AI consultancy that ends in a decision, not a slide deck.

Stakeholder interviews with the people doing the work rather than only the people describing it. An audit of the data you already hold and what state it is in. A map of the opportunities ranked by return rather than by novelty. And a roadmap from proof of concept through to production, with prices and timelines attached.

The output is a written document, not a presentation. It is specific enough to build from, and it is yours to take to any supplier, including one that is not us.

The pilot is the easy part. What kills projects is everything after it: integration with the systems people actually use, cost that scales unpredictably with usage, and nobody owning the thing once the initial enthusiasm fades.

A demo runs on one laptop with one person watching and forgiving. Production runs unattended, on real data, for people who will not tolerate being told the wrong answer. The gap between those two states is where budgets are lost, and it is the part a strategy exercise almost never covers.

Ask most teams what one customer costs them to serve on an AI feature for a month. Very few can answer, and the ones who can usually answer with an average, which is not the same thing.

We instrument cost down to the individual request, attributed to a customer, a feature and a model. We route routine work to economical models and reserve expensive ones for work that needs them. And we enforce spending ceilings in real time, tested by watching them stop something in production rather than assuming they would.

A deck describes what could be done. This produces a scope specific enough to build from, with numbers attached to each option and a recommendation on what to do first.

If the honest answer is that nothing is worth building yet, that is a valid outcome and we write it down plainly. Not every problem needs AI, and a supplier who has never said so is not being straight with you.

What does AI consultancy cost?

Augustova prices a fixed scope audit at £2,500 to £6,000, delivered in one to two weeks. That covers stakeholder interviews, a data audit, an opportunity map ranked by return, an ROI model and a written roadmap you keep regardless of who builds it.

What do we receive at the end?

A written report and roadmap, an ROI model for each opportunity, a prioritised backlog and a recommendation on what to do first. It is specific enough to build from and yours to take to any supplier.

Can you tell us what our current AI features cost per user?

Yes, and most firms cannot. We instrument cost per request attributed to a customer, a feature and a model, then set spending ceilings that have been watched firing in production rather than assumed to work.

What if AI is not the right answer for us?

Then we say so and write down why. Plenty of problems are better solved with ordinary software, a process change, or nothing at all. The audit output is yours either way, so an honest no still leaves you with something useful.