01
Detection
Client + People PotentialSelects which tensions are genuinely live and where the friction concentrates.
Illustrative demonstrationFictional organisation · synthetic data · composite employee voices
PEOPLE POTENTIALCONVERSATIONAL INTELLIGENCE · DETECTION LAYER
Your organisation is already speaking. Can you hear the signal beneath the noise?
Every day, people signal where work is becoming difficult, trust is thinning, decisions stall and energy is being lost. The challenge is not collecting more noise. It is knowing where to look and starting the conversations that bring the real issue into view.
A People Potential product · Extraction powered by Haven, a FeedbackRocket innovation
ONE ENGINE · FOUR LAYERS
Measurement becomes meaning. Meaning becomes a defensible decision.
THE TENSION LENS
One method. Configured around the live question.
Detection decides where to look. Focused conversation extracts lived experience. Identification names the tension and Recommendation presents practical next options.
Download the brochure02
Extraction
Haven · a FeedbackRocket innovationRuns focused, adaptive conversations aimed at each participant’s own signal.
Experience Haven for yourself03
Identification
FeedbackRocket + People PotentialAttributes patterns, emotional register and source layer; distinguishes process from people.
04
Recommendation
People PotentialTurns verdicts into sequenced options with named AI–HI modes and success metrics.
WHAT THE DETECTION LAYER IS
It sets the scene for the conversations that matter.
It does not diagnose or identify the tension. It frames the context, decides where to look and creates focused conversation starters around the live organisational challenge.
Extraction
Focused, adaptive conversations reveal lived experience: pressure, hesitation, rework, avoidance and what remains unspoken.
Identification
Evidence identifies the tension, where it sits and the critical question.
Recommendation
The evidence distinguishes a process problem from a feeling, then proposes options rather than presuming the intervention.
THE FULL TENSION LENS
From the signal people live with to the action they can agree.
The Lens does not assume the intervention. It creates a disciplined route from an organisational challenge to evidence, shared understanding and practical next options.
Detection
Set the scene. Configure the context and conversation starters around the challenge.
Extraction
Have focused, adaptive conversations and hear the lived experience beneath the noise.
Identification
Name the tension, where it sits - strategy, tactics or operations - and the critical question.
Distinguish
Is it a process problem requiring redesign, or a feeling such as fear, mistrust or overwhelm requiring a different response?
Recommend
Present evidence-based options. The recommendation is the proposal, not the build.
Learn
Use Tension Labs and Haven Pulse to implement, re-check and see whether the tension is shifting.
ONE PRACTICAL USE CASE · AI-HI INTEGRATION
AI adoption without integration is cost, not investment.
Lumenbridge Advisory shows how the Detection Layer can be applied when the tension is known: AI is being adopted, but the work, decisions and human experience are not yet integrating around it.
A lens—not a wholesale rebuild
First-principles thinking gives people permission to look beyond inherited processes without assuming the organisation must start again. We explore what could be possible, then compare it with the existing business and its real strengths, constraints, readiness and appetite for change.
Imagine
Temporarily set aside today's process. What might purpose-led work look like with AI available?
Compare
What already works, what creates friction, and what exists for good reasons?
Choose
What should we keep, improve, redesign, stop or explore before committing?
Sequence
What can change now, what needs evidence, and what should be considered later?
AI–HI FIRST-PRINCIPLES REVIEW
Imagine what could be. Compare it with what is. Decide what can change.
A facilitated working session for leaders and people close to the work. Together, we explore the possibilities AI creates, surface inherited assumptions and compare the future opportunity with how the organisation works today.
The output is not a target-state blueprint or a list of AI tools. It is a practical opportunity map showing what to protect, what may be ready to change, what needs evidence and what belongs on a longer horizon.
THE REVIEW PRODUCES
- Strengths, sound controls and valuable practices to keep
- Immediate improvements that are realistic and worthwhile
- Processes to redesign before AI accelerates them
- Experiments and deeper changes requiring further evidence
ONE ENGINE · FOUR LAYERS
Measurement becomes meaning. Meaning becomes a defensible decision.
THE TENSION LENS
One method. Configured around the live question.
Detection decides where to look. Focused conversation extracts lived experience. Identification names the tension and Recommendation presents practical next options.
Download the brochure01
Detection
Client + People PotentialSelects which tensions are genuinely live and where the friction concentrates.
02
Extraction
Haven · a FeedbackRocket innovationRuns focused, adaptive conversations aimed at each participant’s own signal.
03
Identification
FeedbackRocket + People PotentialAttributes patterns, emotional register and source layer; distinguishes process from people.
04
Recommendation
People PotentialTurns verdicts into sequenced options with named AI–HI modes and success metrics.
STRUCTURED MEASUREMENT
Scale and distribution
- Which tensions are present?
- Where do signals concentrate?
- How do Strategy, Tactics and Operations differ?
ADAPTIVE CONVERSATION
Lived experience and cause
- Where does the work actually come apart?
- What is felt, withheld or contradicted?
- Is the origin process, people - or both?
ONE EVIDENCE BASE
Defensible action
This is not numbers versus opinion. It is two empirical views of the same organisational reality.
WHERE THE FRICTION ORIGINATES
Strategy, Tactics and Operations are three connected layers.
We locate the source of a tension so the response addresses its cause. A problem experienced in daily work may have been created by a management practice or a strategic contradiction further upstream.
Strategy
Why, where and toward what?The organisation's purpose, value proposition, priorities, choices, risk appetite and boundaries. Strategy decides what matters and what the organisation will not do.
Examples: growth choices, client promise, AI ambition, investment priorities.Tactics
How is the organisation designed to deliver?The structures and management system that translate strategy into coordinated action: decision rights, policies, resource allocation, leadership practice and controls.
Examples: approval thresholds, governance, team design, performance signals.Operations
How does work actually happen each day?The lived workflow: roles, tools, handoffs, routines, exceptions and the practical experience of employees and clients as value is created.
Examples: case handling, reviews, client delivery, daily use of AI tools.Why attribution matters
If an operational symptom comes from a tactical or strategic cause, adding another tool or training course will not solve it. It may simply automate the problem.
EXECUTIVE SNAPSHOT
Strong intent. Uneven integration.
Scores are illustrative and shown consistently out of 100. Higher scores indicate a stronger positive experience in this section.
Professional pride
82/100Client commitment
79/100Willingness to learn
76/100Quality protection
62/100Manager readiness
57/100Decision confidence
53/100AI role clarity
48/100Experimentation safety
46/100CRITICAL MINORITY SIGNALS
The averages do not carry the whole risk.
have concealed AI-assisted work because expectations feel unclear
are unsure who is accountable when an AI-supported decision is wrong
say faster first drafts have not reduced approval time or rework
WHAT IS WORKING
Protect the capacity already present.
The organisation is not starting from resistance. It has professional pride, client commitment and a strong appetite to learn.
Professional pride
Employees care deeply about the standard and credibility of their work.
Client commitment
AI interest is strongest where it creates more time for valuable client thinking.
Learning appetite
Most employees want practical guidance and role-relevant practice, not generic training.
Peer support
Informal communities are already spreading useful practice across team boundaries.
“I do not want AI to do the thinking for me. I want it to give me more time to do the thinking only I can do.”Illustrative composite employee voice
THE TENSION ARCHITECTURE
Fifteen possible pressure points. Three need action now.
Each tension is a trade-off the organisation must manage, such as speed versus quality. Select one to see where the pressure sits and what the evidence suggests.
Live — strong enough to act on now
Watch — emerging; keep under review
Monitored — present, but not a current priority
The number is a friction signal out of 100: the nearer it is to 100, the stronger or more concentrated the pressure. It is not a percentage of employees.
Live tension
Organisational design
Centralisation vs Empowerment
72/100 friction signal
Centralisation
Empowerment
Where do decision rights actually sit - structurally, not on paper?
Strategy38
Tactics78
Operations71
ORIGINTactics
VERDICTProcess problem
WHAT MEASUREMENT SHOWS
Decision authority is understood at executive level, but becomes inconsistent when translated through management.
WHAT CONVERSATION REVEALS
Employees are not resisting accountability. They are learning decision rights through rejected attempts and repeated escalation.
“The policy says I own the outcome, but every unusual decision still travels two levels up.”Illustrative composite employee voice
“You learn where your authority ends by having the work sent back.”Illustrative composite employee voice
RECOMMENDATION DIRECTION
Design explicit decision rights, then introduce AI-suggested routing for low-risk approvals while managers retain authority.
What “perception gap” means here
Within a tension, it is the fraction of coded passages where an employee describes the organisation’s stated position as out of step with lived experience. It is not the percentage of employees who disagree.
FROM VERDICT TO ACTION
Five recommendations. Each one traceable.
Every proposal names its source tension, originating layer, AI–HI mode and success measures before anything is built.
01Human-led
AI informs; a person initiates and authorises.
02AI-suggested
AI recommends; a person approves before action.
03Agentic + checkpoints
AI runs the process and pauses at defined risk points.
04Fully agentic
AI acts within pre-agreed, low-risk boundaries.
01
Human Judgment vs Automation · Strategy & Tactics
Name AI decision rights
AI-suggested · human-authorisedDefine who initiates, recommends, authorises and remains accountable for each priority use case.
SUCCESS MEASURESBoundary clarity · challenge rate · unresolved accountability cases
02
Centralisation vs Empowerment · Tactics
Redesign low-risk approval flow
AI-suggested · human-authorisedUse explicit thresholds and AI-supported routing to remove back-and-forth without removing managerial authority.
SUCCESS MEASURESCycle time · escalations avoided · decision-confidence score
03
Speed vs Quality · Strategy–Tactics boundary
Reset the quality signal
Human-ledAlign leadership language, management behaviour, delivery thresholds and recognition with the stated quality commitment.
SUCCESS MEASURESRework · quality exceptions · perception-gap movement
04
Judgment + Psychological Safety
Create safe challenge practice
Human-led · AI diagnosticMake it legitimate and easy to question, override and disclose AI-supported work without diffusing accountability.
SUCCESS MEASURESOverrides disclosed · near misses surfaced · psychological safety
05
Capability vs Readiness · Tactics & Operations
Build practice, not just access
Agentic with checkpointsCreate role-based practice loops, exemplars and peer learning around real client work, with defined risk checkpoints.
SUCCESS MEASURESAdoption quality · repeat use · time redirected to higher-value work
WHERE AI CARRIES THE LOAD
Mundane, critical work made sustainable.
AI supports the administration beneath implementation. People retain ownership of every decision, relationship and act of trust-building.
- Humans own the decision
- AI carries repeatable administration
- Nothing is added unless a task stops
PHASE 2 · THE LIVING SYSTEM
Inspire. Activate. Elevate.
The hub becomes the single source of truth for implementation, evidence, ownership, progress and the next organisational question.
Inspire
From non-believer to informed participant.
- Leadership narrative
- Use-case theatre
- Safe experimentation
- Role-level relevance
Activate
Turn accepted recommendations into working practice.
- Decision-rights design
- Tension Labs
- Manager enablement
- Workflow implementation
Elevate
Measure efficacy and redirect capacity to higher-value work.
- Haven Pulse
- Lead & lag indicators
- Capability maturity
- Next Detection cycle
YEAR ONE SEQUENCE
Name the tensions and earn the right to be believed.
Build informed participation and role-level relevance.
Implement accepted decision, workflow and manager changes.
Read early behaviour and friction movement.
Compile lead indicators and correct the course.
Redirect capacity and mature AI–HI practice.
Refresh the evidence after action.
HAVEN PULSE
Testing the lead indicators.
A short, targeted version of the original conversation re-runs against the same tensions and success measures. It is not a fresh diagnostic; it tests whether the human signal is moving after implementation.
Detect→Build→Pulse↺
ONE SOURCE OF TRUTH
What the client receives.
The hub makes a substantial evidence pack usable on demand. Each layer can be updated once without circulating new document versions.
Executive evidence overview
The answer, the scale, what is working and the material risks.
Tension & attribution report
All themes, continuums, layer patterns and source-linked evidence.
Recommendation traceability
Every proposal mapped back to its finding, origin and success measure.
Implementation workspace
Owners, actions, progress, decisions, Pulse results and the next cycle.
CONVERSATIONAL INTELLIGENCE
Is the machine in the right part of the loop?
The demonstration is fictional. The method, tensions and decision discipline are real.