Why not just decide this ourselves?
Some firms can, and if that is you, you do not need me. In practice, every initiative inside a business has someone who chose it, and sunk cost has a way of writing its own review. What I add is not intelligence your team lacks. It is distance: no history with any of the decisions, no tool I want to see survive, and the same standard of evidence applied to everything. It also changes the conversation outside the building. A vendor negotiating a renewal argues differently against an independent review than against an internal opinion.
Can you help us decide where to invest in AI next?
Yes. The same standard applies. I start with a business outcome, workflow or constraint rather than a catalogue of AI use cases. I establish the baseline, identify where better performance would matter, test whether the improvement is plausible and economically worthwhile, then recommend investing, deferring or rejecting the opportunity.
Why is there no day rate?
Because you are not buying my days, you are buying a decision and what follows from it. A day rate pays me for taking longer and pushes you to buy fewer days than the question deserves, so we would both be optimising the wrong thing. The fee is fixed before we start. If the work takes me longer than I planned, that is my problem, not your invoice.
How do you arrive at the fee?
From the value of the decision, not the hours behind it. In the first conversation we establish what the outcome is worth to the business: cost removed, capacity redeployed, speed, quality or throughput gained, revenue or customer value increased, useful capability created or a poor investment avoided. The fee is set as a fraction of that. If the value is not comfortably larger than the fee, I will say so and we will not proceed.
What if our existing AI is already working?
Working is the starting point, not the conclusion. The evidence may support maintaining it as it is, improving the value it creates, expanding it into another workflow or using what you have learned to test an adjacent opportunity. A successful initiative can still leave material value on the table.
What access do you need, and how is sensitive information handled?
Typically the relevant people, your current tool and project costs, selected usage data and screen-shared walkthroughs of the workflows in question. Access is limited to what the agreed scope requires, and confidentiality terms are settled before anything starts.
What makes the advice independent?
I receive no software commission, hold no reseller agreements and carry no licence quota. The only thing I am selling is the quality of the decision, which means the evidence can take me wherever it goes.
Who implements the recommendations?
Your existing partner or a specialist. I hold them to the agreed measure. I stay on the decision. I do not pick up a delivery quota.
What if the honest answer is that we wasted the money?
Then that is the answer you get, in writing, with the evidence behind it, and you stop paying for it from that month. Stopping is a legitimate outcome and often the fastest to pay back.
