01 / Collect
Collect
Bring in scattered material from documents, decks, notes, transcripts, links, and data.
Q3-retro.md14 kb
standup-oct-14.vttlive
figma.com/…/auth-v2new
Introducing Orient
Meaning infrastructure for organisations.
Generating content is easy now. Preserving meaning is not. Orient turns scattered material into structured understanding, shapes it into clear communication, and shows where meaning held or broke across the organisation.OutcomeTeams stay aligned, make clearer decisions, reduce rework, and see where communication is breaking before it becomes costly.
Comprehension
0.31Only 31% of Engineering understood the Q3 strategy shift. 4 of 6 teams scored below threshold — the message did not land.
STRAT-Q3 · Engineering org · 247 recipientsView breakdown →
The new conditionAI made generation abundant.
AI made generation abundant.
Keeping teams aligned is still hard.
Documents, summaries, decks, reports, and decisions are now cheap to produce. But fluent output does not preserve provenance, resolve contradiction, or protect against drift. As meaning moves across teams and systems, organisations lose confidence in what is grounded, aligned, and safe to act on.
Most organisations still lack the infrastructure to keep teams aligned.
Where did this claim actually come from?Why does the deck contradict the strategy memo?What version of this did the executive team see?Do engineering and product mean the same thing by this?
Where did this claim actually come from?Why does the deck contradict the strategy memo?What version of this did the executive team see?Do engineering and product mean the same thing by this?
Where did this claim actually come from?Why does the deck contradict the strategy memo?What version of this did the executive team see?Do engineering and product mean the same thing by this?
Where did this claim actually come from?Why does the deck contradict the strategy memo?What version of this did the executive team see?Do engineering and product mean the same thing by this?
Has the board-approved plan changed since last quarter?What evidence supports this recommendation?Who last changed this, and why?Is the summary faithful to the source?
Has the board-approved plan changed since last quarter?What evidence supports this recommendation?Who last changed this, and why?Is the summary faithful to the source?
Has the board-approved plan changed since last quarter?What evidence supports this recommendation?Who last changed this, and why?Is the summary faithful to the source?
Has the board-approved plan changed since last quarter?What evidence supports this recommendation?Who last changed this, and why?Is the summary faithful to the source?
What did the field teams actually hear?Which interpretation is authoritative now?Was this decision documented anywhere?How did this number end up in the report?
What did the field teams actually hear?Which interpretation is authoritative now?Was this decision documented anywhere?How did this number end up in the report?
What did the field teams actually hear?Which interpretation is authoritative now?Was this decision documented anywhere?How did this number end up in the report?
What did the field teams actually hear?Which interpretation is authoritative now?Was this decision documented anywhere?How did this number end up in the report?
Provenance
Where did this come from?
How did it change?
Sourcelinked
Editlinked
Summarylinked
Claims
What is being asserted?
What supports it?
Claimasserted
Evidence
Support
Gap
Contradiction
What conflicts with what we already know?
What remains unresolved?
Known
≠
New
unresolved tension
Drift
What changed as the idea moved?
Across people, formats, and workflows.
Source
↓
Deck
↓
Handoff
↓
Team
In motion
Meaning does not fail in the source.
It fails in the transfer.
More output does not create more understanding.
More decks. More docs. More decisions. AI lets organisations generate content at speed, but they still struggle to preserve provenance, verify claims, and detect meaning drift as ideas move.
What looks like content growth is often synthetic authority.
The result isn’t just noise.
It’s institutional confusion at speed.
Provenancedisappears
Documentscirculate,buttheirsource,rationale,andsupportingevidencefadeastheideagetsrewritten.
Strategy Brief v7
Source
Claim
Rationale
Evidence
Decision
Traceable origin38%
Claimsgounverified
Assertionsmovequicklyintoplansanddecisionsevenwhenevidenceispartial,missing,orcontradictory.
> BRIEF Market risk memo
> CLAIMS 5 surfaced
> STATUS unverified
Revenue at risk64%
Root cause clear41%
Evidence attached22%
Contradictions resolved11%
Decision grounded4%
Verified claims: 22%
Meaningdriftsinmotion
Oneideabecomesseveralinterpretationsasitmovesacrossfunctions,formats,andworkflows.
Core decision
after redistribution
Leadership
Ops
Productshifted
Salesshifted
Supportshifted
Shared meaning intact: 12%
What it costsMisalignment is the most expensive line item
Misalignment is the most expensive line item
no one tracks.
People have always produced unclear, partial, or weakly grounded communication. AI makes that problem far more dangerous by generating fluent outputs that look credible, even when the underlying meaning is wrong, unsupported, or unverifiable. At organisational scale, that produces false confidence, alignment loss, repeated work, weaker decisions, and confusion with the appearance of truth.
20%
of knowledge work is still spent searching, reconstructing, and reinterpreting what already exists.
→ €5M+/year lost in productivity (1,000 employees)
McKinsey
$4.5M/year
lost when onboarding, handoffs, and undocumented knowledge leave teams guessing what is true.
20–200%
of salary lost when expertise departs and the reasoning behind decisions leaves with it
66% rely on AI output without checking accuracy
Credible language is often acted on before it is verified.
→ KPMG AI trust study
$1.2 Trillion Lost Annually in the U.S.
→ According to McKinsey and IDC estimates, U.S. companies lose over $1.2 trillion annually due to failures in knowledge work and internal communication.
Why it matters
Misalignment does not stay contained.
As communication moves across teams, systems, and decisions, small misunderstandings become repeated work, slower execution, and costly mistakes.
Looks alignedSounds authoritativeMoves fast
Surface signal
Confidence rises faster
than coherence.
Hidden cost
Outputs can look complete while provenance is lost, claims go unverified, and shared understanding quietly fragments underneath.
01
Synthetic authority
Generated material sounds complete and confident even when its provenance is unclear or its claims are unsupported.
02
Alignment loss
A single idea becomes multiple interpretations as it moves across teams, formats, and decisions.
03
Institutional confusion
Outputs accumulate faster than shared understanding, so confidence rises while coherence falls.
04
Meaning drift
Every handoff, format change, and audience shift alters what people think was meant.
The missing layer
The missing layer is meaning infrastructure.
AI has made generation cheap. Understanding is still expensive.
Organisations do not just need more content. They need a system that can preserve provenance, verify claims, hold contradiction without collapse, and detect when meaning shifts as communication moves.
That is the layer Orient provides.
Provenance
Preserved
Keep where meaning came from and how it changed.
Claims
Verified
Track what is asserted and what supports it.
Contradiction
Handled
Surface tension instead of flattening disagreement.
Drift
Detected
See when meaning shifts across people and formats.
What is Orient
A meaning system for organisations.
Orient helps teams turn scattered material into structured understanding, shape it into clear communication, and measure whether the meaning actually got through.
Stages of the loop
Six moves from scattered material to communication that holds.
0106
02 / Understand
Understand
Turn fragmented material into structured understanding: claims, evidence, concepts, contradictions, and context.
CLAIMAuth latency caused Q3 churn spike.
EVID.p95 > 1.4s over 9 days, 3 tickets.
CONFL.Ops attributes it to a third-party outage.
CTX.New auth provider shipped Sep 28.
03 / Frame
Frame
Choose the right communication shape for the audience, purpose, and level of detail.
AudienceExec
ToneDirect
DepthSummary
Context-ledDecision-first
04 / Render
Render
Render clear communication for the channel: decks, audio stories, videos, capsules, articles, and pages.
Deck
Audio
Video
Capsule
Article
Page
05 / Distribute
Distribute
Route communication to the right people on the right surfaces, with traceable intent.
Engineering · 12
Product Ops · 6
Leadership · queuedT-2h
06 / Measure
loop
Measure
Measure whether the meaning actually got through — and where it broke.
Understood88%
D−7drop · slide 4D0
Measure feeds back into Collect. Meaning is governed in a continuous loop.
The outcome
Shared understanding.
Organisational alignment.
Measurable comprehension.
Orient turns communication from a one-way act of publishing into a system of understanding, alignment, and feedback.
How it works
From scattered material to meaning that holds.
Orient gives organisations a repeatable loop: understand what is there, ground it in evidence and contradiction, shape it into communication, and measure whether the meaning made it through.
Understand
01
Build a structured model of meaning from fragmented source material.
- Claims and arguments
- Evidence and support
- Concepts and context
- Contradictions and tensions
Ground
02
Evaluate what is supported, uncertain, inconsistent, or missing.
- Epistemic analysis
- Claim validation
- Confidence and risk
- Narrative integrity
Render
03
Turn structured meaning into the right communication for the right audience.
- Decks and briefings
- Memos and articles
- Audio stories and video
- Format-specific outputs
Measure
04
Track whether communication was understood, and where meaning held or broke.
- Comprehension signals
- Clarity and coherence
- Audience fit
- Meaning decay
Improve
Turn measurement into better future communication and stronger shared understanding.
05
Memory
A shared memory for organisational meaning.
Orient preserves not just outputs, but the reasoning, evidence, revisions, and interpretation behind them.
Every interaction.
Every edit.
Every revision.
Every claim and its supporting evidence.
Every question asked against the material.
Every change in interpretation over time.
Why it is different
Most systems generate content. Orient governs meaning.
Before Orient
Start from generated drafts and disconnected documents
Treat content as the unit of work
Rewrite the same ideas across formats and audiences
Lose provenance as communication moves
Flatten contradiction to keep things moving
Stop at publish and hope meaning held
With Orient
Starts from structured understanding
Treats meaning as the underlying system
Preserves provenance and visible reasoning
Renders many outputs from shared meaning
Keeps contradiction available instead of hiding it
Measures whether communication actually got through
Use cases
Where meaning must survive movement.
Biotech · R&D
Expertise transfer
Capture expert reasoning so knowledge survives handoffs, hiring, and organisational change.
Government · Policy
Executive briefings
Turn complex material into clear, audience-specific communication without losing what matters.
Government · Regulation
Policy and regulation
Translate dense source material into grounded briefings while preserving provenance and contradiction.
Enterprise · Leadership
Strategy communication
Ensure strategic intent holds as it moves from leadership into teams, functions, and execution.
Finance · Research
Research synthesis
Structure large bodies of research into reusable understanding rather than disconnected summaries.
Technology · Operations
Cross-functional alignment
Give product, design, engineering, and leadership a shared model instead of parallel interpretations.
Foundations
Meaning has structure.
Orient is built on decades of work across the disciplines required to preserve, interpret, transmit, and evaluate meaning inside organisations.
Cognitive science
How people process, retain, and act on information.
Linguistics & semantics
How meaning is encoded, shifted, and preserved in language.
Information theory
How signals degrade across channels, contexts, and audiences.
Knowledge representation
How claims, evidence, concepts, and relationships can be structured.
Machine intelligence
How to extract, organise, and measure meaning at scale.
Orient changed how we handle communication. We no longer treat understanding as something we hope for after sending. We can structure what matters, communicate it clearly, and see whether the meaning actually held across teams.
See what communication looks like when meaning holds.
Bring provenance, clarity, and measurable comprehension into the system.