Targeted Insights

Broad instruments find patterns; they do not explain them. Why are meeting rooms booked but empty? Why is the quiet floor avoided? Why does reported crowding coexist with 40% occupancy? Why does hybrid attendance produce no team overlap? Each contradiction has multiple plausible causes with different strategic consequences — and acting on the wrong cause wastes capital. Targeted Insights is the discipline of resolving specific contradictions with proportionate, focused investigation before they harden into wrong conclusions.

• Whenever the Dynamics–Occupancy triangulation produces a divergence with strategic consequences
• When a behavioural signature (avoidance, hoarding, overflow) has multiple plausible causes
• When one team's pattern deviates sharply from organisational norms
• Before Define locks assumptions that rest on an unexplained contradiction

• Strategist framing the hypothesis set per investigation
• Affected teams and their community-group representatives
• FM and IT for environmental and systems evidence
• Kept deliberately small: these are focused probes, not new research programmes

• The divergence analysis and question set from the Occupancy Assessment
• Relevant Dynamics, culture and interview findings for the population in question
• Environmental data where conditions are a candidate cause (acoustics, temperature, air quality)
• Explicit hypotheses: every investigation starts with competing explanations, not an open question

1. Frame each contradiction as competing hypotheses. Example — avoided quiet floor: (a) acoustic quality fails, (b) location friction, (c) social norms penalise absence from the team zone, (d) settings misdesigned for actual focus tasks, (e) booking friction.
2. Design the minimum investigation that discriminates between hypotheses: short structural interviews, in-situ observation, environmental spot measurement, micro-surveys to the affected cohort, booking-pattern analysis.
3. Investigate in the space, not the meeting room — walk the floor with the people who avoid it.
4. Weigh evidence per hypothesis; accept mixed causes (most contradictions have two).
5. Convert the finding into a strategy consequence: what does this change in the demand model, the Kit of Parts, the policy assumptions or the change plan?
6. Document resolved contradictions in the evidence base; escalate unresolved ones as explicit risks, not silent gaps.

• The convenient explanation arriving first — the cause that requires no budget or no difficult conversation deserves extra scrutiny
• Environmental causes masquerading as cultural ones and vice versa
• Policy artefacts: behaviour produced by a rule (auto-release settings, booking rights) rather than by preference
• Contradictions that dissolve on inspection — sometimes the data definition, not the behaviour, was the problem
• Findings that generalise: one team's explained contradiction often diagnoses a building-wide condition

• Resolved-contradiction register: finding, evidence, confidence level
• Corrections to the demand model and assumption set
• Specific design and policy consequences per finding
• Residual-uncertainty register for unresolved cases

Corrections flow directly into Dynamics Alignment and Calibration — this method is quality control on the evidence base before Define consumes it. Design consequences feed the Kit of Parts and Architectural Guidance. The residual-uncertainty register carries into Scenario Development as explicit risk. In Recalibrate, the same discipline reappears as the standing investigation capability.

Hypothesis framing is the judgement: knowing which five explanations are plausible for an avoided floor comes from experience across organisations, not from data. So does knowing when to stop — when the finding is confident enough to act on and further investigation is procrastination.

• Skipping the method and letting Define inherit unexplained contradictions as silent assumptions
• Launching a full new survey to answer a question twenty minutes on the floor would resolve
• Accepting the first plausible cause without testing competitors
• Investigating everything — targeted means chosen by strategic consequence
• Reporting causes with false confidence; mixed and partial explanations are legitimate findings

An insurer's data showed employees reporting severe crowding while measured occupancy averaged 52%. Competing hypotheses: peak concentration, zone-level density, acoustic load misread as crowding, or survey artefact. Spot observation and micro-survey resolved it: two of six zones ran at 95% while four sat near-empty — team anchoring norms made the empty zones socially unavailable. The consequence was a zoning and norms intervention, not the floor expansion the crowding reports had implied.