The manager is the medium

Every promise your organisation makes travels through one relationship. Almost nothing about that relationship is designed.

Marshall McLuhan gave the twentieth century one of its most durable aphorisms: the medium is the message. What carries the content shapes the content, so profoundly that the carrier ends up mattering more than the cargo.

I’ve come to believe there’s no better lens for understanding why employee experience programs succeed or fail, because organisations are transmission systems too. They generate enormous amounts of intended message: EVPs, values, strategies, policies, change narratives. And all of it, every promise in the portfolio, reaches an actual person through one carrier.

Their manager.

This week’s perspective is about what happens when you take that seriously. Not as a truism (“managers matter,” every organisation nods along), but as a structural fact with structural consequences, most of which point somewhere other than where the nodding usually leads.

The misconception: people experience the organisation directly The working assumption behind most people strategy is that the organisation communicates with its people, plural, directly. Craft the message well, cascade it properly, and the intended experience arrives intact. Under this assumption, the EVP is experienced as written, the values as launched, the change program as announced. The manager appears in this model as one stakeholder among many: important, needing to be “brought along,” but essentially a relay.

The lived reality is almost the inverse. From where any individual employee sits, the organisation is a distant abstraction that occasionally emails. What they actually have is a manager: the person who sets their work, reads their mood or doesn’t, approves or sighs at their requests, explains or fails to explain the decisions arriving from above, and does or doesn’t notice when they’ve gone quiet on a Tuesday afternoon.

Run any organisational promise through this lens and watch what happens to it.

The EVP promises flexibility. That promise has no existence in anyone’s life until the moment one manager responds to one request. The response is the promise, as far as that person will ever experience it. A generous policy delivered through a sighing manager is a stingy policy. A modest policy delivered through a genuinely supportive one is a generous policy.

The content was fixed; the medium rewrote it.

The values promise courage. Whether courage actually exists in a team is decided in the challenge moment we examined in the culture essay, and the manager is usually the person being challenged. Their reaction in those fifteen seconds is the values statement, locally ratified or locally repealed.

The change program promises transparency. But people don’t hear change from the CEO; they hear the CEO, then turn to their manager and ask “what does this mean for us?” If the manager learned about the change from the same email, the answer is a shrug, and the shrug is what transparency feels like in that team, whatever the comms plan said.

This is why the same organisation, with one EVP, one set of values and one culture deck, contains teams having wildly different employment experiences. There is no single organisational experience. There are as many organisations as there are managers, and every employee works for exactly one of them.

The evidence for this is sitting in public, in the exercise we ran three essays ago. Open your reviews and count the appearances of one five-word phrase: “depends entirely on your manager.” It’s the most common structural observation in employer reviews across every industry, and it’s not a complaint. It’s a correct description of the architecture.

The nuance: the medium has weather Now the complication, and it’s the piece of this argument I’ve been sharpening most recently, because it surfaced in a run of discussions about attrition that kept snagging on the same problem.

The problem is this: a good manager inside a poor subculture. Or the reverse. And an employee who cannot tell them apart.

Because the manager isn’t the only thing between the organisation and the person. Teams, sites, functions and portfolios each develop their own subculture: local norms about workload, safety to speak, how leaving is talked about, whether the official story is respected or quietly mocked. A capable, caring manager can be transmitting beautifully inside a subculture that contaminates everything anyway: the roster pressure they don’t control, the neighbouring team’s cynicism, a portfolio-level norm that says everyone’s temporary here.

And when the employee eventually leaves, and someone asks why, they can’t cleanly separate the channel from the weather. The exit survey ticks “management,” and a good manager wears a failure that was atmospheric.

This has a hard practical consequence for anyone who measures experience: organisational averages wash the signal out. If the manager is the medium and the subculture is the weather, then experience data only becomes readable at the level where mediums and weather actually vary, the team, the site, the portfolio, not the enterprise. An organisation-wide engagement score is an average of different climates, which is to say, a number about nowhere. The counting-versus-listening essay argued for following the score down to the story. This is the structural reason why: the story only exists locally, because the experience is only produced locally.

So the clean aphorism needs its footnote. The manager is the medium; the subculture is the weather; and any serious attempt to close the Promise Gap has to read both, at local resolution, without blaming the wire for the storm.

The wrong conclusion and the right one Here’s where most organisations, having nodded along to “managers matter,” take the exit ramp to the wrong destination.

The wrong conclusion is the hero-villain model: if managers are the decisive factor, then manage the managers harder. Train them more, measure them tighter, hold them “accountable” for engagement scores, and when experience is poor, locate the failure in the individual. It feels rigorous. It’s actually the organisation outsourcing its Promise Gap to its most exposed layer.

Because ask the question the hero-villain model never asks: what condition is the medium in?

What span of people is this manager carrying? A manager with eight reports can know each of them: notice the quiet Tuesday, remember the thing said at interview, keep the attention promise underneath all the other promises. A manager with twenty-five cannot, no matter their gifts. Somewhere between those numbers, noticing becomes structurally impossible, and everything downstream of noticing, the Maria moments this series began with, becomes impossible with it.

What load are they under? A manager whose own week is triaged survival will transmit triaged survival, whatever the culture deck says. Stress is the most contagious message in any workplace and it travels down the same wire as everything else.

What decision rights do they hold? Can they actually say yes to a reasonable request, flex a workload for a hard month, spend a small amount on their team without a business case? Or has every humane decision been escalated out of their hands, so that the medium can carry the promise but never keep it?

Span, load, permission. Three conditions, all set by the organisation, none chosen by the manager, and together they determine most of what the medium can transmit. A brilliant manager in impossible conditions delivers a poor experience. An average manager in well-designed conditions delivers a decent one, reliably, at scale. Which means the highest-leverage employee experience decisions in any organisation aren’t made in HR programs at all. They’re made wherever spans, structures and delegations get set, usually as cost decisions, by people who have never once thought of themselves as designing the employee experience.

They are designing it. They’re designing the only channel it travels through.

Blaming managers for the Promise Gap is blaming the wire for the blackout.

The counterargument: doesn’t this excuse bad managers?

The pushback writes itself, and it deserves a straight answer: “You’ve just built an alibi.

Some managers are genuinely poor: careless, political, unkind. Conditions don’t explain them, and your model hands them an excuse.“

Two things are true at once here.

Yes, individual manager quality is real and non-negotiable. Careless or unkind managers do disproportionate damage precisely because of everything this essay argues: they’re not one bad employee, they’re a corrupted channel through which every organisational promise reaches eight or fifteen people. Selecting managers for the actual job (noticing, explaining, keeping promises) rather than as a reward for individual output, and acting decisively on the genuinely poor ones, matters more under this model, not less.

But look at the order of operations. You cannot assess the wire until you’ve ruled out the storm. An organisation that holds managers accountable for experience while setting spans of twenty-five, briefing them last, and escalating their decision rights away has built a system where good and poor managers produce converging results, and then evaluates them as if the results were individual. Fix the conditions first, and manager quality becomes visible, developable, and fairly judgeable. Skip the conditions, and “manager accountability” is just the Promise Gap with a name attached to it.

The model doesn’t excuse bad managers. It stops good ones being spent as fuses.

What to do with this on Monday

Audit spans before you audit managers. Pull the distribution of direct reports across the organisation. Flag everything above the number at which noticing is plausible for your kind of work. That list is your employee experience investment plan, before a single program is bought.

Brief managers before you announce anything. One structural rule: no significant change reaches the all-staff channel before team managers have had it, with time to understand it and licence to answer questions. People will always turn to their manager to ask what it means. Decide in advance whether that moment transmits clarity or a shrug.

Hand managers the promise inventory. The careers page audit from four essays ago produced a list of every commitment your organisation makes in public. Give it to your managers, plainly: these are the promises you’re carrying. Most have never seen the promises they’re expected to keep.

Return one decision right this quarter. Find one humane, low-risk decision that currently escalates (a flexibility yes, a small team spend, a workload flex) and push it back to the manager. Permission is the cheapest capability program there is.

Read your experience data at medium resolution. Whatever you measure, cut it by team and site before you read the average. The variance between managers is the finding. The average is the place findings go to hide.

The sentence to keep

People don’t experience your organisation. They experience their manager.

Every promise in your portfolio, the EVP, the values, the change story, is cargo. The manager is the carrier, the subculture is the weather, and the organisation, mostly without knowing it, is the designer of both. McLuhan’s point was never that content doesn’t matter.

It’s that the carrier decides what arrives.

Next week, the essay this whole series has been walking toward: the moment the carrier delivers its final message. Every departure has an audience, and it isn’t the person leaving.

You measure the departure. Nobody designs it.

If you’ve worked for both organisations at once, the one on the intranet and the one your manager ran, tell me about the distance between them. Reply or comment. That distance is this publication’s home ground.

Shehzad Bhanji writes The Promise Gap, a weekly perspective on the relationship between organisational promises and lived experiences. Across a 25-year international career spanning marketing, customer experience, employer brand, HR technology and people experience, he has worked across Australia, Asia, Europe, the Middle East and Africa.