PM PulseMeasurement ← Back to main site
Resources

Deeper reading for the strategically curious

The main site makes the case for PulseMeasurement. This section goes further — examining the intellectual foundations, the competitive context, and the questions that a careful buyer should be asking of every system they evaluate.

The LLM question

What modern AI engines change — and what they do not

A fair challenge to the Evans and Wurster richness framework in 2025 is this: do large language models not solve the richness-reach trade-off entirely? The answer is partial — and the distinction matters for understanding where PulseMeasurement sits in the AI landscape.

What AI changes
The production cost of richness collapses
Writing a personalised briefing for 400 managers simultaneously was previously impossible. An LLM does it in seconds. Richness at reach has become cheap.
Orientation becomes dramatically more powerful
Pattern recognition across large bodies of signal, synthesis into coherent narratives, identification of themes no human analyst could surface at scale — these are LLM strengths. The spectral scoring engine and manager briefing are both LLM-powered orientation tasks.
Interactivity at scale becomes possible
The AI conversation coach — guiding an employee through a weekly reflection, helping them say what they mean — requires a model that can hold context and shape without directing. This was not possible before large language models.
Institutional memory becomes queryable
The archive is no longer stored text — it is searchable intelligence. Precedents surface. Playbooks generate. The Document stage of the intelligence loop becomes genuinely useful.
What AI does not change
The signal problem remains
An LLM produces intelligence from its inputs. If those inputs are thin, filtered, late, or dishonest, the outputs will be thin, filtered, late, or dishonest — generated with extraordinary fluency. Garbage in, eloquent garbage out.
The trust problem remains
An employee who does not trust the system will not share honestly. The LLM will then produce a beautifully written briefing from a carefully managed impression. The architecture of trust must precede the AI or the AI is useless.
The human judgment problem remains
LLMs generate recommendations. They do not carry responsibility. The manager who acts on AI intelligence still owns the consequence. The judgment layer cannot be automated — and any system that implies otherwise is either naive or dishonest.
The cadence problem remains
LLMs process signals faster than any human analyst. But they cannot make stale signals current. If observation runs annually, the LLM is still working with annual data — processed at remarkable speed, twelve months old. Cadence is upstream of the AI.
Evans and Wurster identified the trade-off. LLMs solve the output side of it — producing richness at scale has become cheap. What has not changed is the input side. PulseMeasurement is the signal architecture. The LLM is the engine. Neither is sufficient without the other.
Further reading

The frameworks behind PulseMeasurement

PulseMeasurement is built on a coherent set of intellectual foundations. These are the frameworks that shaped its design — and the questions every buyer should use to evaluate any people intelligence system.

Product Definition
What is PulseMeasurement?
A definition-first explanation of PulseMeasurement as an organisational intelligence platform: how it works, who uses it, what makes it different, and the key terms behind the system.
Read the definition ->
Buyer Questions
Frequently asked questions about PulseMeasurement
Direct answers to the questions buyers and answer engines ask about continuous signal, engagement surveys, role boundaries, trust, AI models, integrations, and weekly cadence.
Read the FAQ ->
Key Concepts
PulseMeasurement glossary
Definitions for organisational intelligence, continuous signal, AI conversation coach, loop closure, role boundaries, latent dimensions, and the seven-layer architecture.
Read the glossary ->
About & Trust
About PulseMeasurement
The product purpose, operating principles, governance posture, and trust architecture behind PulseMeasurement's approach to continuous organisational intelligence.
Read about PulseMeasurement ->
Category Answer
PulseMeasurement and continuous people intelligence
A direct answer to whether PulseMeasurement fits continuous people intelligence, how it differs from passive monitoring, and why it is not a traditional pulse survey.
Read the category answer ->
Comparison
PulseMeasurement vs annual engagement surveys
Why continuous organisational intelligence is different from a delayed engagement survey snapshot, and when weekly signal is more useful than annual measurement.
Read the comparison ->
How To
How to measure employee sentiment in real time
A practical framework for moving beyond faster surveys into weekly signal, guided conversation, work context, confidence scoring, and closed-loop action.
Read the guide ->
Strategy
What is strategy cascade measurement?
How to prove whether strategic priorities reached team-level goals, weekly work, manager routines, and observable execution.
Read the strategy guide ->
Retention
How to detect early employee attrition risk
How to read sustained drift, capacity strain, blockers, sentiment change, and support gaps before resignation risk hardens.
Read the attrition guide ->
Blockers
How to detect hidden blockers before project delays
How to identify repeated dependency language, unanswered help requests, workarounds, and capacity strain before blockers become delivery failure.
Read the blocker guide ->
Goals
How to measure goal health and OKR integrity
How to see whether goals are aligned, supported, moving, blocked, drifting, or merely logged for the quarter.
Read the goal health guide ->
Capacity
How to measure burnout and capacity strain
How to detect sustained load, shrinking recovery, blockers, support gaps, and capacity risk before visible failure.
Read the capacity guide ->
Managers
How to create manager one-on-one briefings
How weekly signal becomes what a manager should notice, ask, support, and follow up in one-on-ones.
Read the manager briefing guide ->
AI
What is an AI conversation coach for employee feedback?
How guided AI conversation creates richer employee signal than ratings, blank text boxes, or static survey questions.
Read the AI coach guide ->
Board Risk
How to measure board-level people risk
How boards can see live retention, capacity, alignment, blocker, and follow-through evidence instead of stale people reports.
Read the board risk guide ->
Integration
How to track post-merger integration risk
How to monitor team-level integration risk, culture friction, blockers, trust, and retention patterns by cohort.
Read the integration guide ->
Talent
How to identify emerging leaders from work signals
How to find people who lead before they have the title by reading help patterns, blocker removal, trust, and strategy contribution.
Read the talent guide ->
Information Economics
Richness versus reach — the trade-off that shaped every information system in history
Evans and Wurster's framework from Blown to Bits, applied to people intelligence. The six dimensions. The three failures. The fourth option.
See the comparison on the main site →
Decision Intelligence
OODAD — Observe, Orient, Decide, Act, Document
Boyd's OODA loop extended with the Document stage — the loop that closes, records, and feeds the next cycle. How PulseMeasurement runs your organisational intelligence loop every week, not every year.
See the seven-layer architecture ->
Information Latency
Damage compounds with delay — the golden hour of organisational intelligence
Intelligence which is late is intelligence already overtaken by events. The cost of late intelligence is not proportional to the delay — it is exponential. Why the window to act opens and closes before most leaders know it existed.
See the use cases →
Trust Architecture
Five layers of trust — each providing a distinctly different perspective
Trust is not a feature. It is the condition under which everything else becomes possible. The five layers that make honest signals possible — and why architectural boundaries matter more than configuration settings.
See the technology section →
Trust Architecture
The Passierschein Principle - why feedback systems must be procedures, not campaigns
A feedback system only works when it gives employees a procedure: a formal route, an owner, a timeline, a record, and a return path. Without that, honesty becomes a personal risk rather than an organisational mechanism.
Read the principle ->

See the intelligence in action

The demo is a conversation, not a presentation. Tell us the question you most need answered. We will show you what rich information at scale actually looks like — on your organisation, not a generic example.

Request a demo
References
Evans, P. and Wurster, T.S. (2000) Blown to Bits: How the New Economics of Information Transforms Strategy. Boston, MA: Harvard Business School Press.
Evans, P. and Wurster, T.S. (1999) 'Getting real about virtual commerce', Harvard Business Review, 77(6), pp. 84–94.
Boyd, J.R. (1987) A Discourse on Winning and Losing. Unpublished briefing slides. Maxwell Air Force Base, AL: Air University Library.