Evidence and Its Presentation

The instinct when a case matters is to bring everything. Forty slides, three appendices, every figure that supports the position.

This weakens the case, and it does so in ways that are easy to demonstrate and hard to believe while assembling the deck.

What volume does

It signals uncertainty. A person confident in two decisive figures presents two figures. Forty communicates that no individual item was sufficient, which is an accurate signal and not one you intended to send.

It supplies targets. Every additional item is something a sceptic can question, and one weak figure in a set of thirty becomes the subject of the meeting. The strength of a case is closer to its weakest presented element than to its average.

And it buries the decisive material. If two numbers would settle the matter, they are now on slides nineteen and thirty-one, competing with twenty-eight others for a decision-maker’s remaining attention.

The discipline is to identify the two or three items that would move somebody who is undecided, present those, and hold the rest for questions. Everything you brought and did not use was still worth preparing.

Different audiences move on different evidence

Not because some are more rigorous. Because they are accountable for different things.

Boards and finance. Quantified, comparative, and tied to risk. What it costs, what it returns, what happens if it fails, and how that compares to the alternative. A board member has to defend the decision later and needs the defensible version.

Operators. Whether it works in practice, from somebody who has done it. A supervisor is unmoved by a study and moved considerably by a comparable operation that tried it and by what went wrong there.

Technical specialists. Method before conclusion. A specialist who cannot see how you arrived at a number will not accept the number, and presenting the conclusion first reads to them as concealment of the working.

The public. Concrete consequence for identifiable people, in plain figures. Aggregate percentages are almost meaningless in a public meeting; what a thing costs a household is not.

Presenting the wrong type is the ordinary failure. A carefully quantified case delivered to a room of operators, or a practical anecdote offered to a finance committee, will both fail while appearing to be about the merits.

The example and the number

A single vivid case is more persuasive than an accurate statistic and is frequently less true. Anyone presenting evidence should understand that they hold an instrument capable of misleading people who are reasoning properly.

The honest use pairs them. The rate is what establishes the claim; the example is what makes the rate legible. Fourteen percent of jobs in this category ran over last year, and here is what that looked like on the Caldwell project.

The dishonest use is an example standing in for a rate that was never established, or that would not support the point. That is effective, it is recognised as a technique by anybody experienced, and it costs the presenter permanently once noticed.

State the uncertainty

Presenting a projection as a single figure invites two failures. Either it is treated as certain and planned against, or somebody identifies the uncertainty you did not mention and the whole case becomes suspect.

Giving the range and the basis costs nothing and strengthens the position. Between two-forty and three-ten, and the spread is driven almost entirely by whether the second tenant signs.

That sentence also directs attention to the thing that actually matters, which is the tenant rather than the number, and a decision-maker who understands the dependency will make a better decision than one who received a point estimate.

Source and be prepared to show it

Every figure should have a traceable origin, and a presenter who cannot produce one on request has lost more than the figure.

This applies with particular force to numbers acquired secondhand: figures from a vendor’s material, a statistic recalled from an article, an industry benchmark nobody has checked. Each of these is likely to be examined by exactly the person you least want examining it.

Where a number is soft, say so before somebody asks. This is the vendor’s own figure and I have not verified it independently is a stronger position than the same admission extracted in a meeting.

The order

Strongest first, and against the instinct to build.

Attention declines through a presentation, and the last item is heard by fewer people with less capacity than the first. Anything held back for a climax will land on a room that has already formed its view.

The exception is the concession, which belongs neither first nor last. Buried at the end it looks like an afterthought; leading with it gives the objection the frame. It sits best in the middle, stated plainly, once.