Priya, thank you for raising this. I have strong feelings about it!

In my view the root cause is the matrix model itself. As long as specialists are shared across many projects simultaneously you will always have this problem. The individuals gets pulled in every direction and nobody can honestly say what percentage of their time goes where.

The solution I believe in is dedicated project teams. You assign people to one project at a time, for its full duration or for a defined phase, and you do not split them. Yes it requires more planning and sequencing at the portfolio level. But the gain in focus, quality and predictability is worth it in my experience.

Has your organisation ever experimented with dedicated allocation, even partially?

Yusuf, I respect your experience but I think dedicated teams are a luxury that most organisations cannot afford. If you have twenty projects running and six Java developers, you cannot assign one developer to one project at a time. The queue would be enormous and the business would not accept it.

The problem Priya describes is not the matrix model - it is the lack of capacity visibility. Nobody knows in real time who is at what utilisation. Project managers make resource requests based on availability they cannot actually see. The result is exactly what Priya describes: the same people get pulled everywhere because they are the ones people know and trust.

The fix in my organisation was implementing a capacity management tool that shows live allocation by person and by project. It did not change the structure but it made the overloading visible, and visible problems get solved faster.

I agree more with Aleksei here, but I think both of you are describing symptoms of the same underlying issue: the PMO is not in the conversation early enough.

By the time a project is approved and resourcing is discussed, commitments have already been made. The specialist's manager has already said yes informally, the project manager is already planning with that person in mind. The PMO then tries to enforce a capacity rule that conflicts with a handshake that happened three weeks earlier.

The structural fix I have pushed for is mandatory PMO sign-off before any resource is committed, even informally. It creates friction at the beginning but prevents the bigger friction later. Has anyone managed to implement something like this?

Aleksei, I understand the practical argument, and yes I have seen organisations where pure dedicated teams are not possible. But I would push back on the tool-based solution a little.

In my experience, even when you have perfect visibility of who is at what capacity, the pressure to use the best people does not go away. A project manager will still request Specialist A even if they are at 120% because Specialist A is the one who can do the job properly and the PM knows it. Visibility tells you there is a problem. It does not give you the authority or the mechanism to say no.

Nkechi's point about early PMO involvement is closer to what I am describing. The structural fix is not just a tool, it is a governance decision about who has the authority to commit people to work.

Yusuf, you are right that visibility alone does not change incentives. I admit I oversimplified. What I should have said is: visibility is necessary but not sufficient.

What we added on top of the tool was a policy: no individual can be allocated more than 80% to project work. The remaining 20% is protected for operational tasks, knowledge transfer, and personal development. When a PM requests someone at 90% utilisation the tool flags it and the request goes to a capacity review panel before it can be approved.

It took about six months for people to stop trying to work around it. But the overloading did come down meaningfully. The key was that the panel had real authority to say no, not just to flag concern.

Nkechi that is a great question and honestly it hit home a little. For most of my career I have been in advisory PMOs. The shift to dedicated teams that I described earlier - we were only able to do that because we had a new COO who genuinely believed in it and gave us the authority to enforce it.

Without that sponsorship, I am not sure we could have held the line. PMs would have found ways around the policy. The tool would have been ignored or gamed. The governance panel would have been bypassed.

So perhaps the honest answer to Priya's situation is: before asking what system or policy to implement, ask what executive authority you have behind you. Because without that, even the best system will not stick.

Yusuf, yes I think we are converging. One more practical thing I would add: do not underestimate the functional managers.

In a matrix organisation the specialists belong to the functional managers, not the PMO. If the functional manager says 'my people can handle 150% - they are strong,' then no PMO policy will stick. You need those managers to understand that overloading their people is not a sign of organisational strength, it is a risk they are carrying on behalf of every project that depends on those people.

We spent a lot of time educating functional managers on what burnout and context switching actually cost in delivery terms. It helped more than the tool did in the first year.

The tool and the policy will only hold if the culture allows them to. In my organisation the 80% ceiling worked because leadership had already decided that overloading people was a management failure, not a sign of strength. That belief had to exist before the policy could be enforced.

Where that belief is absent, even a well-designed system gets gamed quietly. Priya, what is leadership's current attitude toward overloaded people in your organisation? That probably tells you more about what is achievable than any governance model.