Working together
Depending on where you stand in the sequence: a decision to settle in the coming days, a choice of solution to investigate, a pilot to run through to the industrialisation decision, or a multi-year commitment to secure over time.
Format 1 · ahead of phase 1
You have a committee, a signature or a board meeting ahead of you, and a file already prepared by others. I read the file, trace back to what was actually measured, and tell you what the demo does not prove.
What you receive
A short note, readable in committee: what is established, what is not, and the questions to ask before signing.
Order of duration
A few working days, up to a few weeks if the file is substantial.
Format 2 · the whole of phase 1
The five steps of the upstream phase, with your domain experts. We build the evaluation set, put the candidates through an identical protocol, and you leave with the selection decision and the tooling.
What you receive
The evaluation set, the completed scorecard, the protocol you can rerun, and the decision note.
Order of duration
A few weeks, paced by the availability of your experts.
Format 3 · the whole of phase 2
The solution has been chosen and you now need to know whether it holds up in your organisation. Framing with indicators fixed before the start, shared diagnosis, a pilot on a bounded scope, measurement, then a reasoned decision to industrialise, adjust or stop.
What you receive
The framing note, the diagnosis, the pilot protocol, the scorecard held against the indicators, and the decision note.
Order of duration
A few weeks of preparation, then the pilot itself, bounded at framing.
Format 4 · recurring
The commitment runs over several years and the technology keeps moving. We rerun the campaign at regular intervals on the same set, to measure real progress, catch regressions and keep a negotiating position.
What you receive
A campaign report at each iteration, the comparative trend and the regressions identified.
Order of duration
Campaigns spaced out in line with the vendors' release cycle.
Formats 2 and 3 follow on naturally, but nothing requires taking them together. Many organisations arrive with the solution already chosen and need only the third. Scope and terms are set on the first call. If your need falls outside these four formats, I say so, and point you elsewhere if I know someone more relevant.
The sequence is the same whichever phase applies. The milestones are not calendar dates, they are handovers.
What you are trying to decide, by when, where you stand in the sequence. I tell you whether the method applies, on which phase, and what it requires from your side. This call is free and carries no obligation.
One page: scope, candidates or hypothesis to test, decision criteria fixed from the start, experts to involve, schedule and terms. Nothing begins before that document is signed off on your side.
The densest part. Building the case set and ground truth in phase 1, shared diagnosis and pilot protocol in phase 2. Your experts do the thinking, I structure and facilitate.
Candidates run through the protocol, or the pilot runs on the bounded scope. Blind scoring wherever possible, then consolidation. You see the raw results, not just the summary.
The decision note, presented to whoever you wish, including the committee that decides. Then the handover: the scorecard, the set and the protocol stay with you, with whatever session it takes for your teams to be able to rerun them.
What is needed from you
What I do not do
An evaluator who sells the solution being evaluated is not evaluating anything. That is the condition on which this work has any value.
Confidentiality
Your cases, your data, your results and the names of the candidates evaluated stay with you. Nothing produced in an engagement is published, cited as a reference, or turned into an example on this site. A confidentiality agreement is signed if you want one.
No. They answer two different questions and can be taken separately. If the solution is already chosen, phase 1 has no purpose and we start directly with the pilot framing. Conversely, a market evaluation can conclude without any pilot following.
No, and it is often better if they are not. The shortlist is built at the first step, once maturity has been qualified. Arriving with three names already settled sometimes means missing the simplest option, which is occasionally to buy nothing.
Cases reconstructed by your experts are better than nothing, and far better than the vendor's examples. The set can also be built on an anonymised sample. We look at this at framing: it is a question better handled before starting than after.
Beyond five or six, the scoring effort becomes disproportionate and quality drops. The first step exists precisely to bring a long list down to a defensible one.
Then the recommendation is not to buy, argued like any other. It is one of the outcomes provided for from the outset, on the same footing as industrialisation. A process that could not conclude in favour of stopping would not be an evaluation.
No, and it is a common situation. We go back to framing, retrospectively: what were the implicit indicators, what is actually known today, and what would need measuring to be able to decide. The aim is to get out of the middle ground, one way or the other.
Yes, with time on site for the expert workshops and the readout, which gain a great deal from being held in person. I am based in the south of France and I travel.
French and English, including workshop facilitation and written deliverables. The method itself is language independent: what matters is that the evaluation set is written in the language of the people who will use the solution.
The first call is there to qualify, not to sell. If there is nothing there, I tell you on the call, and you leave with the scorecard anyway.
