A contingency practice has an unusual problem: for most of a case's life, it looks free.
Nobody is billing hours. There is no running total that grows as work happens. The cost is real — it is staff time, and staff time is the largest line in the firm — but it accumulates invisibly and is only compared against the outcome long after every decision that mattered has been made.
So firms end up with a portfolio where some cases are quietly subsidising others, and no way to tell which is which.
The expensive case does not look expensive
Ask which files consume the most time and the instinct is to name the difficult ones. Contested liability, a defence firm that fights everything, an injury that takes eighteen months to stabilise.
Those are demanding, but they are demanding in ways that are visible. Somebody is deciding to spend that time, and the spending is proportionate to something.
The cases that quietly cost the most tend to be ordinary. Nothing about them is hard. They simply require the same small piece of work to be done eleven times: the records that were requested and not chased, the client who calls because nobody told them what was happening, the form filled in from information the firm already holds in three other places.
None of that appears in a case-difficulty assessment. All of it is time.
Three costs nobody counts
Chasing. Every request that leaves the firm and does not come back has to be followed. Chasing is not skilled work, it is not billable in any sense, and it scales linearly with caseload. It is also almost never measured, because each individual instance is trivially small.
Re-explaining. A client who does not know what is happening asks. The question is reasonable and the answer takes ten minutes, including the part where somebody reads the file first to be able to answer. Firms that treat this as an unavoidable cost of client service are half right — it is unavoidable that clients want to know, and entirely avoidable that they have to ring to find out.
Re-entering. The same fact — a date of loss, an adjuster's name, a policy number — typed into a letter, a form, a portal, and a spreadsheet. Each entry is a chance to introduce a discrepancy that somebody will later spend an hour reconciling.
Individually all three are too small to notice. Collectively they are, in my experience, where most of the year goes.
Why "efficiency" is usually the wrong frame
The instinct is to make people faster. It rarely works, because the people are not slow.
What is slow is the shape of the work: a task that requires four systems, a piece of information that lives nowhere authoritative, a next step that has to be decided before it can be done. A capable person navigating that is already performing well. Asking them to perform better is asking them to absorb the design.
The question worth asking is not "how do we do this faster" but "why is this being done at all". Chasing exists because a request had no automatic follow-up. Re-explaining exists because the client had no way to see. Re-entering exists because one fact is stored in four places instead of referenced from one.
Each of those is a structural answer, not an effort answer.
What to do first, if you do one thing
Pick a single case that closed recently and reconstruct where the time went — not from memory, from the file. Every note, call, letter and request, grouped by what it was for.
Most people who do this exercise find the same thing, and find it uncomfortable: the largest category is not legal work. It is coordination — arranging, following up, and telling people things.
That is not an indictment of anybody. It is what happens when a process grows a piece at a time. But it is very hard to keep believing the cost is in the hard cases once you have seen where an ordinary one actually went.
We are fielding a benchmark study of how personal injury firms spend time across a case, and will publish the findings openly. If your firm would like to take part, or to receive the results, get in touch.