Should you hire an operations manager, or outsource it?

Somewhere between eight and fifteen people, most advisory firms have this conversation. The principal is still the escalation point for routine work, the service team is at capacity, and the obvious move is to hire somebody whose whole job is operations.

It is usually the right destination. The question is whether it is the right next step, and the answer turns on one thing: whether the job you are hiring for has been written down.

What goes wrong with the hire

The role gets filled before anyone has defined what it covers. The new person spends three to six months learning how the firm works, mostly by asking the people who have been carrying it, and then they start doing the work the way they learned it.

The process has now moved out of the principal’s head and into theirs. That is a real improvement in the principal’s week and no improvement at all in the firm’s dependency. You have the same problem with a different name on it, and you are now paying a salary for it.

Hiring into an undocumented operation does not remove key person risk. It relocates it, and adds payroll.

The comparison, side by side

Set against each other honestly, the two options differ less in cost than in what you are left holding:

• A hire is a permanent commitment to salary, benefits, and payroll tax, made before the role is defined. An outside build is a fixed scope and a fixed fee.
• A hire takes three to six months to become useful. An assessment runs in two to four weeks and the first systems are in use during the build.
• When a hire leaves, you are back where you started. When a documented system is in place, the procedures and the training stay.
• A hire gives you capacity. A system gives you capacity that does not need managing.

We set this out in more detail on the what we do page, because it is the comparison most firms this size are running, and it deserves to be made in the open rather than avoided.

What the role costs

Firms tend to compare a salary against a project fee, which understates the hire considerably. The real commitment is the salary plus employment taxes and benefits, plus the recruiting cost, plus three to six months of reduced output while they learn the firm, plus the management time the role consumes permanently.

It is also a permanent commitment made at the point of least information. You are defining the role before you have measured the work, which is how job descriptions at this size end up being a list of everything nobody else wants to do.

What to measure before you decide

The decision gets much easier with one week of data. Have whoever is currently absorbing the operations load mark every task against three categories:

• Work that needed judgement, meaning somebody had to decide something
• Work that needed a person but not judgement, such as a client conversation
• Work that was moving information from one system to another

The third category is the one that answers the question. If it is a small share, you have a genuine capacity problem and should hire. If it is a large share, you are about to pay a salary to do work that does not need a person, and the same exercise is set out in how many households one advisor can serve.

When hiring is clearly right

There are firms where the answer is obviously a person. If the work that is piling up genuinely needs judgement, if it is client-facing, if it changes shape every week, or if you are large enough that somebody needs to own vendor relationships and staff supervision full time, hire.

The distinction worth drawing is between work that needs a person and work that currently has one. Most operations roles at this size are a mix, and the mix is rarely examined.

The sequence that works

Build the systems and write the procedures first, then hire into a defined role. The person you recruit walks into a job with an actual description, learns it from documentation rather than from whoever has time, and spends their first month doing the work rather than reverse-engineering it.

You will also hire a different person. The role you are now describing is smaller and more specific than the one you would have advertised, and you are no longer looking for somebody to absorb an undefined amount of institutional memory.

The failure mode nobody plans for

A hire made into an undocumented operation creates a single point of failure with a notice period. Two or three years in, that person knows things nobody else does, and the firm is more dependent on them than it ever was on the principal, because at least the principal was not going to resign.

That is not a hypothetical. It is the most common way firms of this size arrive at us, and the fortnight it produces is described in what to do when your operations person resigns.

The short version

Hire when the work that is piling up genuinely needs judgement, and hire after the processes are written down rather than before. If a meaningful share of the load is information moving between systems, that part is not a staffing problem and a salary will not fix it. The broader framing is in the operations guide.