You have noticed that buyers in your market are asking ChatGPT, Perplexity or Google’s AI answers which supplier to use, and you are not sure you are in the answer. There are four realistic ways to deal with that: hand it to the SEO agency you already use, buy a visibility tool, give it to someone internally, or hire a managed programme to run it.
Before going further: we sell the fourth option. Everything below is written with that interest in plain view, and in several situations the honest answer is one of the other three. Where it is, we say so.
Start with the problem, not the channel
“AI visibility” is not a problem. It is a symptom, or a measurement. The problem is usually one of these:
- A discovery problem. You are missing from, or misrepresented in, the answers buyers see. They never hear your name.
- A pipeline problem. You may be visible, but the enquiries that arrive are irrelevant, too few, or the wrong size.
- An execution problem. You already know what is wrong, perhaps from a tool or an audit, and nobody has the time to fix it.
- A measurement problem. You cannot tell whether any of the work you are already paying for produces enquiries your sales team wants.
Each option below is good at some of these and poor at others. Pick the one that matches your problem, not the one with the most persuasive demo.
Option 1: extend your existing SEO agency
What it buys you. People who already know your site, your CMS, your analytics and your history. Much of what helps you appear in AI answers is ordinary search work done well: accessible text, sound internal linking, indexing that works, structured data that matches the page. Google says plainly that its AI features need no special file or special schema. An agency that does those things competently is already doing part of the job.
Where it struggles. Three places, usually. First, measurement: rank trackers do not see AI answers, and a single screenshot of ChatGPT is noise, not data, because the same question gives different answers on different runs. Second, off-site work: getting accurately represented on the third-party pages engines read is slow relationship work that a content-production retainer often leaves out. Third, the commercial question: most SEO retainers report traffic and rankings, not which enquiries your sales team accepted.
When it is the right answer. Your agency is good, they will accept a buying-question brief rather than a keyword list, and they are willing to report on enquiries as well as positions. Ask them directly. If the answer is yes, you may not need anyone else.
If the agency wants to say yes but lacks the capacity or the measurement, that is a different arrangement: they keep the relationship and bring in execution for that one client. We run a named-client pilot for exactly that case.
Option 2: buy a visibility tool
What it buys you. Measurement at a scale you cannot do by hand: a set of prompts sampled across several engines on a schedule, charted against named competitors. The better tools show you which sources the engines cite, which is useful for deciding where to spend effort. Entry tiers are inexpensive; we keep a comparison of the main trackers with prices linked to each vendor’s own pricing page.
Some tools now go further than dashboards. Profound, for example, documents agent workflows that connect to a CMS such as Webflow and help produce content. So “tools only measure” is no longer a fair description of the whole category.
Where it struggles. A tool cannot decide what your business should be known for, verify a technical product claim, rewrite a quote form, or tell you whether the enquiries that followed were any good. It measures one layer, discovery, and that layer is several steps away from revenue. A tool also needs someone to act on what it shows. Many teams that buy one end up with an accurate chart of a problem nobody is fixing.
When it is the right answer. You have someone with the time and judgement to act on the findings, and your main gap is knowing where you stand. Or you want an independent check on whoever you hire. Before buying anything, run the measurement by hand once. It takes an afternoon and tells you whether you need a tool at all.
Option 3: build it in-house
What it buys you. The deepest knowledge of your product, your buyers and your sales process, which is exactly what makes buying-question pages useful rather than generic. Your own team can also talk to sales every week about which enquiries were worth having. No outside party can match that feedback loop.
Where it struggles. It is not one skill. Doing this properly needs research on what buyers ask, writing that a technical buyer will trust, technical work on the site, repeated measurement, and the patience to work off-site. Small teams usually have one or two of those. The work is also easy to deprioritise: it compounds slowly, and the urgent thing always wins.
When it is the right answer. You have a marketing or growth person who owns acquisition, a developer who can ship changes, and a clear backlog. In that case, an outside party adds cost more than capability. A one-off assessment to set the backlog may still be worth buying; an ongoing programme probably is not.
Option 4: a managed programme
This is what we sell, so read this section with that in mind.
What it buys you. Someone accountable for turning a commercial objective into shipped work and measured enquiries. You give us an objective (more qualified quote requests for a product family in your served regions, say), the buyer you want, a budget, and permission to execute. We map the buying questions, find where you are absent or unconvincing, assemble and ship the work, and report the enquiries your sales team accepted alongside the visibility underneath them.
Our Demand Programme is typically £4,000–£8,000 a month ex VAT. That is a fixed fee agreed in advance plus an execution allowance spent only on work you approve. Unused allowance is credited, returned or rolled forward. We split it that way so we have no reason to spend money just to spend it. The pricing page has the detail and a break-even calculator.
Where it struggles. It is the most expensive option per month, and it only works if you do your part. We need someone who can approve spend, a clear description of a good and a bad enquiry, access to the site and analytics, and honest feedback on which leads were useful. Without that feedback we can report visibility and traffic, but not outcomes, and you would be paying programme prices for a tool’s output. It also cannot guarantee anything an engine does. Google does not guarantee crawling, indexing or serving even for sites that follow its rules, and no engine sells placement in its answers.
When it is the right answer. You have a specific, valuable offer, an identifiable buyer, capacity to take more work, and a budget, but nobody inside with the time to run the whole loop. The segments where we see this most are specialist industrial suppliers, rentals and distributors and technical software and API businesses.
When it is the wrong answer. You want a guaranteed ChatGPT recommendation. You cannot say what a good enquiry looks like. You cannot implement changes on your own site. Or you will not tell us whether the leads were useful. In any of those cases we will tell you not to buy it, because a client with budget but no workable route to an outcome is not a good engagement for either side.
A simple way to decide
Ask three questions, in this order.
- Do you know where you stand? If not, find out cheaply first, by hand or with a tool. Our free demand snapshot does this for one buyer journey and says what it cannot tell you.
- Do you know what a qualified enquiry is worth? Take what a won deal contributes, multiply by how often a qualified enquiry becomes a deal, and you have it. If a programme would need an implausible number of extra enquiries to pay for itself, no option on this page is worth a monthly budget yet.
- Who will actually do the work? If it is an existing team or agency with the capacity, extend them. If it is nobody, that is the gap a managed programme fills.
If the answers are “no, roughly, and nobody”, a bounded Demand Diagnostic at £750–£1,500 is the proportionate next step. It ends in a go/no-go, and “no” is one of the possible answers.
Found this useful? Tell us what you sell and who buys it, and we will check one buyer journey for you, free.
Get a demand snapshot