Flow, Not Status: What Kanban Reveals About Programme Governance
Stop asking people how they feel about their work, and start watching how the work moves.
The Ritual We All Recognise
Every few weeks, in programmes across every sector, the same ritual plays out. A steering group assembles. A pack is circulated — thirty slides, sometimes eighty. Each workstream lead narrates a row of coloured indicators: green, amber, red. The board asks a question or two, notes the reds, and closes with a shared sense that the programme has been governed.
It has not. It has been reported on. In my experience these are not the same activity, and confusing them is one of the quieter reasons large programmes drift.
A status report is a snapshot of intentions dressed up as a snapshot of reality. It records what each lead believes to be true about their part of the work on the morning they filled in the template. It is, by construction, a lagging and subjective account — assembled by the very people whose performance it describes, at the moment they are least able to be objective about it.
The trouble is not dishonesty. The trouble is that status answers the wrong question. It asks how do you feel about your workstream? when governance needs to know what is actually happening to the work?
A status report tells you what people believe about their work. A board that shows flow tells you what the work is actually doing — whether you ask it or not.
Status Is a Noun; Flow Is a Verb
The distinction I keep returning to is grammatical. Status is a noun — a state, captured at a point in time. Flow is a verb — the movement of work through the system that produces it.
When you govern by status, you are looking at a series of still photographs and trying to infer motion. When you govern by flow, you are watching the film.
The practical difference shows up the moment you put the work on a wall. Not a reporting wall — a working wall, where every meaningful piece of work is a card, and every card moves left to right through the stages it must pass: analysis, build, test, ready, done. Columns have limits on how many cards they may hold. The wall is updated by the people doing the work, as they do it, not by a coordinator the night before the board meets.
What that wall reveals is precisely what the status pack conceals:
- Where work actually sits. Not amber, but in test for eleven days, and the column is full.
- Where the queues are. Cards pile up in front of the constraint, and the constraint is visible to anyone who walks past.
- How much is in flight at once. The most common pathology I see in troubled programmes is not slowness but overload — everything started, nothing finished.
- How fast work is really completing. Count the cards leaving the last column each week and you have a delivery rate that no amount of narration can spin.
None of this appears on a red-amber-green report, because a status colour is an opinion and a queue is a fact.
Why Programmes Resist the Wall
If flow is so much more honest, why does programme governance cling to status? The answer, I think, is that most governance was designed to reassure, not to decide.
Consider what a steering group is implicitly asked to do in the status model. It receives an assurance that things are broadly on track — or a flagged confession that they are not — and its role is to accept the assurance or to apply pressure. The board is an audience. The report is a performance. And like any performance, it rewards presentation over substance: the lead who can hold amber steady with a confident narrative outperforms the one who honestly shows a stalled queue.
“Governance built to be reassured will always prefer a confident amber to an honest red.”
Flow breaks the performance, and that is exactly why it meets resistance. A board wall does not let a workstream lead narrate progress; it shows the cards, or the absence of them. It exposes overload that leaders would rather not admit to, because admitting it means someone has to stop starting new work. It surfaces the uncomfortable truth that a programme reporting eight greens can be delivering nothing, because all eight streams are stuck in the same downstream bottleneck that no single stream owns.
There is also a scale objection, raised almost every time: our programme is far too large for a wall of sticky notes. It is a fair challenge and it deserves a serious answer.
Flow Does Not Stop at the Team Boundary
The mistake is to assume flow is a team-level idea that cannot survive being scaled up to a programme. In my experience the opposite is true — the larger the endeavour, the more governance needs to see flow rather than status, because the number of hand-offs where work silently queues grows with every additional team.
What scales is not one enormous wall but a discipline of tiers, each showing flow at its own altitude:
- The team board shows the flow of individual pieces of work through delivery.
- The workstream board shows the flow of features or capabilities across the teams that build them — and, critically, the queues between those teams.
- The programme board shows the flow of the outcomes the programme exists to deliver: how many are genuinely in progress, how many are blocked, how many have actually landed.
At each tier the questions are the same, and they are flow questions, not status questions:
- How much is in progress versus finished?
- Where is work waiting, and on whom?
- What is our completion rate, and is it steady, rising, or falling?
Roll-up in this model is not a colour aggregated from the colours below it. It is a genuine view of movement — the same cumulative picture, drawn at a coarser grain. A board can look at the programme tier and see, without a single slide, that the completion rate has been flat for six weeks while the amount of started-but-unfinished work has doubled. That is a governable fact. Three ambers and a red is not.
The Numbers That Replace the Colours
Boards are not wrong to want numbers; they are wrong to accept opinions disguised as numbers. Flow offers a small set of measures that are hard to game, because they are read from the movement of the cards themselves.
- Work in progress — the count of items started but not finished. A rising number is the earliest warning a programme gives, and it appears weeks before any milestone slips.
- Lead time — how long a piece of work takes to travel from started to done. When it lengthens, delivery is congealing, whatever the status colours say.
- Completion rate — how many items actually finish each week. This is the only honest measure of pace, and it is invisible on a report that counts activity rather than throughput.
- Age of blocked work — not merely that something is blocked, but for how long. A blocker three days old is a nuisance; one three weeks old is a governance failure.
Track those four across the tiers and the steering pack begins to write itself — and it writes itself in facts rather than adjectives.
Governance as a Decision Engine
Underneath all of this is a claim about what governance is for. I have come to believe that the organisations which struggle most are the ones that treat governance as a reporting mechanism — a channel for moving assurance upward. The ones that deliver treat it as a decision-making engine — a place where the constraint is made visible and the next intervention is chosen.
Flow-based governance changes the texture of the steering meeting entirely. The conversation stops being talk me through your status and becomes the test column has been full for three weeks — what decision unblocks it, and who owns that decision today? The board stops grading confessions and starts removing impediments. Its authority is spent on the queue, which is the only place spending authority actually changes the delivery rate.
This is not an argument for abolishing reporting. Sponsors, funders and assurance functions have legitimate needs that a wall alone does not meet. It is an argument about primacy: let the flow be the source of truth, and let the report be derived from it, rather than the other way round. When the numbers on the summary slide are read straight off the board the teams already live by, the theatre quietly ends. There is nothing left to perform.
The Quiet Shift
The change I am describing is not a methodology to be installed. It is a shift in what governance chooses to look at. Stop asking people how they feel about their work, and start watching how the work moves. Stop counting whether milestones were declared on time, and start counting whether things are finishing at all.
It is a modest-sounding adjustment, and its effects are not modest. When a board can see flow, it can no longer un-see overload, and the endless proliferation of half-finished work — the most reliable predictor of a programme in trouble — becomes impossible to ignore. The wall does not let you look away.
That, in the end, is the case for it. Not that visual management is fashionable, but that it tells the truth in a room that has grown too comfortable being reassured.