The Identity Gap: Why AI Describes Your Business Wrong and What It's Costing You

The Machine Already Introduced You. It May Have Got You Wrong.

Here's an experiment worth ninety seconds of your time. Open ChatGPT, Gemini, and Google's AI answers, and ask each one, in plain words, what your company does and who it's for. Then read the three answers side by side.

If you're like most businesses we audit, you won't get one company back. You'll get two or three. A slightly wrong category. A location that's really just your registered address. Credit for a free tool or an old blog post you'd rather forget. Sometimes a founder's name attached to a stranger who happens to share it.

That's the identity gap, and it isn't new. Search engines have always built their own picture of your business from the signals you feed them. What's changed is that the picture is now read aloud. Instead of a list of blue links you could scan and correct for yourself, a buyer gets a short, confident paragraph at the top, written as fact. And the small "feedback" link that used to let someone flag a wrong answer is mostly gone. Whatever the machine decided you are tends to stand.

We think this is the most under-priced risk in Australian marketing right now. Not because AI is new, but because the cost is invisible. You can't see the buyer who asked an AI about your category, got a muddled answer, and quietly chose someone else.

How Big Is This Really? The Australian Numbers

Big enough that "we'll deal with it later" is now a decision with a price tag.

According to Roy Morgan, 13.6 million Australians (58% of people aged 14 and over) used AI tools in an average four-week period in the March quarter of 2026, with ChatGPT alone reaching 10.5 million people. Telsyte puts the figure higher again, at 17.4 million Australians, with 5.2 million now using AI every single day.

Our read: this stopped being an early-adopter story a while ago. The detail that matters most for business owners is who these users are. Roy Morgan found usage is heaviest among Australians aged 25 to 34 (74%) and 35 to 49 (72%). Those are the people who sign off on purchases, approve suppliers, and hold budgets. Roy Morgan also found that 51% of AI users work full-time, against a little over 40% of Australians overall, and they earn more, with a median personal income of $67,000 versus $53,000 for the population. The people forming an opinion of your business through a machine are disproportionately the people who can buy from you.

It's already shaping how they shop. According to PayPal Australia, 48% of Australians have used AI assistants for product searches while shopping online, rising to 66% of those under 45, and 78% expect these tools to become a mainstream part of shopping. More telling for our purposes: PayPal found that 59% of Australians have received AI shopping results that were unexpected or, in their words, downright laughable, a figure that climbs to 81% of Gen Z and 69% of Millennials.

Sit with that last number. The machine is confidently wrong often enough that most younger buyers have already watched it happen. When it's wrong about a lasagne recipe, nobody's hurt. When it's wrong about what you sell, you've been mis-introduced to the exact audience you're trying to win.

What Is the Identity Gap?

The identity gap is the distance between four versions of your business that are supposed to match but rarely do:

• what your business says it is
• what search engines file it as
• what AI cites it for
• who your buyers actually are

We steer by rankings and assume the rest follows. It doesn't. These four get decided in different rooms by different teams. Product writes one story, the sales deck tells another, the homepage says a third thing, and the blog you started three years ago for traffic quietly outvotes all of them because it's where most of your pages live. Every one of those is a signal. When they disagree, the disagreement becomes noise, and machines are very good at repeating noise back to people.

AI didn't create this gap. It inherited it from search and stripped out the buyer's ability to correct it. That's the whole problem in one sentence.

The Three Ways the Gap Shows Up

You don't need a technical audit to recognise the symptoms. You need to know what you're looking at. There are three patterns, and most businesses with a gap have more than one at once.

1. The Machine Mislabels You

This is the most literal version. The engines have the wrong idea of the business itself: the wrong category, the wrong location, sometimes a different company entirely because someone shares your name.

How to spot it: ask each AI engine plainly who your company is, then check what Google's knowledge panel (if you have one) anchors to. Does it lead with your product, or with a free tool and your blog? Do the "people also search for" suggestions point at competitors and complaints, or at the thing you actually sell? Line up ChatGPT, Gemini, and Perplexity on four points: category, location, founder, and what you sell. The wider they disagree, the deeper the mislabel sits.

Our view: this is fixable, but not by writing more content. It's fixed by making your identity unambiguous in the places machines trust, so there's only one sensible conclusion for them to reach.

2. You Attract the Wrong Crowd

Here the traffic isn't the problem. The traffic is the problem's disguise. You're pulling in plenty of visitors, but they're a different population from the people who buy.

This is the trap that catches good marketers, because the dashboard looks healthy. Sessions are up. A free calculator or a top-of-funnel guide is doing numbers. Meanwhile the questions that actually close deals never appear in your analytics, because the buyer phrases them a hundred different ways at the bottom of the funnel and no keyword tool aggregates them.

How to spot it: put the queries and pages that bring you traffic next to the reasons people give in your sales calls and your CRM. Ask one question. Do these describe the same person? If the traffic clusters around "how do I..." and the revenue clusters around "can you handle my compliance / migration / audit risk," you have an audience mismatch, and no amount of extra traffic fixes it.

Our view: this is where marketing has to leave its own lane. The people who can tell you who the buyer really is sit in sales and support, not in the analytics. If you spend all your time inside the algorithm, you lose sight of the person it's meant to reach.

3. AI Credits You for the Wrong Things

The newest of the three. The AI does recognise you, and does mention you, but for things you no longer sell or never made money from. Old content. An abandoned free tool. The reputation you've spent two years trying to outgrow.

How to spot it: ask each engine what your brand is known for and best at. Write down what it names. Beside that, list the products that actually pay the bills, ranked by revenue. The distance between the two lists is the drift. Run it more than once, on different days, because AI answers are inconsistent and one snapshot will mislead you.

Our view: this is the symptom that catches the light, because it's the one everyone can see in a chat window. But it's a symptom, not the disease. The AI is reading your accumulated signals back to you. If it credits you for the free tool, it's because the free tool is shouting louder than your product across everything you've published.

Why You Can't Fix This With a Chatbot Trick

Because the mismatch doesn't live in the chatbot. It lives upstream, in your positioning and in the signals you've been sending for years.

We understand the instinct. A new channel appears, so you want to optimise for it the way you once optimised for Google. But two things make chasing the chatbot a waste of money.

First, the answer won't hold still. Ask an AI for a brand recommendation today and tomorrow and you'll often get different lists. You can't win a position that doesn't survive two identical prompts. What you can do is make the underlying business so clear that you turn up more often, whatever the wording.

Second, ranking no longer guarantees you're the answer. According to the Pew Research Center, when an AI summary appears at the top of a Google search, people click a traditional result on just 8% of visits, down from 15% when there's no summary, and only 1% click a source cited inside the summary itself. Being on page one matters less when page one is a paragraph the buyer never scrolls past.

Here's the part that should reassure you, though. The fix is upstream, which means it's inside your control. You don't have to reverse-engineer a model. You have to decide who you are and say it consistently. That's strategy, not a hack, and it's the same work whether the buyer arrives through Google today or an AI conversation next year.

The Fix: One Source of Truth

Closing the gap starts with a document, not a campaign.

Before anyone touches a website or writes a line of content, the business, marketing, product, and sales teams have to agree on three things and write them down: who your company is, what it sells, and who it's for. Most organisations have never done this. So the argument gets re-fought on every campaign, every page, every release, and each team settles it slightly differently. That's where the noise is born, and that's what the machines pick up.

At Bushnote, we call that agreed document a source of truth, and it's the thing that keeps the four signals from drifting apart again. Once it exists, the rest of the work has a spine:

• Fix the mislabelling, so the engines stop confusing you with a free tool or a same-named stranger.
• Close the gaps where real buyer questions (migration, risk, cost, "will this actually work for someone like me") go unanswered on your site.
• Prune the content that's dragging your identity toward the traffic magnet and away from what you sell.

That pruning feels backwards, and it's the part clients resist most. You're choosing to lose some traffic on purpose, because the traffic was noise. But when you clean the signals, two things happen at once. The engines start to recognise you for what you actually are. And your real buyer starts to find you. Those turn out to be the same movement, seen from two sides.

What Closing the Gap Is Worth

Two currencies. You pay the first cost in demand that never converts: attention earned against the wrong terms while the buyer's real questions go unanswered. You pay the second in being cited for the wrong things, so the product that funds the business stays invisible in the one place a buyer is now forming their first impression.

Neither gets fixed with more content or a cleverer bot. Both get fixed by deciding who you are and making every signal agree.

The upside is that the work pays twice. A business organised around its buyer's actual problem gets found in the search someone runs today, and again in the AI conversation they have tomorrow. According to PayPal Australia, 30% of Australians say they'd only engage with AI shopping tools from brands they already trust, and 61% would trust AI to recommend a product. Trust and clarity are the currency now. The businesses AI describes accurately are the ones that decided, internally, who they are. The ones it garbles never settled the question.

You can settle it. That's the whole opportunity.