Nobody can see a culture. They can only see what it makes people do.

Culture is an iceberg in weather. Values are the submerged mass: the organising principle, larger than what shows, shaping everything above it. Skills are the visible peak, the behaviour a colleague can actually observe. Problems and celebrations are the weather: a problem is a storm that strains both halves, and a celebration is sunshine that lifts them. And the waterline is where the daily insight works, where the submerged and the visible meet.

That is four nouns, and the useful thing about them is that they are not a vocabulary you have to learn. They are the four things any manager already talks about - what we believe, what we are good at, what is going wrong, and what is going well - arranged so it is obvious which one you are looking at.

Everybody draws the iceberg with values driving behaviour. That direction is real and it is also the half that has produced twenty years of disappointing culture work: if values come first, the intervention is a values exercise. A new mission statement. An offsite. A poster in the stairwell. And then a year later, the same company.

The arrow points both ways. Behaviour practised often enough reshapes the values underneath it, and that is the direction with the evidence behind it. It is what implementation intentions are, in the research they come from - a plan that fires because it was rehearsed, not because it was agreed. And it is what a team does when it chooses its own norms and then actually practises them, until the way people approach a problem changes.

This is why the product is a daily rep rather than a workshop. You cannot argue a company into a value. You can give it a small amount of practice, every working day, for long enough that the practice becomes the thing it believes about itself.

A re-org, a missed quarter, a redundancy round, an acquisition rumour: storms. A launch that shipped, recognition that landed, a target beaten: sunshine. Neither is culture, and both change what culture looks like from the inside while they last.

This matters more than it sounds, because it is the single most common way a culture measurement gets misread. A team that goes quiet during a restructure has not suddenly become disengaged; it is a team in weather. Reading the dip as a verdict on the people is how companies end up treating a storm with a performance process.

Here is the part most culture products will not say out loud. The submerged half is exactly what an engagement signal cannot see. Nothing we measure reads what anybody privately believes, and nothing ever will, because the only honest inputs are what people opened, answered and said was useful.

So the claim we make is deliberately smaller than the one the picture invites: we work at the waterline. We read and train the observable surface. When our own reporting says a team engages with skills material and goes quiet on values material, that is a statement about attention - about which content lands - and it is never evidence that anybody does or does not hold a value.

The alternative is to ask people to rate their own beliefs on a five-point scale four times a year, which is what the rest of this category does, and what a survey measures is the survey.

The daily insight reads and trains the surface. One short thing every working day, to the whole company, on the themes it chose. It is the multivitamin: it holds the posture whether or not anything hurts, and it is never targeted and never diagnostic.

A module treats below the line. When something goes wrong - a named problem, the vocabulary a manager already diagnoses in - the team that needs improvement gets two weeks on it: a short course of ten daily lessons on that one problem, booked onto the team and never onto a person, with two ten-minute debriefs its lead runs from a written guide. It works downward by working on the surface, which is the return leg of the arrow doing its job.

Which teams a module is for, and what each one books, is on the homepage.

The iceberg has a real lineage. Edward T. Hall used it for culture in the 1970s; Edgar Schein’s levels of organisational culture are the reference most practitioners actually mean when they draw one. We cite it as after Schein and Hall and never as Schein’s model, because his has three levels - artifacts, espoused values, underlying assumptions - and ours compresses to two.

That compression is a real simplification and worth naming rather than hiding: the distinction Schein draws between what a company says it values and what it actually assumes is a genuine one, and our two layers fold it into the submerged half. We drew two because two is what a product can act on. The weather is the addition, and it is ours.

The model also earns its place by organising something that already existed rather than replacing it. The themes a customer picks from have always been typed - values, skills, problems, celebrations - so adopting the iceberg changed no data at all. It changed what the four types mean in relation to each other: problems and celebrations are the weather, and booking a module on one is the moment it arrives.