Four things worth counting in your next hard conversation

Ask a manager how the conversation went and you get one of three answers. Fine. Better than I expected. Awful.

That is the entire measurement system most of us run on, and it has the same flaw as every other self-report: it records how you felt, and how you felt is mostly a record of how frightened you were beforehand. Relief is not evidence.

The alternative is not a scorecard. It is four things you can count, which have the useful property of being true whether or not you enjoyed yourself.

1. How long before you said what it was about

Start the clock at hello. Stop it at the first sentence that names the actual subject.

Most people are startled by this number the first time. Four minutes, six minutes, a full ten if there is a holiday to ask about. All of it spent in a state both of you recognise as a run-up, which means they have been braced the whole time and are now listening to you rather than to the point.

Under a minute is a reasonable aim for most conversations. Thirty seconds if the subject is someone’s performance. That is not brutality, it is the opposite: the fastest way to make it less frightening for them is to stop making them wait.

2. How many times you took it back while saying it

Count the hedges. Just, a bit, sort of, kind of, maybe, slightly, I mean, I think, no big deal, if that makes sense.

Some of these are normal social lubricant and a conversation with none of them sounds like a robot delivering a verdict. The number is not supposed to be zero. What matters is the change in the number: most people hedge twice as much in the ninety seconds after the other person pushes back, which is exactly when the message needs to hold still.

So count them twice. Once in your opening, once in the two minutes after the first sign of resistance. The gap between those two counts is the most honest thing you will learn about yourself all month.

3. How much of it was you

Roughly what share of the talking was yours? A third, half, most of it?

In a conversation where you need information, which is nearly all of them, more than half is a problem. Zenger and Folkman’s analysis of 3,492 participants in a development programme found the best listeners were not silent sponges: they asked questions that moved the conversation on. But asking questions and holding the floor for eighty per cent of the meeting are difficult to do at once.

Stanford’s work on over-talking points at the mechanism. Most over-talking is anxiety management. You keep going because stopping would leave a gap, and the gap feels like failure.

4. The longest silence you let stand

This is the one people find hardest, and the one with the most interesting evidence behind it.

Jared Curhan and colleagues at MIT Sloan measured silences in negotiations, using a threshold of at least three seconds. Pauses of that length tended to come just before breakthroughs, and appeared to interrupt the reflexive, zero-sum mode of thinking on both sides in favour of something more deliberate. Both parties did better, not just the one who paused.

That was negotiation, not a performance conversation, so do not treat it as a law. Treat it as permission. Three seconds of silence feels, from the inside, like about ten. Ask your question, then count to three before you rescue anybody.

How to get the numbers without becoming strange about it

Do not record a colleague to collect data on yourself. In most places that is a policy problem, and everywhere it is a trust problem, which is the exact thing you were trying to build.

Three honest ways:

• Count them in rehearsal. Practise the conversation out loud with a colleague or into a voice memo of yourself, and count everything on the playback. Nothing stops you being accurate about a recording of you.
• Estimate straight afterwards. Five minutes after the real one, write four numbers down before the memory smooths itself over. They will be rough. Rough and written down beats precise and imagined.
• Borrow someone’s eyes. In a meeting where a colleague is already present, ask them beforehand to note two things: when you got to the point, and who talked more.

Compare with yourself, and nobody else

There is no good benchmark for these numbers across people, and anyone selling you one is guessing. A quiet manager and a talkative one can both be effective.

What is meaningful is the same person over time, which is also what the training research keeps pointing at: Lacerenza and colleagues found that leadership training worked better when it started from a needs analysis and included feedback. Your four numbers are a small, cheap version of both.

They also fix the problem in the self-assessment evidence: people judge themselves more accurately when the question is specific. “Am I good at this?” is unanswerable. “Did I get to the point inside a minute?” has an answer, and sometimes the answer is no.

Pick one

Four at once is too many. Choose the one you already suspect, run it for a month, and write the number down every time.

For most first-time managers it is the first one, because getting to the point feels unkind right up until the moment you watch someone relax when you do it.