A new category attracts new noise
Whenever a new commercial category emerges, bold claims follow close behind. AI visibility, AI discovery and generative-engine optimisation are no exception. As more providers enter the space, buyers are increasingly presented with impressive-sounding figures — visibility multipliers, ranking improvements, discovery rates — that are difficult to verify and, in some cases, simply invented.
We think buyers deserve a clear way to tell the difference. The following distinctions are the ones we apply to our own work, and we would encourage any organisation evaluating a provider — including us — to apply them too.
Illustrative, real, and fabricated are three different things
An illustrative example demonstrates a method. It shows how an analysis is structured or how a report reads, using a hypothetical or sample subject, and it is clearly labelled as such. It claims no outcome. It is honest, and it is useful for showing capability before client results exist.
A real result reports a measured outcome for an actual, identifiable engagement, with enough context to understand what was measured and how. It can be substantiated.
A fabricated claim looks like a real result but is not one. It presents numbers with no verifiable basis — a specific multiplier, a precise percentage, a named benchmark — for work that was never done or cannot be evidenced. The danger is that fabricated claims and real results look almost identical on a webpage. Confidence is not evidence.
Questions worth asking any provider
A few simple questions separate substance from noise. Is a striking figure attached to a specific, describable engagement, or does it float free of any context? Are sample reports and demonstrations clearly labelled as illustrative, or are they implied to be client outcomes? If a provider claims a benchmark or a result, can they explain what was measured, and how? A credible provider will welcome these questions. Precise-sounding numbers that evaporate under them are a warning sign.
Our own position
We hold ourselves to the same standard. Where we show sample reports, we label them as illustrative. We do not present invented statistics as client outcomes, and we will not claim results we cannot stand behind. As we complete real engagements, our evidence will grow — and it will be genuine.
In a category this new, integrity is not only the right thing; it is a competitive position. The providers who last will be the ones whose claims survive scrutiny. We intend to be one of them.