top of page

What Does a Deadline Ask of Everyone?

Writer: Riann Smith
Riann Smith
4 hours ago
4 min read

Part 3 of 3


What organisation wouldn't want approval rights?


It makes complete sense. They're paying for a change they'll have to live with, run and explain, and they want to be sure it serves them. It also gums up the works in the worst possible way.


Approval is a way of controlling the flow of work. Controlling the flow also gives you the ability to block it. Nobody tends to see that at the moment the process is agreed, because an unanswered question looks like an item on a list, something to come back to after the next working group. Meanwhile an engineer is switching tasks, a feature is waiting at 98%, and a test team can't validate the journey.


"Late decisions may delay delivery." True. Also useless. It's so beige it blends into the wall. Everyone nods at it, because it's accurate, and nobody takes in what it means for them on a Tuesday afternoon.


How long can a small choice take?


Six months. That's how long it took to agree which product attributes customers would see when comparing products.


Designs were signed off in the spring and the question went to the client straight away. It then sat for a couple of months, which was fair enough, because they had plenty of other things competing for their attention. It reached a working group workshop in September, went off to committee afterwards, and came back with more questions, when all we'd been hoping for was a list.


What moved it was a mindset conversation. We weren't the experts on what makes a useful data point, and we had neither the time nor the budget to do the exploratory work and come back with a recommendation. So I pushed on a different question: what's the MVP? What's the smallest set of attributes that's useful? We all knew more could be added quickly and easily later. This was about what they'd go live with, not what they'd have to live with, and since their own team was being brought in to support the build, they'd have the power and the knowledge to manage it afterwards. Knowing they could change their minds is what finally got it decided.


None of that makes reversible decisions free. Some small choices deserve proper investigation. But how reversible a decision is, and what it costs to keep waiting, belong in the conversation too.


What happens when nobody is actually objecting?


On a university programme, IT were fairly sure a senior leader, the CIO or the CMO, wouldn't like a particular route. We flagged it as a blocker in a senior meeting and asked the obvious thing: could you help us understand the objection, so we can work with it or around it? And we were told, in that session, that they didn't care. Not that they weren't invested in the project. They just had no strong view on this point, and were happy to go with our recommendation as long as we documented it.


It was a game of telephone. An opinion somebody believed somebody else held, which turned out not to exist. Nobody was at fault. The person who passed it on did so in good faith, trying to protect the project and its progress, from inside a world of extraordinary red tape where that caution was exactly how every other project survived. This one had more visibility, more vested interest and more pressure to show a return on the investment, so some of the old rules simply didn't apply, and they had no way of knowing that. It seemed logical to me, but that's my world, not theirs.


That conversation is what led us to ask for delegated authority for the relevant team, so the people who could actually decide were the people in the room. The first release went live inside twelve weeks. The digital marketing lead demoed it live to a large audience, having gently warned the room about the curse of the live demo. It behaved.


Railway tracks curve and branch through a junction.


What do we each owe the date?


As a delivery partner, we owe the client early sight of dependencies, so there's time to act on them. Where we have the expertise, we owe them a recommendation. Where we don't, we should help make the choice and its consequences clearer. Sometimes that's a decision matrix, a method for weighing the options, or coaching on what matters now and what can safely be revisited later. They have the expertise and the authority. Our contribution is to help them use both without getting stuck trying to settle every possible future question. We also owe them a plain explanation of what each decision holds up, and an honest stop on presenting a date as dependable when the assumptions behind it aren't.


The client owes something too. Where it keeps a decision or an approval, the timing of that decision becomes one of the programme's delivery conditions. If the timing moves, the effect on dates, sequence, effort or cost needs to be acknowledged, rather than quietly absorbed by the people downstream.


I'm not arguing for more governance, because another register or another approval step is how safeguards become part of the delay. What I'd carry forward is questions, and the judgement to use them. Which decisions need broader approval, and which can sit with the people closest to the work? What do those people need to know? When does the sequence need an answer, and what happens if it doesn't arrive? A decision with little consequence doesn't need the treatment that one controlling several integrations and a release does.


And when delegation works, say so, and be specific. Telling a senior leader exactly what their quick decision made possible does more to repeat the behaviour than any process will.


Engineers need usable work. Decision-makers need information and the authority to act on it. Test teams need something complete enough to validate. Sponsors need to face the trade-offs when the conditions change.


A date depends on commitments from people across the programme. The forecast becomes believable when those commitments are understood, achievable and owned.



Earlier in this series: Part 1 · Part 2

 
 
 

Comments


bottom of page