Dark Matter

On reading what we cannot see.

In 1933, an astronomer called Fritz Zwicky was looking at the Coma Cluster - a swarm of more than a thousand galaxies, bound by gravity, about 320 million light years away. He measured how fast the galaxies were moving relative to each other, and the numbers didn’t make sense.

The cluster shouldn’t have held together. Given the visible mass - the stars, the dust, everything that emitted or absorbed light - there wasn’t nearly enough gravity to keep the galaxies from flying apart. By Zwicky’s calculation, the cluster needed roughly four hundred times more mass than anyone could see, just to stay bound.

He called the missing thing dunkle Materie - dark matter. And then the field largely ignored him for forty years.

In the 1970s, Vera Rubin started measuring the rotation curves of spiral galaxies, and the same problem came back. This time it was impossible to dismiss.

Stars at the outer edges of a galaxy should orbit more slowly than stars near the centre, the way the outer planets in our solar system move more slowly than the inner ones. That is what gravity does - its pull weakens with distance. But Rubin found the outer stars moving just as fast as the inner ones, in galaxy after galaxy. Either gravity itself worked differently at galactic scales, or there was a vast invisible halo of mass wrapped around every spiral galaxy, providing the extra pull. Her data was clean enough that the second explanation became the only one that made sense.

Zwicky had been right. There was something else there.

Once you accept that something invisible is bending the universe into the shape it’s actually in, the natural next move is to try to find out what it is. So that is what we’ve been doing.

For the last fifty years, cosmologists and particle physicists have been building instruments designed to catch dark matter directly. Vast tanks of liquid xenon, suspended in underground vaults in Italy and South Dakota, watching for the flash of light that would mark a single particle drifting through. Detectors buried a mile down in disused mineshafts, shielded from the cosmic noise that would otherwise drown out a real signal. Particle accelerators built to slam protons together hard enough to produce a dark matter particle in the wreckage. Decades of work, billions and billions of dollars, some of the most sophisticated and sensitive instruments ever built.

And nothing. Not a single confirmed detection.

We know it’s there anyway, with more confidence than almost anything else in cosmology. Without it, galaxies fly apart. Without it, light doesn’t bend the way the universe shows it bending. Without it, the cosmic microwave background carries the wrong fingerprint and the structure of the universe doesn’t form in the way we observe it forming. The maths only works if dark matter is in it.

Roughly 27% of the universe is made of something we have never directly observed, and we know it’s there because of everything around it that would not behave the way it does if it weren’t. The evidence is in what holds, what moves, what doesn’t fall apart.

Why am I talking about dark matter? Because that sentence describes something I have struggled to articulate about the most consequential work I do. The work has mass. It shapes what holds inside the organisations and systems it touches. And like the thing the cosmologists are looking for, it registers nothing on any instrument the industry has built to measure it.

Now, I am fully aware that comparing my strategy work to one of the most profound unsolved problems in physics is, on the face of it, ridiculous. But bear with me. Because someone asked me, recently, how I measure whether my work has any value. And I have been trying to answer them ever since.

---

The question came as a comment underneath something I’d written on LinkedIn. Phrased politely - Beyond the experience itself, how do you measure whether or not your thoughts and opinions have value? - underneath was something else. Prove it. Justify your perspective. Show why anyone should be paying you to think.

The answer I gave at the time was true but partial. I said most of my work is governed by NDAs designed to keep it from circulating. I said my work is not input-output. There is no clean line between a single document and a quarterly result, no campaign that ran, no sales spike attributable to my involvement. I said the work is catalytic. The most useful thing I do often happens months after I have left the room, in a meeting I am not in, in a decision I will not be told about, in language that surfaces years later in someone else’s strategy without me anywhere in the document trail.

All of that is accurate. None of it is enough.

My answer described the work. It didn’t answer what was actually being asked. Which was whether this work is genuinely beyond measurement, or whether we have just never bothered to figure out how to measure it. Catalytic is either a true description of what the work does, or it is a word the field has been hiding behind for a very long time. I don't think I can keep using it without knowing which.

---

Let’s start with the work I actually do.

A client recently brought me in as a futurist-in-residence to work across five territories of change shaping the next decade of their business. On paper, a series of horizon-scans. Behind that, the harder question - how an institution built for one set of conditions should think about itself in a world that no longer matches them.

A different client brought me in to design a leadership development module for their top 50 leaders. The brief was a blank page. The direction I came back with was how a generation of leaders, promoted in a more stable era, were supposed to lead now — when the social contracts they inherited are breaking, the playbooks they were taught are no longer working, and the certainties they built their careers on are gone.

And then there are the myriad keynotes. The ones that matter aren’t really performance. They are something closer to thinking out loud, on stage, with a room of executives sitting with a question they need to work through together. Not the official theme of the conference, but the thing they are actually trying to work out.

What kind of macro environment the next decade actually is, when the comfortable assumptions about globalisation, governance and growth have all been quietly retired. What discernment means when truth itself is in flux, edited at source by systems no one elected. What it costs a leadership team to keep mistaking intelligence for wisdom, and what gets lost in the trade. Where sovereignty actually sits when the most consequential systems of the next decade are being built by companies who feel they owe nothing to the publics they reshape. What this generation of leaders is actually leaving behind, and whether the inheritance is one anyone would knowingly accept.

The talk is what gives the room something to think against. The thinking is the work.

Three engagements, three different shapes. None is a campaign. None produces a metric at the end. All of them are about what an organisation, a system, a leadership team will be capable of on the other side.

---

At its simplest, the work creates better conversations, which produce better decisions. A leadership team gets new angles, new nuances, new watch-outs, new layers of information that weren’t in the room before. They start arguing differently. They start asking each other things they hadn’t thought to ask. A direction that would have been taken doesn’t get taken. A direction that wouldn’t have been considered, does. The organisation walks out of the engagement having made one or two structural choices that will compound for a decade.

Underneath that, something more durable. The work leaves residue. Language the team starts using that they didn’t have before. A framing they keep returning to. A question someone keeps re-posing in subsequent meetings, sometimes for years. Sometimes the residue hardens into something formal - a piece of governance, a policy, a principle that ends up written into how decisions get made. Sometimes it just stays in the air, but it stays. The work I do leaves a vocabulary, a shape, and a set of reference points that the organisation continues to use long after I’ve gone.

And around that, depending on the engagement, something wider. A youth sport that produces fewer young people who leave it thinking less of themselves, because the system that produced them has been rebuilt. A constituency that gets represented in a place that had written it off. A board that is asking better questions about AI than the boards around it, which means the next ten years of an institution moves in one direction rather than another, which touches everyone whose lives intersect with that institution. The impact is not always commercial. Sometimes it is structural. Sometimes it is generational. Sometimes it is both.

This is what I am being asked to prove. Not whether a campaign moved a metric. Not whether costs came out of a P&L. Whether better conversations happened, whether the language and frames they produced are still in use, and whether the system on the other side of the engagement is operating differently because the work was done.

---

The existing measurement tools were built for different work, by the people whose work they measure.

The marketing world has spent decades building one set. Brand tracking, attribution modelling, marketing mix modelling, mental availability, Ehrenberg-Bass, the proprietary frameworks the holding companies sell back to clients as evidence that what was paid for produced something. The whole apparatus answers one question: did this campaign work? It was built by people running campaigns, for people paying for them. Anything that happens upstream of a campaign - anything that decides whether the campaign is even the right thing to be doing - registers nowhere on those instruments. By design, not by accident. The work that decides what a campaign should be, or whether there should be one at all, happens somewhere else entirely.

None of which has much to do with what I do. I’m not in the marketing department. The work I do happens several rooms over - org design, technology strategy, talent decisions, what futures the business is building toward, what bets it should be making. The brand is a downstream consequence of those decisions, not the thing being decided. A different conversation, with different stakes.

The consulting world has built a different set. McKinsey, BCG, Bain. Their measurement is for the upstream territory - cost taken out, capabilities built, transformations delivered, case studies stacked into pitch books for the next pitch. Whitepapers that establish the firm as the authority before the procurement conversation even starts. Independent benchmarks that conveniently confirm the firm’s own diagnosis. Whether the work is actually any good is almost beside the point. The story around it is defensible in a board pack, repeatable across engagements, and propped up by a global firm with the scale to make its narrative the narrative. That’s what gets the next engagement.

I sit closest to this world, without quite belonging to it. The consultancies sell certainty - playbooks, frameworks, “this is what worked at the other client.” I sell something closer to the opposite. Every engagement starts from scratch, with a specific question that no other engagement is going to ask in quite the same way. The work is bespoke, which is exactly what the consulting measurement story can’t accommodate. Their story is built around scale and repeatability. Mine is built around neither.

Which leaves me - and everyone else doing this kind of upstream thinking outside a firm - with no story at all. The consultancies built theirs, however thin. We haven’t built anything. And that nobody has built anything, including me, is the part of this I have found hardest to admit.

---

So what would it take to build the apparatus the work deserves?

It’s catalytic. It’s not really attributable. That’s a long-term play. You can’t really measure that kind of thing. The value is in the conversation. You’ll know it when you see it.

Every one of those phrases has done an enormous amount of unearned labour, for an enormous number of practitioners, over a very long time. Several have done some of that labour for me.

The honest answer is that the work isn’t unmeasurable. But the work isn’t one thing.

Over the years I have come to think about it in terms of registers - different altitudes the work operates at, each one with its own logic. The same practitioner can move between several of them in the same week, sometimes in the same engagement. The mistake - the one I have been making, and the one the field has been making - is to treat them as if they were one register, judged by one set of measures. They aren’t. Each one comes with a different unit of value, a different time horizon, and a different kind of evidence.

There are four, as I can currently distinguish them.

The institutional register.

Example - A board commissioning a strategic review of how their organisation should restructure itself for the next decade. The output isn’t a campaign or a transformation programme. It is a new shape, written into governance.

This is work that reshapes an organisation. The board curriculum that shifts how the most senior leaders think about a structural transition. The strategy document that reorganises a function. The leadership development module that changes the assumptions a top cohort is operating from. The framework the C-suite now uses to think about its relationship to a technology, a category, a future state. The work leaves real things behind inside the organisation - artefacts that exist, governance that didn’t, language that holds, decisions that get made differently because of architecture that wasn’t there before. The evidence is auditable, if anyone bothered to audit it. Which functions exist now that didn’t eighteen months ago. Which decisions reference frameworks that came out of the engagement. Which roles were created. Which committees were dissolved.

The ecosystem register.

Example - A regulator running a pilot programme that, if it works, will become the template the wider sector is held to.

This is work that reshapes a system, an industry, a sector. A new code of practice that an industry quietly converges on within five years. A research initiative that changes how a whole field thinks about a category of problem. A pedagogical model in one school district that gets adopted across thirty others. A framework one NGO develops that becomes the way every NGO in adjacent territory ends up working. The evidence is whether the system has moved. Whether the field's defaults have shifted. Whether other organisations operating in similar territory are now doing things differently because of work that was done somewhere else.

The speculative register.

Example - An energy company commissioning a deep scenario exercise on what the next forty years look like under three radically different climate-policy regimes.

Work that changes what an organisation can ask. Futurist-in-residence engagements. Horizon-scanning. Cross-domain reading and synthesis that gives a leadership team coordinates it didn’t have. The evidence here is not an artefact and not a system shift. It is the range and quality of questions the organisation is now capable of asking. Whether the organisation is planning further out than it used to. Whether the strategy now anticipates conditions it previously assumed wouldn’t arrive. Whether the questions surfaced in the engagement are still being pursued two years later, in rooms the strategist has long since left.

The cultural register.

Example - A piece of public writing that names what an industry has been refusing to look at, and a year later half the field is using its language without remembering where they got it from.

This is work that changes what a field is willing to say about itself. Public writing that surfaces what the industry has been refusing to look at. Investigative pieces that name specific failures and force a reckoning. Frameworks that become the way the next cohort of practitioners describe their own work. Essays that travel past their original platform and end up cited in conversations the writer wasn’t in. The evidence here is language travelling. Conversations shifting. Practitioners arriving with different defaults than they would have, because the field’s conversation has moved.

---

Four registers is not the answer, I know that. But it’s the start of one. The work happens at four different altitudes I can currently distinguish, with four different units of value, time horizons, and kinds of evidence. Anyone looking properly will find more. But the point is not the taxonomy. The point is that we have been treating a multi-altitude practice as if it were one thing - measurable by one apparatus, judged by one kind of metric, defensible inside one kind of board pack. That is the category error underneath the measurement question. And as a result, we have defaulted to “it can’t be measured.”

But when cosmologists realised they couldn’t detect dark matter directly, they didn’t conclude the work was over. They built a different science. Galaxy rotation curves, gravitational lensing, the structure of the cosmic microwave background, the dynamics of galaxy clusters - none of these observations detect dark matter itself. They detect the effects of something. And the proof, when it arrived, was not a single instrument finally catching the particle. It was the convergence of multiple independent observations all pointing at the same missing mass. The thing was confirmed not by being seen but by being inferred from the shape of what surrounded it.

The same architecture, I think, applies to the work I am trying to describe. No single register is the evidence. The evidence is in the convergence. An institutional artefact that exists, alongside an ecosystem that has moved, alongside questions the organisation is now able to ask, alongside language entering the field’s bloodstream - multiple independent registers pointing at the same source - is structurally more credible than any of them in isolation. The proof, if proof is the right word, is in the pattern that emerges when several different effects line up.

What I genuinely don’t know is how we measure this. What the equivalent of gravitational lensing would actually be in this domain. What the equivalent of the Bullet Cluster - the observation that finally settled dark matter for most physicists - would even look like, for upstream strategy work.

But even though I don’t know that, I am sure of the prior point. The work is not unmeasurable. We have been using a tape measure on a cathedral, complaining the cathedral doesn’t fit, and concluding that nothing better can be built. That isn’t true. It’s just convenient.

---

Which brings me back to the comment that started this. The question - how do I measure whether my work has any value - was a fair one. My answer was partly fair too. The NDAs are real. I can’t show specific work, and that constraint isn’t going anywhere. But what I can’t defend is the rest of what I said - the catalytic stuff, the years-after-I-leave-the-room, the it-just-happens-in-language. That part wasn’t measurement. That part was skirting around the question.

What I should have said is that I don’t yet have a good way to measure my impact, but that is a job worth doing rather than a fact about the work. That distinction is the thing I missed.

And sitting in this space has spurred me into a different practice. The questions I will be asking myself, and asking of the work, are different now. What holds, what moves, what doesn’t fall apart, on the other side of the work? What is the organisation, the sector, the field able to do that it couldn’t do before? What is the convergence of evidence pointing at, even if no single piece of evidence is decisive on its own? What would a measurement practice look like that took those questions seriously, instead of reaching for the nearest available number or denying measurement is possible at all?

And yes, this will take a long while - the tools and approaches for it don’t yet exist. But it’s a practice I am committed to from this point on. Because how can I survive as a consultant at this level if I don’t? In a world of data overload, where numbers have become the religion and rigour gets confused with measurability, anyone working at this altitude either finds a way to make their work legible or stops being a serious option.

Inference is a new world for this field. But it is going to be a critical one to crack.