one and upload it to ChatGPT. It will often be able to read the part number from the photo and tell you what it is compatible with or where to find a good replacement.
If a household appliance is coming to the end of its life and you want to find a like-for-like replacement, show AI a photo and ask. It can tell you the key specifications to look for and suggest options across a range of price points.
If you are replacing something on a budget — a light fitting, a garden tool, a piece of tech kit — ask AI whether a less well-known version of the same thing would perform just as well and cost significantly less. It often will.
The pattern is the same every time. Take a photo, describe the situation, ask for the right replacement and a fair price guide. If the price turns out to be a problem, ask for alternatives.
Open ChatGPT on your phone or computer. Start a new chat. Tap the paperclip or photo icon and upload a clear photo of the item you need to replace — the closer and clearer the better, especially if there is a label, serial number or part code visible.
Then type your question. Something like this works well for most situations:
"I need to replace this [describe what it is]. Can you tell me the correct replacement, what to look for and what a reasonable price range should be?"
If you are replacing a car part or vehicle component, add the vehicle details:
"This is from my [make, model and year]. What is the right replacement part and what should I expect to pay for it?"
If the recommended item proves hard to find at a fair price:
"The [item it recommended] is proving difficult to source at a reasonable price. Are there alternatives that would do the same job — and is there any reason to prefer one over another?"
That last question is the one worth remembering. It is what turned a frustrating search into a better outcome than I had started with.
AI gives you a guide price based on what it knows, and that knowledge has a cut-off date, so treat the range it gives you as a sense check rather than a guarantee. Prices change and vary by supplier. If every quote comes in significantly above what AI suggested, it is worth questioning — either the market has moved, or you are looking at the wrong type of supplier for what you need.
In my case, the price guide turned out to be accurate. The issue was simply that I had been looking for the wrong part. AI found me the right one.
Kevin also runs free library presentations across Norfolk — a good chance to see AI in action, ask about anything covered in the blog and find out what it could do for you.
If you would like to go further, Kevin's book covers exactly this kind of practical AI use in detail — written in plain English, full of real examples and designed for people who want genuinely useful results rather than just theory.
See upcoming events Find out about the book