Satisfaction and Climate

A finished study produces two indices, not one. They are computed over the same responses and they answer different questions.

What each one measures

Satisfaction measures how content people are with the various aspects of their work. Its categories, and the way they combine, are defined by the theoretical model you chose. See Theoretical models.

Climate measures how the working environment is perceived as a whole. It has its own model and its own categories, independent of Satisfaction's.

They are two different questions about the same people. An organization can come out well on one and badly on the other, and that is not a contradiction: it is information.

Why they never share a chart

Both indices come out on the same numeric scale, and that coincidence is a trap: they are built with different methods, so the difference between them means nothing.

«Climate is six points below Satisfaction» produces a number, and the number describes nothing about the organization — it describes the distance between two ways of calculating.

That is why the product separates them in charts: two lines on a shared axis are, visually, a claim that the distance between them can be read. The chart shows one index at a time.

They do coexist in tables, and that is not a contradiction. A table with both columns does not propose a common axis: each column is read on its own, against its own history. What is forbidden is the shared axis, not the adjacency.

So how do you read them

Each index against its own series. The question that has an answer is «did the Northern Region's Climate go up or down compared to last year?», and the one that does not is «is Climate better than Satisfaction?».

The same rule holds across nodes: comparing the Climate of two branches is legitimate — same method, same scale, same instrument. What you never cross is the two indices against each other.

You can measure just one

Measuring both is not mandatory. At the top of Settings → Survey forms there are two switches, «Measure Satisfaction» and «Measure Climate», just like the one for eNPS: turning one off removes that section from every questionnaire, and new studies stop asking it and stop computing that index.

• One always stays on. eNPS alone is not enough for a study, so the switch that is the only one left on cannot be turned off.
• It cannot change while a study is running. It would change the questionnaire people are answering; change it before launching or after finishing.
• Each study remembers what it measured. What gets measured is fixed at launch, so moving the switches later changes nothing about a study already launched.

A study that did not measure an index is a gap, not a break. In Statistics, looking at an index shows only the studies that measured it, and in tables that carry both the missing one reads «Not measured». The series runs from one study that measured it to the next one that did, and so do targets: the target of the next study that measures it is computed from that index's last result. See The historical series and Improvement targets.

Turning an index off deletes nothing. Studies that already measured it keep their results, and if you turn it back on, its history continues where it was.

Each has its own model

Because they are parallel indices, choosing the theory behind one says nothing about the other: each questionnaire points at two models, one per index. It also means a historical series break is evaluated separately for each — changing the Satisfaction model does not touch Climate's history.

See The historical series.