I sat in a post-mortem a few years ago that I have never quite forgotten, and what made it memorable was that nobody in the room said a single untrue word.
A major client implementation had gone badly. Deadlines had been missed, the client was threatening to escalate, and the board wanted answers. The leadership team gathered to establish what had happened, and within ten minutes the meeting had taken on a shape I have seen more often than I care to admit. The commercial director explained that the scope had been clear and the timeline agreed, so this was fundamentally a delivery matter. The head of operations replied that the scope had changed three times after the contract was signed, and that the resource implications had been flagged on each occasion. The finance director, invited to explain why the budget overrun had not been caught sooner, observed with considerable dignity that the variance had been reported in the monthly pack. It was on page twelve.
Every word of that was accurate. Each function had, by its own lights, done its job. The client was still furious, the deadlines were still missed, and the one sentence the situation actually required never arrived. Nobody said: this is our failure, collectively, and here is what we are going to do about it.
I have thought about that meeting a great deal since, because it captures how organisations actually fail. Rarely through incompetence, and almost never through villainy. They fail through the problems that sit in the gaps between functions, where no single leader’s name is attached and accountability quietly evaporates. The most dangerous problems in any organisation are the ones with nobody’s name on them.
Everyone did their job, and the organisation failed anyway
Silo behaviour under pressure is one of the most common things I see in leadership teams, and one of the least examined. Nobody walks into a Monday meeting having decided to put their department ahead of the company. What happens is quieter than that. Pressure arrives and activates an instinct that is entirely reasonable at the individual level: protect what you are directly responsible for. Operations secures operations. The commercial director shields the pipeline, and finance tightens its grip on the reporting. Each of them is doing something defensible. Collectively they are doing something ruinous, because while every leader is busy securing their own area, nobody is securing the enterprise.
I recognise the pattern because I have lived it from the inside. There was a stretch, whilst I was running my own business, when our small leadership team pulled back into functional territories without any of us ever announcing it. I retreated into client relationships, which I told myself was where I added the most value. My operations lead disappeared into process, and my finance person burrowed into the numbers. On paper we were a team. In practice we were three people defending three budgets, and the places that needed all of us, the seams where selling met delivering met getting paid for it, had nobody watching them at all. Those seams were precisely where the business began to fray. Not because anyone was doing a poor job inside their own area, but because nobody was doing the job of holding the areas together.
It would be comforting to call this a character flaw, since character flaws conveniently belong to other people. It is actually a structural problem. Functional metrics reward leaders for optimising their own patch, so under pressure they optimise their own patch. Careers rise and fall on how a function is perceived, so distancing your area from a failure feels like survival rather than politics. Budgets, headcount and executive attention are all contested, so collaboration starts to look like a luxury the moment money tightens. Underneath the lot sits a quieter driver: the fear that owning a share of a failure will read as weakness. Put four understandable incentives in one room and you get a leadership team that performs beautifully as a set of individuals and fails completely as a collective.
The potholes always appear at the boundary line
There is a road I drive regularly that gets resurfaced in instalments, because it runs across the border between two councils. Each authority maintains its stretch conscientiously, right up to the line on the map where its responsibility ends. The seam where the two surfaces meet is where the tarmac crumbles, every winter, without fail. Both councils are doing their jobs. The road still breaks, and it breaks in the one place that belongs to neither of them.
Organisations wear out in exactly the same spot. The problems that do the real damage almost never sit tidily inside a function, where a competent leader with a clear remit would simply deal with them. They live in the handoffs. Between what was sold and what can be delivered. Between the forecast and the capacity plan. Between a strategy agreed upstairs and the operational reality three floors down. When accountability fragments, these problems get passed around the leadership table like an unwanted parcel. Operations says it is a commercial issue, commercial returns the compliment, and finance points out that it flagged the whole thing months ago on page twelve. The parcel keeps moving, and the problem inside it keeps growing.
Here is the part that should worry a leadership team most: the organisation can see all of it. Employees are remarkably good at reading the dynamics above them. They know which of their leaders cooperate and which merely coexist, which relationships are real and which are performed for the town hall. They could tell you, with unnerving precision, whether the leadership team is actually a team or simply a collection of senior people who attend the same meetings. Their behaviour adapts to what they see. A leadership team that operates in silos will run an organisation that operates in silos, and no poster about collaboration in the lift lobby will argue anybody out of it.
Your first team is not the one that reports to you
The idea I use with leadership teams to counter all of this is one I call Team Number One. Every leader sits on two teams. They lead their function; that is Team Two. They serve the enterprise as a member of the leadership team; that is Team One. Enterprise-first accountability means that when the interests of the two collide, Team One wins, every time, with no committee required to adjudicate.
Written down, it looks obvious. Practised, it is one of the hardest disciplines in leadership, because Team Two is where a leader actually lives. It is where their relationships run deepest, where their loyalty is felt most keenly, and where their performance is most directly measured. Choosing the enterprise can mean carrying disappointing news back to your own people. It can mean accepting a decision that makes your function’s year harder because it makes the organisation’s decade better, and doing it without martyrdom, which is the part most of us find hardest of all.
I watched it done properly once, and the striking thing was how unremarkable it looked. A sales director I worked with had a strong case for additional headcount, backed by real demand. At the same meeting where he intended to make that case, the operations director walked the team through a capacity problem that would start hurting client delivery within three months. The sales director listened, then said his request could wait, since the capacity problem needed dealing with first. There was no speech, and no pointed glance at the managing director to check the sacrifice had been noticed. He had weighed what the organisation needed against what his team wanted, and he said so in one plain sentence.
Contrast that with a managing director I knew who, after a particularly bruising cross-functional argument, stood up and told his colleagues: “We’re behaving like five departments who happen to share a building. We need to start behaving like one company.” It is the kind of statement that should never need saying out loud. The fact that it did, and that it changed the temperature of the room, tells you how far fragmented accountability can drift before anyone names it.
Accountability faces forwards, or it isn’t accountability
One distinction makes all of this workable, and muddling it is what makes collective ownership feel so dangerous: accountability and blame are different instruments. Blame looks backwards. It wants to know whose fault the failure was, usually so the fault can be seen to land somewhere other than on the person asking. Accountability looks the other way up the road. It wants to know what has to happen next, and who is going to make it happen.
The best leadership teams I have worked with hold this as an explicit norm, sometimes stated and sometimes simply understood through years of practice. When something goes wrong, the first question in the room is what do we need to do. Only once that is moving does the second question get asked, namely what happened and why, and even then it is treated as a systems enquiry rather than a hunt. Where were the handoffs weak, which signals got missed, what in our process let this through. Questions of that shape produce answers a team can build with. Whose fault was it produces a defensive essay from every corner of the table, each one factually accurate and collectively useless, which is roughly where this essay came in.
This discipline is also the one that makes the others worth having. A leadership team can be honest, behaviourally consistent, anchored in its convictions and quick to decide, and still fail, if each leader applies those strengths only inside their own silo. Honesty within functions but never across them. Fast decisions in one area that quietly manufacture the next area’s crisis. Enterprise-first accountability points everything else at the right target. It reminds a leadership team that it exists to run one organisation, whatever the org chart says about who runs what, and that every decision, every challenge and every difficult conversation starts from that premise.
Enterprise-first accountability is one of the five disciplines of the RESOLVE framework, which sits at the centre of my forthcoming book, The Leader with Resolve, out on 1 October 2026. Each discipline guards against a specific way that leadership gives way under pressure. This one guards against fragmentation, the slow separation of a leadership team into territories that pressure will eventually prise apart.
If parts of this essay felt uncomfortably familiar, I would be glad to compare notes; a diagnostic conversation with me is exactly that, a conversation rather than a pitch. A lighter first step is the free RESOLVE diagnostic, which takes about two minutes, asks nothing in return, and will show you where your own team holds together and where the seams need attention.