One challenge · One baseline measurement · One factual decision
Start small. Prove it fast. Then scale up.
We start with a concrete management challenge, measure the current situation and only build further when the value is demonstrable. Not a year-long program, but a decision after 45 days based on numbers.
How we start
Eight steps, from first question to scaling.
01
Decision scan
Ten questions online. Your biggest bottlenecks get named and put into words, with a first estimate of available information, feasibility and value. You get advice to sharpen, to investigate further or to scale up.
02
Conversation
An introduction, plus an explanation of what we do and how the process runs. We go through your report and test whether the bottlenecks it names are the bottlenecks you recognize.
03
Diagnosis
Where are the biggest decision bottlenecks, what information is available and where is the value demonstrable? We use the ten questions, a questionnaire and short interviews. We weigh opportunities on revenue, margin, cost, time, customer impact, risk, complexity, investment and payback time.
04
Build
We pick one application and build the tool and the AI agent for it. One, not five, because that is the only way to establish afterward what made the difference.
05
Prove
We test with real users and real management questions, and compare the outcome with the baseline measurement we took up front.
06
Decide
After about 45 days comes the evaluation. The agents collect the feedback from actual use themselves, so the assessment rests on recorded data rather than on impressions.
07
Embed
Once the value is proven, we anchor the application in the processes and in the management rhythm, and we monitor the indicators and the governance.
08
Scale up
New tools and agents run on the same digital foundation. From about five applications onward, a Digital Business Twin becomes possible, and with it an overarching agent: the sixth seat itself.
A sample of the diagnostic questions
Six questions where it usually gets stuck.
- Where does margin pressure arise, and which customers, products or projects explain it?
- Which sales opportunities deserve priority based on likelihood of conversion, margin and capacity?
- Which projects are at risk of delay, hours overruns or cash flow pressure?
- Where does growth become hard to manage operationally?
- Which knowledge is vulnerable because it depends on a few key people?
- Which information is missing at exactly the moment a decision has to be made?
The evaluation after 45 days
What we measure is not AI usage, but business results.
We take a baseline measurement up front, build the application and then measure what actually changes in decision making and execution. Three things are central.
Decision lead time
How fast does a signal move from analysis to discussion to a decision taken?
Validated value opportunities
Which margin leaks, missed revenue, avoidable costs or risks become demonstrably visible?
Decision quality and governance
Are analyses traceable and verifiable, and have people explicitly reviewed them?
- In addition we measure time saved, process lead time, conversion, customer value, error reduction, forecast quality and the dependence on key people.
- After the evaluation there are three possible outcomes: stop, sharpen or scale up. All three are good outcomes, because all three are backed by evidence.