Real systems, real evidence. Aviators, not autopilot.

How a DragonAI engagement actually runs

Nothing here starts with a tool. It starts with a scored read of where the business is, because until that exists every AI decision is a matter of opinion and whoever is most confident in the room wins.

One — The read

We run your company and its industry against 58 defined metrics across eight weighted domains. You get a score, the reasoning behind each domain, and a list of what we could not establish. Every figure carries a label saying how it is known: your own data, a named public source, a derivation we show you, or an industry assumption. The proportion that is assumption is printed on the report rather than buried in it.

Two — The shortlist

From the read we produce a short, costed, ordered list of what is worth doing with AI and what is not. Usually it is fewer things than clients expect, and the list of what to leave alone is the part that saves the most money. This is AI consulting, and it is deliberately separate from building anything.

Three — The build, the training, or both

Some items on the shortlist are systems, and those become agentic workflows with a human at the approval point. Others are not systems at all, they are people who need to get better at a tool they already have, and those become AI Aviators training sessions run by function. Most engagements need both, and knowing which is which is the whole point of doing the read first.

Four — The review

A quarter later we run the read again and compare. Scores that moved, scores that did not, workflows still running, workflows quietly abandoned. Because the framework is fixed, the second read is directly comparable to the first, which is the only honest way to answer whether any of it worked.

What DragonAI is not

It is not a software product you license, and there is no seat to buy. It is not a resale arrangement: we take no commission from any AI vendor, so the recommendation to keep what you have and train your people on it costs us the same as the recommendation to build something. And it is not a replacement for judgment. The analysis narrows the field and shows its working; the decision is still yours, and the aircraft still needs a pilot.

DragonAI sits alongside two other pillars. DragonONE is the strategy, brand and creative foundation; DragonIQ is the demand and visibility layer. AI is applied inside both rather than sold as a thing of its own.