I was trained to ask one question: how do you know that?
I came to this from engineering and auditing rather than organisational development. That means I spent years being paid to ask one specific question: how do you know that?
In an ISO audit, nobody cares what you believe is happening. You show the record, or the finding does not hold. Most work on communication lacks that habit. It is full of confident claims about culture and personality, and almost none of them can be checked, repeated, or compared to last quarter.
On plant floors and project sites, I rarely saw a shortage of good communicators. I saw capable people whose information had nowhere useful to go. I saw leaders told to fix a trust problem that was really a format problem, and training problems being handled with more meetings. Every time, the missing step was the diagnosis.
So the method is built the way an audit is built. It applies a fixed standard the same way for everyone, traces findings to dated examples, and states its limits openly. If I cannot back something with evidence, I will say so instead of giving you my opinion.
Nirishka Swart — Founder
B.Eng. · PMP · ISO 9001 Lead Auditor

Distance is one of the reasons a breakdown survives.
The work happens on plant floors, on job sites, in control rooms and in the offices above them, and with leaders responsible for locations they only visit occasionally. The further a leader is from where the work happens, the longer a problem can go unnamed.
Language is where communication reliably breaks.
The work is delivered in English and French. In a Quebec plant with French-first floor leadership and English-first corporate functions, language is one of the places communication reliably breaks, so we assess it as a dimension of its own.
Five commitments, written to be held to.
A claim invites comparison. A commitment can be checked against what actually gets delivered, and the method holds your organisation to the same test.
Cause before symptom
Every finding traces to a mechanism, because each mechanism needs a different kind of fix. Getting the mechanism wrong is the expensive failure in this field. It is how an organisation spends a year building trust when somebody needed training.
Evidence, not impression
Nothing is asserted. Findings are backed by dated, traced examples, and the serious ones need at least two independent sources before they are written down at all.
One standard, applied the same way
Every client gets the same dimensions, mechanisms and standard. That consistency makes a finding defensible when someone challenges it, and makes this quarter comparable to the last.
Systems that outlast me
The goal is a system you run yourselves. If the improvement depends on me being in the room, it is a dependency, and you already have one of those.
The limits get said out loud
Where the evidence shows association rather than proof, we say association. Where a pathway is a working hypothesis, we label it as one. Overclaiming is the fastest way to lose the one thing this method has going for it.
Answered directly, including the awkward one.
There is no published client case study yet. That question is second on the list, because a hedged answer would undermine everything above.
- What methodology are you using?
It is a hybrid, and each part is adapted from an established source. Tracing real incidents through the organisation comes from ECCO analysis. The psychological-safety mechanism draws on Amy Edmondson's research. Sorting findings into individual, interpersonal and organisational layers is a simplified version of the Downs–Hazen framework. Where individual processing preferences matter, we use a validated behavioural instrument instead of an improvised questionnaire.
The combination is ours: four mechanisms, ten dimensions across Me, We and Us, and a fixed evidence standard that decides when a finding is allowed to exist. The structure is the same for every client, which is what makes findings comparable.
The limit is that no published instrument validates this specific combination. The research on the individual mechanisms is strong in places and thinner in others, and we would rather tell you which is which than present the synthesis as settled science.
- Why should I trust this without a published case study?
You shouldn't, on my word alone. There is no client case study I have permission to publish yet, and you will not find an invented one here: no statistics we cannot source, and no testimonials from companies that do not exist. That is a deliberate choice.
Here is what you can check instead. The method is fixed and written down, so it asks you the same questions it asks anyone else. The evidence standard is explicit, so you can see when a finding falls short of it. And the entry point is a two-and-a-half-hour session on your own recent breakdowns, which is the cheapest way to find out whether this finds something you had not already named.
If it does not, you will know inside an afternoon.
- Is this a personality assessment?
No. Behavioural profiling is used where it explains a specific friction, such as two people whose information protocols are genuinely incompatible. The individual profile belongs to that person. It goes to them privately, with their consent, and appears in the shared report only as a team-level pattern.
The diagnosis is about the system, not the people in it. Findings are written as operational design problems, because those are the ones somebody can fix.
- What happens to what people tell you?
Quotes are anonymised and patterns are described at team level. No finding is attributed to an individual in anything the organisation sees.
Where ongoing measurement is part of the engagement, it reads counts and timestamps from business records you already keep, at team level only. It never reads message content or individual-level data. The system is built to accept nothing else, so this does not rest on a policy about what we choose to look at.
- How much of our time does this take?
The Audit is one session: two and a half hours, up to five people, and nothing to prepare beyond a willingness to talk about real recent incidents.
A full Diagnostic means a structured interview with each person in scope (or a representative sample on a larger team), plus a review of records you already produce, such as meeting cadence, decision logs and channel usage. It is designed to run alongside the operation without stopping it.
- Do you work in French?
Yes. Delivery is bilingual, and on a bilingual site that matters more than it sounds. Where the floor works in one language and the corporate functions in another, the language line is often where the breakdown is, so it is assessed as a dimension in its own right.
- How small or large a team does this work on?
The Audit is built for up to five people. Diagnostics usually run on teams of four to twelve. Beyond roughly fifteen to twenty people, the flow map covers a representative sample instead of every individual, and we agree that scope with you openly.
Still the fastest way to check: an afternoon.
Delivered in English and French, from Quebec, Canada. If a session on your own recent breakdowns finds nothing you had not already named, you will know by the end of it.
