Early Users Report 40-60% Less Time on Report Preparation. Here Is What That Number Is Worth.

Every prospect asks the same question in the first ten minutes: how much time does this actually save? Here is our honest answer, together with an equally honest account of how much weight it can carry.

The professionals who ran Casealyze on real matters through our private beta — roughly two months of ordinary caseload each — put the saving at 40 to 60 percent of the time they previously spent on the pre-writing phase.

Why we are not dressing this up as a study

It would be straightforward to present this as research. Pick a midpoint, add a decimal place, put it on a chart, and it acquires an authority it has not earned. That is common in this market and it is precisely the habit that has made buyers in legal technology sceptical of every vendor claim, including the true ones.

It is also a bad idea when your buyers are lawyers. The professional instinct of anyone who prepares dispute reports is to ask what a figure rests on. A claim that cannot answer that question does not merely fail to persuade — it costs you the rest of the conversation.

So: a range, attributed to the people who gave it, with the weaknesses stated. Four of them matter.

• Recollection, not records. Users estimated at the end of the trial, comparing against their memory of how these matters used to go. Memory is not a timesheet.
• No baseline. Nobody recorded pre-writing hours separately before adopting the tool, so there is nothing to subtract from.
• Selection effect. These are people who chose to keep using it for two months. Anyone who stopped after a week is not in the sample.
• Small numbers. A handful of practitioners is a signal worth reporting and not a basis for a percentage anyone should plan around.

Where the time was going in the first place

The estimate is more credible when you look at what it refers to. Preparing a dispute report splits into two very different kinds of work, and only one of them is affected.

That split is also why the range is wide rather than tight. A four-thousand-page construction bundle is mostly mechanical work, so the saving sits at the top of the range. A short, well-organised matter that turns on one contested clause is mostly judgement, and the saving is far smaller. Anyone quoting you a single precise figure across all matter types is quoting you an average of things that are not alike.

How to get a number you can actually rely on

Ours is a hypothesis. Yours would be evidence. If it matters to your decision — and for a firm-wide rollout it should — the method is not difficult:

1. 01Record pre-writing hours separately from drafting hours on the next few matters, before changing anything. Without a baseline there is nothing to compare against, and this is the step everyone skips.
2. 02Keep the task breakdown constant across the before and after periods: first-pass reading, chronology, evidence indexing, report assembly.
3. 03Discard the first matter or two after adoption, or report them separately. Those measure learning, not the tool.
4. 04Report the range and the sample size, not just the midpoint — and note what drove the spread.
5. 05Ask what did not improve. If every number in a comparison moves the same way, the comparison is not measuring carefully enough.

We will publish measured figures when we have them, with the method stated and the limitations intact. Until then this is what we have, described accurately.

The techniques behind the change are set out in five ways to cut case file review time, and the structure the recovered time feeds into is in how to structure a legal case report.