See what grows
What should hold.And what holds on Monday morning.
Every organisation has an idea of what should hold. Culture Code shows how much of it reaches everyday work: what works, what slows people down, and the strengths nobody ever wrote down.
Example figures
What should hold
Every organisation has two cultures. The first lives in the mission statement, the values page and the leadership principles. We record each statement with its source. If nothing is written down, the management answers a few questions about what should hold. What it approves is the yardstick for the comparison.
- Step 2 of 3. Everyday work.
What employees experience
The second lives in everyday work. A guided conversation asks about what the mission statement says and follows up where it gets concrete. It is anonymous, runs on each person’s own phone and takes about 19 minutes. Broad like a survey. Deep like an interview. Anonymous like neither.
- Step 3 of 3. The difference.
Where the two match and where they do not
Some things match, some work better than planned, and some never arrive. That difference is what we measure. Every finding appears in the report with the figures it rests on.
Openness & trust · How often it happenedWhen you make a mistake, how often can you raise it without fearing negative consequences?
- NeverContradicting
- RarelyContradicting
- SometimesMixed
- OftenAligned
- Almost alwaysAligned
- Does not happen to meNot scored
- I cannot judgeNot scored
When someone picks an answer at the edge of the scale on a core question, the system asks once about the specific incident. Tap an answer:
Follow-up generated from the last answer
“Never”: What specific case is behind that?
Write here … (free text, names are never asked for)
Why a garden?
Culture is often described as a machine to be fixed, a patient to be treated or a game to be won. None of these fits. A machine has no life of its own. A patient implies illness. A game needs losers.
A garden is alive. It grows, sometimes exactly to plan, sometimes in surprising directions. It needs care, not repair.
And in a garden, everyone tends. There is no single gardener to blame or to praise. That is why Culture Code describes patterns, never people.
When it pays off
Sixty people become two hundred and fifty, a second site opens, or the founding generation hands over. The mission statement dates from a time when everyone knew each other, and nobody quite knows how much of it still holds on Monday morning.
Then the management wants to know what still works in the way things have been done, where things catch between sites, levels and new people, and whether any of it was ever in the mission statement.
With the report it can decide which principles to hand on to the next leadership, where to clarify how decisions are made, and which part of the mission statement to rewrite.
Culture Code also suits mergers and restructuring. For the pilot we focus on growth and handover.
A finding fromthe sample report
A complete report for a fictional logistics company, built from test data. No real person stands behind any figure, and it is laid out like a real report. It is written in German.
- 9
- findings
- 120
- completed interviews
- 190
- employees in the example company
Aspiration rarely experienced: described in company documents and rarely experiencedInvolvement & decisions
„Wer für eine Aufgabe zuständig ist, kann kleine Verbesserungen im eigenen Bereich selbst umsetzen und wird an Entscheidungen beteiligt, die die tägliche Arbeit betreffen.“Eigenständige Verbesserungen und Beteiligung an Entscheidungen werden selten erlebt. Der Unternehmensanspruch, kleine Verbesserungen selbst umzusetzen und an Entscheidungen zur täglichen Arbeit beteiligt zu sein, wird kaum erlebt. Nur 13 Prozent erleben das oft oder fast immer.
Finding 4 of 9, quoted in the original German
The pilot
In the pilot we support a few companies in person. Each covers one site or one unit with 50 to 300 employees, about six weeks from the first document to the joint review.
What you bring
- the documents that set out the standard
- one contact person in the management
- approval of the survey, by the works council, the management or data protection, depending on the company
- a way to send the link to everyone, including shifts and the warehouse
What you have afterwards
A report whose findings are ordered by size, each with its figures beneath it, and a list of next steps that came out of the joint review.
In the pilot, an experienced organisational diagnostician also evaluates the same answers independently. In the final meeting we compare both evaluations: where they agree, where they differ, and why.
- No evaluation below 5 people per group. Smaller groups are merged with a larger one. This is fixed in the software.
- Every question in advance, word for word. Whoever approves the survey receives every fixed question and the framework of the follow-ups before the first person answers. See the sample approval document (German)
We agree the terms in the first conversation. Week by week
What peopleask first
The rest, about the method and its limits, are answered on the method page.
Is it really anonymous?
No name is stored and no personal profile is built. What is recorded is area, line management and tenure in broad groups, and which poster or message the link was opened from. A group of fewer than 5 people is merged with a larger one. This already happens in the database query. Free text is deleted once the report is complete; quotes appear only if each one was anonymised and approved beforehand.
Who approves the survey, and what do they see first?
If there is a works council, it approves the survey, and we involve it from the start. Without a works council the survey is still possible; the company then decides who approves. That can be the works council, the staff council, the management or the data protection officer. Before the first person answers, that body receives every fixed question word for word and the framework within which follow-up questions may arise. The document is generated directly from the question catalogue and describes the survey in full.
How much of the workforce’s time does it take?
A guided conversation in which the AI follows up at most 3 times. By our estimate about 19 minutes, in the browser on their own phone, without a login, and it can be paused at any time. People who work shifts need neither an appointment nor a computer.
Does it replace our existing employee survey?
It measures something else. An engagement survey asks how far people agree and compares with other companies. Culture Code asks about everyday experience and compares it with what the company has written about itself. So there is no industry comparison and no overall score.
Do we get recommendations?
Yes, in the joint review. At the end of the pilot we agree next steps with the management that can be checked: what to watch, by when, and how a change would show. The report itself derives no measures from the figures, because survey data does not show why something is the way it is. Written recommendations in the report can be booked in addition; a person checks and signs off each one.
Who we are
What happens in your organisation on Monday morning?
Let’s find out together. A short conversation is enough to see whether Culture Code fits your situation and where a first pilot would help most.
