01
Business Value Fit
Is AI or organizational change being applied where there is meaningful business value?
Looks at
expected outcomeseconomic valueprioritizationvalue leakagewhether effort is concentrated on the right problems
How it works
From business question to decision-ready diagnosis.
Every GrundMind engagement starts with a specific question leadership needs to answer. We collect evidence from the people, workflows and organizational context behind that question, diagnose where value is being created, lost or blocked, and translate the findings into a prioritized roadmap for action.
The output is not another survey score. It is a structured diagnosis of what is happening, why it is happening, and what should change next.
01 — Define the question
We do not start with a survey.
We start with the decision.
Before collecting evidence, we define what leadership needs to understand.
The question
What does leadership need to understand?
Shapes the diagnostic
- 01Why is AI investment not producing expected value?
- 02Where are workflows creating unnecessary friction?
- 03Is the organization ready for a new AI investment?
- 04Why is a transformation initiative stalling?
- 05Where are managers and employees seeing the organization differently?
- 06What should a newly appointed executive address first?
- 07Where is value leaking after an acquisition?
- 08Is the organization operationally ready for new AI governance requirements?
02 — Context
Understand how the work actually happens.
The diagnostic is anchored in the real operating environment rather than abstract maturity questions. Depending on the engagement, GrundMind may examine:
workflows and handoffs
roles and responsibilities
AI tools and current usage
decision points
repetitive work
approval bottlenecks
governance constraints
existing processes and documentation
sources of rework, delay or duplication
business priorities and expected outcomes
AI adoption, transformation and organizational performance cannot be diagnosed in isolation from the work itself.
The objective is to understand the system people are operating inside before interpreting their responses.
03 — Evidence
Collect evidence without averaging disagreement away.
GrundMind gathers role-appropriate evidence from employees, managers and leadership. Different perspectives remain separate. We do not assume that the manager's view is correct. We do not assume that the team's view is correct.
And we do not blend contradictory evidence into one comfortable average.Disagreement is often one of the most useful diagnostic signals.
Stream 01
Employee evidence
How the work is experienced day to day
Stream 02
Manager evidence
How the work is understood from above
Stream 03
Workflow context
How the work is actually structured
Stream 04
Organizational context
Priorities, constraints and governance
Kept separate where necessary.
Compared where useful.
Combined only where the evidence supports it.
04 — Diagnose
Look at the organization through five connected lenses.
The lenses are layers around the same business outcome, not five independent maturity scores.
02
Workflow Fit
Does the way work is structured allow the technology or intervention to create value?
Looks at
handoffsduplicationreworkbottlenecksapprovalsworkflow frictiontasks that should be redesigned before being automated
03
Cognitive Fit
How well does the way people interact with uncertainty, control and AI fit the work being asked of them?
Looks at
human-AI interaction styleambiguity tolerancetrust calibrationcontrol preferenceverification behaviorcognitive friction
The four Cognitive Fit archetypes
Co-ThinkerCalibratorControllerAI-Averse
These are not rankings of ability or intelligence. They describe different interaction patterns that require different working conditions.
04
Governance & Risk Fit
Can people use AI or make changes within clear, safe and workable boundaries?
Looks at
policiesaccountabilityapproved toolsdecision rightsdata boundarieshuman oversightoperational clarity
05
Role & Capability Fit
Do people have the role clarity, capability and support required to perform in the future workflow?
Looks at
skillsrole expectationsaccesstrainingownershipsupport structurescapability gaps
05 — Evidence quality
Not every answer deserves a score.
GrundMind uses evidence thresholds before producing quantitative conclusions. If there is not enough evidence, the result is not treated as zero. Instead, the diagnostic distinguishes between sufficient, partial and insufficient evidence. Dimensions without enough evidence are suppressed rather than converted into misleading scores.
Common practice
No evidence = 0
Silence becomes a bad score, and leadership acts on a number that was never measured.
GrundMind
No evidence = insufficient evidence
The dimension is reported as unresolved, with what would be needed to resolve it.
Missing evidence is itself information. It is never silently converted into poor performance.
06 — Interpretation
Find where value is being created, lost or blocked.
The diagnosis connects patterns across the five lenses. Examples might include:
01strong AI usage but weak business value
02capable employees operating inside a poor workflow
03manager-team perception gaps
04governance that is blocking legitimate use
05high-value opportunities with low implementation readiness
06training investment applied to a workflow problem
07AI added where automation or process redesign would create more value
08cognitive mismatch creating unnecessary verification or avoidance
09unclear ownership causing value leakage
This is where GrundMind moves from measurement to diagnosis.
A symptom tells you what is visible.
A diagnosis tells you what to change.
07 — Review
AI can help analyze evidence.
It does not make the final judgment.
GrundMind combines structured analysis with human review. Automated analysis can identify patterns, contradictions and areas requiring investigation, but published interpretations and recommendations are reviewed before they reach leadership.
- 01Evidence
What participants and organizational sources tell us
- 02Analysis
Patterns, gaps and relationships identified across the evidence
- 03Interpretation
What those findings may mean in the specific organizational context
- 04Recommendation
What leadership should consider changing
These layers are intentionally kept separate.
08 — Output
A decision-ready result.
The final output is designed to help leadership act, not simply understand the diagnosis. Depending on scope, outputs can include:
executive diagnosispriority gapsValue Leakage analysisCognitive Fit findingsworkflow findingsmanager-team alignment findingsorganizational risksAI opportunity assessmentprioritized recommendationstarget operating changesownerstimingmeasurable success conditions30 / 60 / 90-day roadmapexecutive readout
The output is not a score.
It is a roadmap for what to change next.
Typical pilot
Start contained. Learn quickly.
1 team
Typical initial scope
All relevant team members
Participants
~1 hour
Per participant for the full diagnostic
~1 week
To executive readout for a typical contained pilot
- 01Understand the work
- 02Hear both perspectives
- 03Diagnose the evidence
- 04Review the findings
- 05Deliver the roadmap
The exact scope is customized around the question being answered.
09 — Trust
People need to be able to tell us what is actually happening.
The quality of an organizational diagnosis depends on honest evidence. GrundMind therefore separates individual, team, manager and organizational reporting layers and applies minimum evidence thresholds before reporting findings.
confidential participant process
secure invitation links
data minimization
controlled access to source responses
reporting thresholds
minimum team reporting group sizes
evidence-aware suppression
immutable reporting snapshots
separation of manager and employee evidence
human review before publication
no use of Cognitive Fit results for employment decisions
Start the conversation
What do you need to understand?
Tell us the decision, problem or uncertainty you are facing. We will shape the diagnostic around the evidence needed to answer it.