What is really happening in a new joiner's first three months?
A firm hires a qualified person with relevant experience. They start on a Monday. Within a fortnight everyone agrees they are capable, pleasant and clearly know the subject. Six months later people are still saying they are not quite up to speed.
Look closely at what they do in those months and very little of it is learning the profession. They already know the profession. What they are learning is this firm: which template is the current one, who signs off what, how the fee note is worded, where the precedent sits, which client hates bullet points, what gets filed and where, and the twenty small conventions nobody thought to mention because everyone else absorbed them years ago.
That is not knowledge work. It is archaeology. The new joiner is reverse engineering the firm's method from old jobs, half remembered advice and whatever they can get out of a colleague between meetings.
Why is ramp-up time a measurement rather than a fact of life?
Here is the useful way to think about it. The time it takes a competent hire to become useful is roughly proportional to the amount of the firm's method that exists only in people's heads.
If the way a scope is written, a report is structured, a file is checked and a bill is raised all live as tacit habit, a new person can only acquire them by osmosis. Osmosis takes months, because it depends on happening to be on the right job, with the right colleague, at the right moment. If those same things exist as a defined sequence with the firm's own templates and checks attached, the same person can do a passable version in week two and a good one in week four.
So ramp-up time is not a fixed cost of hiring. It is a readout. A long one tells you that your method is undocumented, and that every new person has to rebuild it privately from scratch. It also tells you something about how much the job already varies between your existing people, because a new joiner copies whichever version they happen to sit next to.
Why doesn't hiring better people shorten it?
The instinct, when ramp-up is slow, is to blame the hire or the hiring. Next time, recruit someone from a firm that does the same kind of work. Pay more. Be fussier at interview.
This helps less than people expect. A joiner from a similar firm arrives with a fully formed method of their own, learned at their last place. On day one that method is wrong here: different templates, different review culture, different filing, different client expectations. They are not starting from nothing. They are starting from something that has to be unlearned, which can be slower.
The second problem is that the cost of ramp-up falls on the people you can least afford to spend. The questions go to whoever is nearest and most senior, which is usually the same small group who already carry the sign-off load. Every new hire therefore makes the partner bottleneck worse before it makes it better. Firms that hire to relieve pressure often feel more pressure for two quarters, then wonder why.
And every time an experienced person leaves, the undocumented method leaves with them. We have written about what walks out of the door with a departure. A firm with a long ramp-up and no documentation is exposed twice over: slow to absorb people, and badly hurt when they go.
Where does the time actually go?
If you watch a new joiner's week, the lost time clusters in recognisable places.
- Finding the right version. They open three documents that look like the template before someone tells them which is live. This is the everyday consequence of templates that drift.
- Guessing the structure. They write a report in a sensible shape that is not this firm's shape, and a reviewer rewrites it.
- Waiting for an answer. A question sits in someone's inbox for a day because the person who knows is in court, on site or with a client.
- Rework after review. The first few jobs come back heavily marked, not because the thinking was wrong but because the checks were not what the firm does.
- Admin they were never shown. Time recording, file opening, conflict and onboarding steps, all learned by getting them wrong once.
Almost none of this is about judgement. It is mechanical, which means it is the part that can be fixed.
What changes when the core processes are rebuilt?
When a job is rebuilt properly, the firm's method stops being folklore and becomes the way the work moves. The sequence is defined. The current template is the only one the system offers. The checks run as part of producing the output rather than as a reviewer's memory test. A draft comes out in the firm's own structure and language, and a named person reviews and approves it.
For a new joiner, that turns an invisible method into a visible one. They are not asked to guess the house style, because the draft already has it. They are asked to apply judgement to the content, which is what you hired them for. The review conversation is about substance rather than formatting.
Two supporting pieces matter. Written procedures that are actually current, and a way to find the firm's own precedent and guidance without interrupting a colleague. Template control removes most of the version hunting.
How does this compare with simply hiring more?
Put the two options side by side honestly. Hiring adds capacity but adds it slowly, at full cost from day one, with a drag on your senior people while it beds in. Rebuilding a core process adds capacity from the people you already have, immediately, and then shortens the ramp-up of every future hire as a side effect. That second effect compounds and is almost never counted when firms weigh automation against hiring.
What should a firm do about it this quarter?
- Ask your last two joiners what slowed them down. Do it in a conversation, within their first year while they still remember. They will name specific documents and specific unanswered questions. That list is your documentation backlog, written for you.
- Pick the job a new person has to learn first. Usually the recurring output: the report, the file note, the client update, the proposal. Define it once, in the firm's own language. The guide to documenting a process sets out how, and preparing templates covers the inputs.
- Count the supervision time. Hours spent answering the same questions are unbillable hours like any other. The unbillable hours calculator turns them into a yearly figure at your own charge-out rate.
This bites hardest where output is recurring and house conventions are strong, such as accountancy firms and engineering consultancies, where a new person cannot produce anything client ready until they have learned a format nobody wrote down. To find which process is costing you most, the audit will point at one.