Communication skills metrics: five states, proxies, and one direct check

"Did that land?" is the only question that matters about a message, and it is almost never asked, because the answer is uncomfortable and there is no obvious way to check.

There is a way to check. It starts with noticing that landing is not one event. It is five, and each one has its own evidence.

What to take away

The five states, defined

The table is ordered, and each state depends on the one above it. That

Where the proxies break

Read receipts measure a window, not a mind. A message can be open on a

A reply measures politeness at least as well as comprehension. "Sounds good" is a social move. It is compatible with having understood, having skimmed, and having decided to deal with it later.

Attendance measures availability. A meeting with everyone present and no decision recorded has produced a record of a calendar, not of a communication.

Silence measures nothing at all. It is compatible with agreement, with disagreement, with not having seen it, and with waiting for somebody else to speak first. Reading silence as consent is a decision, and it should be declared as one. Reading silence as consent is one of the communication skills mistakes that a readback repairs.

The one direct check

Ask for a restatement. Not "does that make sense", which invites yes, but

something closer to: tell me what you are going to do, and by when.

A restatement is direct evidence: someone who has not understood cannot produce it. It works against the curse of knowledge, the difficulty of imagining what a sentence sounds like to someone who does not already know what it means.

It surfaces the specific place where the message went wrong, which no proxy can do. The cost is a few seconds and a small amount of social friction, which is why almost nobody pays it.

The same technique appears in safety-critical work under other names, and the principle is stable: the sender is responsible for confirming receipt of meaning, not just receipt of words. The communication skills guide treats this as part of listening rather than as a measurement.

Measuring the document rather than the reader

Document-level measures are easier to collect and further from the thing you

care about. Sentence length, word length and structure can be counted; whether a reader can act on the page cannot.

That does not make them useless. Length and structure are worth watching as warnings rather than as scores, and the jargon metrics guide sets out which document measures are worth keeping and why none of them should become a target. What matters here is the boundary: a document measure is evidence about the document, never about the reader.

Federal guidance on testing a document with real readers takes the same position, and recommends watching someone try to use the page rather than scoring it.

Building a check you will actually run

1. Decide, before sending, which of the five states you need. Most messages need understood. Some only need delivered.

Step four is where most people go wrong. Resending an unclear message more loudly is the default response, and it treats a design fault as an attention fault.

What cannot be measured

Trust, willingness to raise a problem, and whether someone felt able to disagree. These are not measurable from the outside, and constructing a number for them produces a figure that moves when people learn it is watched.

The honest position is to say that these matter, that they are not observable, and that the substitute is a standing invitation to contradict you plus evidence that contradicting you has been survivable. That is a practice, not a metric. The skill assessments guide makes the same argument about what no instrument can see.

Common questions

Is a readback appropriate with a senior person?

Yes, if you supply it rather than request it. Say what you understood and what you will do, and invite correction. That puts the burden of restatement on you and gets the same information.

What if the reader restates it correctly and still does nothing?

Then the failure is at agreement or action, not understanding, and the remedy is different. Adding clarity to a message that was already clear is wasted effort and reads as nagging.

Should I count how many messages I send?

Only if you plan to send fewer. Volume is a measure of your behavior, not of anyone's understanding, and it is useful mainly as a check on whether you are substituting frequency for clarity.

How do I know if my writing is the problem?

Watch one person use it without helping them. The business vocabulary guide is a fair test bed: hand someone a paragraph with a term in it and ask what they would do next.