The verdict form is the shortest document in the case and the only one that states, in the court’s own words, what you have to prove. It is also, in most matters, the last thing anyone writes.
That order is backwards, and reversing it costs an afternoon.
The exercise
Take the pattern instructions for each claim you have pleaded. Write out the questions the jury will actually be asked to answer — not your theory of the case, not the complaint’s language, the questions on the form the judge will hand them.
Under each question, list the facts that must be true for the answer to be yes. Be literal. If the element is knowledge, the fact is that a particular person knew a particular thing by a particular date, and you should write down which person and which date.
Under each fact, note two things: what evidence would establish it, and who would be holding that evidence if it exists. The second half is the one people skip, and it is the half that turns into a document request.
Then stop. You now have a grid, and most of it is empty.
The empty cells are the point
A team that has just done this for the first time usually finds the same thing: two or three elements are already well covered, one is covered by a single witness whose deposition has not been taken, and one has nothing under it at all.
That last column is not a gap in your file. It is your discovery plan, and it is a better one than you would have written any other way, because it was derived from what the jury will be asked rather than from what seemed worth asking for.
It also tells you what not to do. A request that produces documents touching no line on the grid will produce volume and nothing else. That is worth knowing before you serve it, because you will be paying to read whatever comes back.
What it does to a production
When the defendant’s first tranche arrives, the grid changes the question being asked of it. Instead of “is this responsive,” which the other side’s lawyers already answered when they decided to send it, the question is “which line does this touch.”
Most documents touch none. That is the ordinary condition of a production and it is not a complaint — but it is only visible if you have the lines written down. A team without the grid reads the same material and calls the day productive because the queue moved.
The documents that do touch a line get filed under the proposition they support, not under the request number they arrived in. Six months later, when someone asks what supports element three, the answer is a folder rather than a search.
Run it monthly
The grid is not a planning artifact to be admired once. Re-run it on the first of the month and record the state of each cell: nothing, contested, or established.
The movement between those states is the only status report worth circulating. “We processed four hundred thousand documents” answers a question nobody asked. “Element three moved from nothing to contested, and here is the document that moved it” answers the only one that matters.
The first few months of that report are uncomfortable, because they say you can prove almost nothing yet. That discomfort is accurate. Having it in month two is considerably cheaper than having it at the pretrial conference.
The obvious objections
The instructions are not settled yet. They rarely are. Use the pattern instructions for your jurisdiction and mark the contested elements as contested. A grid built on the instructions you expect is more useful than no grid, and when the charge conference changes something you will know immediately which cells it touched.
Claims will be added and dropped. Yes, and the grid is how you will price that. A claim dropped is a column deleted, and you can see at a glance which of your evidence was only supporting that column.
It is too early. It is never too early, and it stops being possible to do honestly once the file is large. The exercise is easiest on day one, when nobody has invested anything in a theory yet.
What it costs
An afternoon, once, and an hour a month after that. The output is a page.
Set against the cost of reading a production in the order it arrived, that is the cheapest thing on the case, and it is the only exercise I know of that reliably tells a plaintiff team what it does not yet have.
Draft the verdict form on day one. Then ask of every document that arrives: which line does this touch?