A simple monthly routine for Christchurch business owners: ten realistic prompts, three tools, one spreadsheet, and what to do if your business is not being named.
Write eight to ten prompts your customers might type, run them in ChatGPT, Perplexity and Google, and record which businesses get named in a simple sheet. Repeat monthly. Treat the results as a sample, not a measurement, because answers change between runs, accounts and locations. The point is to spot patterns, not chase a score.
You can check whether AI tools mention your business in about half an hour a month. Write a set of realistic customer questions, ask each one in ChatGPT, Perplexity and Google, and note who gets named. It will not give you a precise score, but it will show you where you stand and what to fix.
Start with a list of 8-10 questions a customer would genuinely type. Mix the styles, because people do not all search the same way:
Say you run a plumbing business in Hornby. Your list might include "blocked drain plumber Hornby", "best plumber in Christchurch for a bathroom reno", "how much does a plumber charge per hour in Christchurch" and a direct question using your trading name. Keep the wording natural, and avoid the common mistake of asking only branded questions, because most new customers do not know your name yet. Do not write prompts that flatter you, such as naming your own services in unusual ways, or the results will tell you nothing.
Results can be shaped by your location, your account and any past conversations. Where you can, use a private window or logged-out session, and write down the date, the tool and roughly where you were. Do not expect to remove every variable. The goal is consistency from month to month, so run the check the same way each time.
Ask all the prompts in ChatGPT, Perplexity and Google. In Google, look at the normal results, the local pack and any AI Overview that appears. Not every search shows one, and Google's documentation on AI features is the place to check what it says about how they appear. Do not paste long prompts or give the tools private information. Just ask the questions as a customer would.
If a tool asks follow-up questions or gives different answers when you rerun it, that is normal. Run each prompt once, and again if the first answer looks odd.
Make a simple sheet with columns for date, tool, prompt, businesses named, whether you were named, and any sources or links the tool showed. Add a notes column for oddities, such as a wrong phone number or a business in the wrong suburb. Five minutes of tidy recording is worth more than a vague memory of "I think it mentioned someone". Screenshot anything surprising, because the same prompt may give a different answer an hour later, and you will want proof of what you saw. Save each month's sheet rather than overwriting it, so you can compare trends over a few months instead of reacting to one odd result.
After the first round, look for patterns rather than individual answers:
Say your Hornby plumber finds that Perplexity names three competitors for blocked drains but never mentions the business, while Google's results show it in the local pack. That is useful: it suggests the Business Profile is doing its job and the gap is elsewhere, perhaps thin service pages or few mentions on other sites.
Be careful with conclusions. Because answers vary by run, you are sampling, not measuring. One missed mention proves little. The same gap across several months says more.
If you are not named, or details are wrong, work through the basics rather than hunting for tricks:
None of this guarantees a mention. For the wider picture, see our overview of GEO and practical visibility tips. Then re-run the same prompts next month and compare.
It will not tell you how many people see AI answers, it will not rank you, and it will not explain why a particular tool chose its sources. Nobody outside the AI companies can say that precisely. What it gives you is an honest, repeatable view of how your business appears to a customer asking today, which is more than most owners have.
If you would like help reading your results and deciding what to fix first, our GEO work for Christchurch businesses starts from exactly this kind of check, or you can request a free SEO audit and we will look at your basics first.
Not honestly. Answers can differ between runs, accounts, locations and over time, so any single result is a snapshot. Treat your monthly check as a rough sample that shows patterns, such as which competitors keep appearing, rather than a number you can compare precisely.
Start by checking that your own website, Google Business Profile and main listings all say the same thing. Wrong details often trace back to old or conflicting information somewhere online. Fix the sources you control and keep your site clear, but there is no way to force an AI tool to correct itself.
No. For a small business a spreadsheet and thirty minutes a month is enough. Paid tracking tools exist, but with answers varying so much, a simple manual routine you actually keep doing beats an expensive dashboard you ignore.
It is a useful finding, not a verdict. Many small local businesses are not named yet. It usually points back to the fundamentals: clear service pages, consistent details, genuine reviews and mentions elsewhere. Nobody can guarantee a citation, so aim to be easy to understand and verify.
Get a free, no-obligation review of your Christchurch business's SEO - what's working, what isn't, and what to fix first.