Why do senior staff become the bottleneck in a professional services firm?

Is the bottleneck really the partner?

Ask the people in almost any firm of 20 to 200 people where work gets stuck, and they will name someone senior. The proposal is waiting for a director to write the approach section. The report is waiting for a partner to rewrite the executive summary. The advice letter is drafted but sits in a queue for sign-off. The firm's throughput is set by the diaries of a handful of people.

The usual diagnosis is that those people need to let go. Partners are told they hold on too tightly, that they should trust their teams and delegate more. There is some truth in that. But in most firms we think the diagnosis is wrong, and that is why the advice rarely sticks. The partner is not the problem. The partner is where the problem becomes visible.

Why does telling partners to delegate not work?

Delegation assumes the work can be handed over in a form someone else can complete. In a lot of professional services work, it cannot. The knowledge needed to finish the job is not in the file, the template or the procedures manual. It is in the partner's head: which clauses this client always pushes back on, which phrasing the regulator responded well to last time, which risks matter on this kind of scheme and which do not.

So the junior drafts, the partner reads it, and the partner rewrites large parts of it. That rewrite is not a failure of trust. It is the only point in the process where the missing knowledge gets applied. Telling the partner to delegate more simply moves the rewrite later, and often makes it longer, because the draft has drifted further from what was needed.

A partner who has been through this cycle a few times draws a rational conclusion: it is quicker to write it myself. Everyone can see that this is expensive. Nobody can see an alternative, because the work has not been set up so that anyone else can do it well.

Where does senior time actually go?

If you track a senior fee earner's week honestly, the time spent on genuine judgement is usually smaller than people expect. A great deal of the rest is assembly: finding the last similar document, pulling figures from two systems, reformatting a colleague's section, checking that names and dates match, rewriting generic paragraphs so they read like the firm. None of that needs a partner. It falls to them because they are the only person who knows what good looks like, and it is faster to fix it than explain it.

This is the cost that rarely appears in a management meeting. Partner time is the most expensive time in the firm and the time most closely tied to winning and keeping clients. Every hour of assembly is an hour not spent on advice, business development or developing the people below. You can put a figure on it by counting those hours, multiplying by the number of senior staff and their charge-out rate, and running it across 46 working weeks. The guide to calculating unbillable hours walks through the method.

Why does hiring more juniors make it worse?

The instinctive response to a backlog is to add capacity underneath. More associates, more assistants, more graduates. If the constraint were drafting capacity, that would work. But the constraint is review and rework, and each new junior adds drafts to the partner's queue without adding any of the knowledge that would let those drafts arrive finished.

A firm that responds to a partner bottleneck by hiring below it often finds the partner busier a year later, with a larger team that is less well supervised. The choice between adding people and changing the work deserves more thought than it usually gets, and we set out the trade-offs in automation or hiring.

What actually takes the pressure off?

The work has to reach the partner in a different state. Instead of a rough draft that needs rescuing, the partner should receive something close to finished, built from the firm's own best material, so that their time goes on the decisions only they can make.

That means moving the knowledge out of their head and into the process. In practice it looks like this:

• The partner's corrections become rules. If they always rewrite the risk section in a certain way, that way is captured once, in the firm's language, and applied every time.
• The best past work becomes the starting point. Not whichever document happened to be open, but the approved version the firm is proud of.
• Assembly happens before review, not during it. Figures, names, dates and standard sections are already in place and already checked when the draft arrives.
• Review is a decision, not a rewrite. The partner reads, amends where judgement is needed and signs off. A named person still approves everything that goes to a client.

This is where AI earns its place in a firm. Not as a chatbot a partner types into when they have a spare minute, but as the machinery that carries their standards into every draft. The partner stays the final word. They stop being the first draft.

Where should a firm start?

Pick the one document that most often waits on senior time. For many firms that is the proposal or bid, where the approach and experience sections sit with a director until the last evening before the deadline. For others it is the report that goes to the client every month, or the scope and engagement documents that open every new instruction. The proposals and bids page shows what a rebuilt version of the most common one looks like.

Then do one thing properly. Sit with the partner and the people who draft for them, capture what the partner actually changes, and rebuild the job so those changes are made before the draft lands. When that single process stops queuing on one person, the firm learns more about its bottleneck than any amount of delegation training would teach it.

There is a longer-term benefit too. Knowledge that lives only in senior heads is a risk the day that person retires or leaves, a problem we come back to in when knowledge leaves. Rebuilding the work around it protects the firm as well as freeing the partner.