Pulse Survey

What a pulse survey is

A pulse survey, also called an employee pulse survey, is a short survey sent to employees at close intervals, whether weekly, monthly or quarterly, with a limited number of questions in each round. It takes the place of a single long survey run once a year: instead of asking everything at one point in the year, the organisation asks a little at a time and asks again at the next interval.

What a pulse survey gains by being short

An annual survey gives a detailed picture of the organisation, but it gives that picture once. By the time its results have been analysed and acted upon, the picture has aged. It describes the organisation as it stood at the moment the answers were collected, and the response to it arrives later, when the situation it described may already have changed.

A pulse survey trades detail for frequency. It asks fewer questions, so each round says less than the annual survey says. In exchange, it produces readings that sit close together in time, and between two close points a direction can be seen: whether the reading is rising, falling or holding steady. A single detailed picture taken once a year shows where things stand at that moment; it does not show which way they are moving.

That exchange makes the pulse survey better suited to following the effect of a specific change than to making a first diagnosis. Where the organisation has introduced a new policy, or has gone through a restructuring, the question it needs answered is whether the situation has moved since the change, and in which direction. A pulse survey answers that question round by round, while the change is still recent. A first diagnosis calls for the breadth and detail that a short survey gives up in order to be short.

The conditions that make a pulse survey work

A pulse survey produces a usable reading only when four conditions hold together:

• Genuinely few questions. Five or fewer. Beyond that, the survey stops being a pulse survey and becomes an annual survey that is repeated at short intervals, with the length of the one and the frequency of the other.
• The same questions in every round. Each round asks the questions the previous round asked. A trend is a comparison of the same measurement at two points in time. If the questions change from one round to the next, there is no trend to measure, only a series of separate measurements that cannot be set against one another.
• A minimum group size before results are shown. A group’s result is not displayed until the number of respondents in that group reaches a set minimum. The purpose is to protect the identity of respondents in small teams, where a result shown for a handful of people can reveal who gave which answer.
• An announced closing of the loop. After each round, the organisation states what changed as a result of the previous round. This is the decisive condition of the four, and the next section describes what follows when it is missing.

How use can spoil a pulse survey

A question that is repeated with no visible effect can produce something worse than not asking at all. Employees answer a round and see nothing change as a result. They answer the following round and again see nothing change. Over successive rounds, the team can learn that answering changes nothing.

Once that lesson has been learnt, the participation rate can fall. The employees who keep answering are those with a strong reason to answer, and that reason runs in one of two directions: enthusiasm or resentment. The employees between those two positions, who had no strong reason to answer in the first place, are the ones who stop.

The result can then be biased in both directions at once, towards favourable answers and towards unfavourable ones, and it is nonetheless taken as the voice of everyone. The figure still looks like the view of the whole workforce, but it now describes only the part of the workforce that still chooses to respond. Nothing in the figure itself shows that this has happened, because a shrinking and polarised group of respondents still produces a number.

What a pulse survey number does not say

A pulse survey shows that the indicator has moved. It does not say why it moved. The figure records a change between one round and another, and the reason for that change lies outside the figure.

The reason comes from a conversation. That conversation can take the form of a stay interview, a one to one meeting, or an open question attached to the round, in which employees explain in their own words what lies behind the answers they gave.

A decision built on the number alone addresses the symptom that was measured, not the cause that produced it. If the indicator falls and the organisation responds to the fall itself, without asking what lies behind it, the response is aimed at the reading on the survey rather than at the condition that moved the reading.

How a pulse survey differs from the annual engagement survey

The annual engagement survey is a diagnostic tool. It is broad, it is detailed, and it draws a map: in a single round it covers a wide range of questions about work and shows where the organisation stands on each of them.

The pulse survey is a monitoring tool. It is narrow, it is repeated, and it tracks movement: it covers a few questions and shows how the reading on them changes from one round to the next.

The two complement each other. The annual survey shows where things stand; the pulse survey shows which way they are moving. Using either one in place of the other loses something: an organisation that relies on the pulse survey alone loses depth, and one that relies on the annual survey alone loses direction.

This is an explanation of the concept, not legal advice.