I sat through a steering committee last year - two hours, fourteen people, a 47-slide deck - and the only decision that got made was to schedule another meeting. The project sponsor turned to me afterwards and said, genuinely puzzled, "I don't understand why nothing moves between these sessions."
I did. The answer was in the room. Or more precisely, the answer was the room.
Fourteen people can't steer anything. Fourteen people can listen, nod, and raise concerns that get noted in the minutes and then ignored. They can create the warm, comforting feeling that governance is happening. But they can't make the decisions a programme needs to move forward. And the gap between governance that feels like governance and governance that actually governs - that's where most digital programmes quietly start to fail.
The core of it is this: most steering committees are designed for reporting, not deciding. They produce status updates. And a committee that produces status updates isn't a governance mechanism - it's a meeting that makes everyone feel better about the lack of one.
We need broad representation on the steering committee to get buy-in across the partnership.
I hear this constantly. And I understand the instinct - especially in partnership structures where the person who feels excluded from a decision can become the person who blocks its implementation. That's a real political risk. But there's a difference between stakeholder engagement and decision-making authority. You can keep a broad group informed, consulted, and heard. What you shouldn't do is put them all in a room and call it steering. Because what you've actually built is a town hall with an agenda.
I've seen enough of these to spot the patterns. They tend to fail in one of four ways, and most failing committees have at least two going on simultaneously.
Too many members. A committee of twelve cannot make a decision efficiently. It can receive information and discuss it - that's about it. Every member above five adds communication overhead and dilutes individual accountability. When everyone is responsible for the decision, nobody feels personally responsible for it. I've seen this in law firms and accountancy practices particularly - the instinct to include everyone with a stake ends up producing a room where nobody will commit to anything.
Unclear decision rights. This one's insidious. A committee that doesn't know what it's empowered to decide will default to two behaviours: escalating everything it's uncertain about, and rubber-stamping everything it isn't. Neither constitutes steering. I worked with a mid-sized professional services firm last year where the steering committee had been running for nine months and nobody - nobody - could tell me what budget threshold they were authorised to approve without going to the board. So every decision above about £5,000 got kicked upstairs. The board got annoyed. The project team stopped asking.
Too infrequent. A quarterly steering committee is reviewing 90 days of delivery history. By the time a problem is visible in quarterly data, it's been developing for eight weeks. You're not steering at that point - you're doing a post-mortem on a patient who's still technically alive but has been deteriorating since March. The governance cadence should match the intervention window, not the convenience of senior diaries.
Agenda dominated by status updates. This is the big one. A steering committee that spends 45 minutes on progress reports has 15 minutes left for the decisions the meeting was supposed to produce. Status is context, not content. If I'm sitting in your steering committee and someone is reading out a slide that says "Phase 2 is 73% complete," I want to scream. Send me that in an email. Use our time together for the things that require human beings in a room with the authority to say yes or no.
The structural choices are straightforward. Not easy - because politics and organisational culture make everything harder - but straightforward.
Four to five members. Maximum. The accountable sponsor who owns the investment outcome. The delivery lead who owns the programme. An operations or finance lead who tracks cost and risk. And one senior stakeholder whose support is critical to delivery. No observers. No deputies. No rotating membership. The moment you allow "observers" into a decision-making body, you've changed the dynamic. People perform differently when they're being watched. Decisions get softer. Conversations get less honest.
What about the partner who's been excluded and might cause trouble? Address it directly. Brief them separately. Give them a monthly written update and a standing invitation to raise concerns with the sponsor. Create a stakeholder reference group if you need to - a separate forum where broader input is gathered and fed into the steering committee's deliberations. But don't solve a communication problem by breaking your decision-making structure. That's like adding more seats to a car because the passengers can't agree on the route.
Written decision authority. Before the first meeting, write down - literally write down - the decisions this committee is empowered to make without escalation. Scope changes above a defined threshold. Budget reallocation within an agreed envelope. Programme pauses or restarts. Then write a separate list of the decisions that must be escalated, and to whom. This feels bureaucratic. It's the opposite. Clarity about authority is what prevents the committee from either overstepping or - more commonly - dithering. I've written about the board-level governance framework this sits within in a companion piece on what every board should know before approving a digital transformation budget, and the two work together.
Monthly rhythm. Frequent enough to catch problems while they're still addressable. Fortnightly is too much for senior people - you'll get cancellations and deputies, which defeats the purpose. Quarterly is too little - you're reading history, not steering the present. Monthly hits the sweet spot for most digital programmes. If you're in a particularly fast-moving delivery phase, you might go to fortnightly for a defined period. But monthly is the default.
Decision-focused agenda. Status is shared in writing 48 hours before the meeting. The meeting itself is for decisions, risks, and priorities - not information transfer that could have been an email.
I've landed on three standing agenda items that produce steering rather than reporting. Every programme is different, but this structure has worked reliably across the B2B service firms we've worked with.
Decisions required. Specifically, the decisions that cannot be made without committee authority and that the programme needs by the next meeting. Named, with the options and their implications prepared in advance by the delivery lead. Not "we need to discuss the timeline" - that's a topic. "We need to decide whether to extend Phase 2 by three weeks or reduce scope by removing the integration with [system X] - here are the implications of each option." That's a decision.
Risks requiring committee attention. Not every risk on the register. The two or three where the committee's authority could actually affect the outcome. The vendor who's missed two consecutive deliverables and might need replacing. The regulatory change that could affect the compliance module. The budget pressure from another programme that might force a reprioritisation. If a risk doesn't need the committee to do something about it, it belongs in the written status update, not on the agenda.
Progress against outcomes. And I mean outcomes, not outputs. Not tasks completed or milestones passed - the commercial outcomes the programme was approved to deliver. If the programme was funded to reduce client onboarding time by 40%, show me where we are against that target. If it was meant to increase qualified enquiries by 25%, show me the trajectory. An honest assessment of whether we're tracking towards the thing that justified the investment. This is where uncomfortable conversations happen, and it's where they should happen.
Three items. Ninety minutes. Decisions documented and distributed within 24 hours. If your steering committee regularly runs over 90 minutes, you've either got the wrong agenda or the wrong number of people.
A steering committee that defers every contested decision is performing governance theatre. Full stop. But contested decisions are inevitable - you've put senior people with different priorities in a room and asked them to make trade-offs. Disagreement is healthy. Paralysis isn't.
The protocol that works, in my experience: the delivery lead presents the decision options with their recommended choice and rationale before the meeting - in writing, distributed with the status pack. Committee members review asynchronously and flag concerns. The meeting resolves the concerns and makes the decision. If consensus can't be reached within the meeting, the accountable sponsor makes the call, with the dissenting view recorded.
That last bit matters. Record the dissent - not to protect the dissenter, but because a decision that turns out to be wrong is far better reviewed in the context of the objections raised at the time. It creates institutional memory. And honestly, it makes people more willing to accept decisions they disagree with, because they know their position is on the record.
I sat in a steering committee a few years ago where the sponsor overruled the finance lead on a scope addition. The finance lead thought it was premature and said so. It was documented. Three months later, the scope addition turned out to be exactly the right call - it unlocked a dependency that would have caused a much bigger delay. But the finance lead's caution had also pushed the team to build a fallback plan they wouldn't have otherwise considered. Both perspectives made the programme better. That's what governance is supposed to do.
Here's something nobody ever puts in the charter, and they should: a steering committee should have a death date.
A committee that continues after the programme it was set up to govern has been delivered is bureaucracy, not governance. The charter should include a defined end date or a completion criterion - something like "this committee dissolves when the programme has been in operational use for six months and has met its stated success criteria."
The ongoing governance of a delivered programme should move to a different mechanism. I've written a companion piece on the quarterly digital review - that's the rhythm that takes over once the programme steering committee has done its job. The transition should be deliberate, not accidental. Too many organisations let steering committees drift into standing meetings about a system that's already live, attended by people who no longer need to be there, reviewing metrics that no longer need committee-level attention. It's the meeting equivalent of keeping scaffolding up after the building's finished. Nobody's quite sure whose job it is to take it down, so it just stays there.
In a partnership structure - law firms, accountancy practices, consulting firms - exclusion from a governance body carries real political weight. The partner who finds out they're not on the steering committee for a £500k technology programme may interpret that as a signal about their status or their influence. And that interpretation, whether justified or not, can create resistance that derails the very programme the committee is trying to steer.
So manage it proactively. A stakeholder briefing programme running alongside the steering committee - monthly written updates, quarterly face-to-face briefings with the broader partnership, a clear escalation route for concerns - gives people voice without giving them a vote on every operational decision. The steering committee makes decisions. The stakeholder programme builds consensus. Different activities, different forums.
Putting twelve partners in a room doesn't build buy-in. It builds frustration, because twelve partners in a room means nobody gets enough airtime, decisions take forever, and the people who shout loudest win regardless of whether they're right.
If you're about to set up a steering committee for a digital programme - or if you're sitting in one that isn't working and you can feel it - here's what I'd do. Get the sponsor and delivery lead in a room for an hour. Write down the five to ten decisions the programme will need made over the next six months. Identify the four people who genuinely need to be in the room for those decisions. Write a one-page charter with the decision authority, the escalation path, the standing agenda, and the dissolution criteria. That's your steering committee.
If you want the structure as a document you can adapt - decision authority list, agenda format, and charter template - download it here. It's designed to be completable in under an hour for a specific programme. We've used variations of it across dozens of engagements, and the feedback is always the same: "I wish we'd had this at the start."
There's also a companion piece on what good looks like halfway through a digital project - it covers the health check side of things, the diagnostic you run between steering committee meetings to make sure the data going into the room is honest.
Because the best steering committee in the world can't steer if the information it's receiving is wrong. But that's a different problem for a different day.