Part 2: Decision Half-Life
Why Decisions Decay Even When No One Disagrees
This article is part of a five-part series on senior engineering leadership.
In the previous piece, decision containers explained why senior conversations fail to converge and why clarity alone no longer closes anything at Director altitude.
But there is a second failure mode that only appears after a container exists. I call this “decision half-life”
Consider a familiar decision: this quarter’s priorities.
The leadership group debates scope, capacity, and trade-offs. Some initiatives are explicitly deprioritised. Others are named as commitments. There is no confusion about what matters most for now. The roadmap is published, people nod and the decision feels done. And yet, as the quarter unfolds, the organisation slowly stops behaving as if those priorities are true.
The Roadmap Loses Gravity
Nothing dramatic happens.
A team takes on a “small” additional piece of work because a key customer is blocked. Another delays a committed item to clear an operational backlog that keeps interrupting sprint planning. A third keeps a deprioritised initiative alive - just two engineers, just until the end of the quarter.
Each move is reasonable in isolation and no one disputes the priorities. Everyone still agrees with the logic. But by mid-quarter, delivery no longer reflects the original decision. The roadmap hasn’t been overturned, but has faded.
This is not a prioritisation failure. It is decision half-life.
Decision Half-Life
A decision’s half-life is the length of time it continues to meaningfully constrain behaviour once it leaves the room in which it was made.
The term is borrowed deliberately from physics. Half-life describes how long it takes for something to lose half its original strength even if nothing interferes with it. The decay is not malicious, but structural.
Prioritisation decisions behave this way in almost every organisation. Once priorities are set, competing pressures immediately begin to act on them: customer urgency, delivery risk, local optimisation, managerial helpfulness. No one has to disagree. No one has to push back.
The system simply finds it easier to behave as if the priorities are guidance rather than constraint.
Why This Feels Like a Discipline Problem (And Isn’t)
Inside a team, prioritisation decisions tend to have long half-lives. Capacity is shared and trade-offs are visible. When something new is added, something else must visibly move. The cost of deviation is immediately felt.
At Director level, prioritisation spans multiple teams with different pressures and incentives. When one team slips, another compensates. When something new appears, it doesn’t obviously displace anything else.
Publishing the roadmap didn’t do the work.
How Decay Actually Happens
Prioritisation decay follows predictable stages.
From commitment to preference
“This is a top priority” becomes “this is what we’re aiming for.”
From trade-off to exception
New work appears without an explicit de-prioritisation elsewhere.
From visibility to background
The roadmap still exists, but decisions are made without reference to it.
From constraint to narrative
The priorities are talked about mostly as context, not as limits.
At no point does anyone say, “We are no longer following the priorities.” The organisation simply stops acting as if they are binding.
Why Repetition Fails
When this drift becomes visible, many new Directors respond by re-communicating the priorities.
The same slides are shown again and rationale is explained more carefully. This almost never restores the roadmap’s force. Because the decay isn’t caused by misunderstanding. Everyone understands the priorities.
The decay is caused by unopposed pressure.
Unless something in the system change, for eg., how new work enters, how trade-offs are enforced, how deviation is surfaced - the same forces will erode the priorities again. This is why some leaders seem able to “hold a roadmap” without constantly repeating it. They are not better communicators. They are counteracting decay.
Extending the Half-Life of a Prioritisation Decision
If decision half-life is structural, the question becomes: what gives a prioritisation decision more staying power?
It is not process, but the conditions itself.
Where deviation becomes cheap
Priorities dissolve fastest when taking on new work carries no visible cost. Long-lived roadmaps make deviation hard, not forbidden.
What is explicitly non-negotiable
When everything is flexible, nothing holds. A small number of clearly non-negotiable commitments give the rest of the roadmap something to anchor to.
When decay is noticed early
Priorities last longer when there is a known moment where the organisation checks whether it is still behaving as if they are true - before drift becomes normal. This is where, for eg., regular check-ins come in.
Where the roadmap lives
Roadmaps decay fastest when they belong to everyone and therefore no one. Persistence requires a place where drift is noticed and named.
How exceptions are treated
Exceptions are inevitable. Unlabelled exceptions are corrosive. When exceptions are explicit, time-bound, and acknowledged as exceptions, they don’t quietly redefine the rule.
None of these are process innovations. They are ways of increasing a decision’s resistance to entropy.
What Changes Once You Expect Decay
Seeing decision half-life doesn’t just explain frustration. It changes how Directors act.
They stop treating decisions as self-sustaining and assume decay unless something actively counters it.
They stop asking who is “ignoring” the priorities and start asking what pressure makes deviation locally rational.
They make decay visible earlier, before it becomes normalised.
They separate reinforcing a decision from re-litigating it, preserving credibility instead of making every decision feel provisional.
And they become selective about which decisions are worth fighting entropy for.
Not every roadmap item deserves a long half-life. Some should decay quickly as conditions change. Others are bets worth sustaining through friction.
Seeing half-life clearly allows Directors to choose where to spend authority and attention instead of being surprised by where it leaks away.
The Shift
At Engineering Manager altitude, the system preserves prioritisation decisions for you. At Director altitude, preserving them becomes part of the work.
The mistake is being surprised when priorities dissolve without anyone actively sustaining them.
Decision containers explain where prioritisation debates are allowed to end.
Decision half-life explains whether the roadmap survives contact with delivery reality.
Once you can see it, prioritisation failures stop feeling like discipline problems - and start looking like physics.
And that changes what you pay attention to.