Mirror, Map and Compass

TL;DR

Mirror, Map and Compass: why one survey reaches from a person to a culture

  • Mirror, Map and Compass are three readings of one survey: yourself, the people you work with, and the organisation as a whole.
  • Culture is "the way we do things around here" (Bower, 1966). What people do follows what they prefer, so culture is measurable through preference.
  • Schneider showed the people make the place: who is attracted, chosen and remains shapes the culture. Aggregate their preferences and the shape appears.
  • Nothing is collected twice. Each person re-takes the survey periodically; pair, team and culture readings are built from the same data.
  • Read against Kotter's eight accelerators, the same data shows how ready the culture is for each one, on a five-point temperature computed by rule.

Where does a culture come from? People. Not the values on the wall, and not the strategy document, though both matter. A culture is what the people in a place actually do, repeated until newcomers copy the behaviour without being told. That sentence has held up for sixty years of research, and the research points somewhere useful: if culture is built from what people do, and what people do follows what they prefer, then a good reading of individual preference is also, in aggregate, a reading of the culture.

That is the argument of this article, and the reason Sariio uses three plain words for one survey. A mirror, for the person. A map, for the people beside you at work. A compass, for the organisation. Three instruments, one source of data. This piece sets out why each reading stands up alone, and why the step from one to the next holds.

One survey, three readings

Start with a definition, because the words carry weight. Sariio calls a person's reading a map: a picture of how someone prefers to connect, think, decide and get things done, drawn from a survey that takes about 12 minutes. The survey asks about language, behaviour and motivation rather than about mood or opinion. Preference, in this sense, is the way of working a person tends to choose when choice is available. Preference says nothing about ability, and a preference is never a verdict on anyone. The full account of what is measured is in What are work preferences?.

What are Mirror, Map and Compass in Sariio?

Three ways of reading the same survey. The Mirror is a person's own reading: how they prefer to connect, think, decide and get things done. The Map places two or more readings side by side, for a pair or a team, so people can see where preferences meet and where they differ. The Compass aggregates the readings of a whole organisation, anonymously, to show the direction the culture already travels. No separate survey is needed for any level; each reading is re-taken periodically and the levels are built from the same data.

The three instruments are three distances from that reading.

Mirror One person's reading. How you prefer to work, in plain language, and what that preference costs and earns. The myMAP.
Map Two or more readings side by side. Where a pair or a team meets, where the preferences differ, and how the people work together. Compare and the teamMAP.
Compass Every reading in the organisation, anonymised and aggregated. The direction the culture already travels. The cultureMAP.

The order matters. A compass reading with no mirror beneath the needle is an average of strangers. A mirror with no map is a private insight nobody else can use. Sariio reads in one direction, from the person outwards, and each level draws on the same readings as the one before. Nothing is collected twice. People re-take the survey at least twice a year, closer to quarterly where coaching is active, because preferences move with context: a promotion, a reorganisation, a difficult project. The movement between readings is one of the most useful things the system shows.

The Mirror: two halves meeting at the centre line, one the reflection of the other.

The Mirror: what a person can see of themselves

Self-awareness is rarer than most of us assume. Tasha Eurich and her team studied nearly 5,000 people across ten separate investigations and found that although most people believe they are self-aware, only 10 to 15 per cent meet the criteria. Eurich separates two kinds: internal self-awareness, knowing what you value and how you react, and external self-awareness, knowing how other people see you. Experience and seniority made neither kind more likely. Power, if anything, made the second kind less likely, because fewer people tell a senior person the truth.

A mirror addresses both kinds at once, which is why the reading has to be plain. Sariio describes a preference in everyday words, then says what that preference tends to earn a person and what the preference tends to cost. Someone who prefers to commit early gets decisions made on time, and sometimes makes a call that another day's thought might have changed. Someone who prefers to think first gets fewer reversals and sometimes waits longer than the moment allowed. Both are preferences. Neither is a grade.

The reason the survey holds up as a mirror is the reason the survey holds up everywhere else: the questions are about behaviour and the language that drives behaviour, not about how a person feels this week. That tradition of reading a person through their patterns of language and motivation is older than Sariio; Sariio's contribution is to measure the patterns as a preference on a scale, without sorting anyone into a type. The distinction between a preference and a type is set out in Trait versus type, and the beliefs beneath the instrument in Ten things we believe about ourselves and work.

5,000 People studied across ten investigations into what self-awareness is and who possesses the quality Eurich, HBR, 2018
10-15% Met the criteria for self-awareness, although most believed they did. Seniority made no difference Eurich, HBR, 2018
2 Kinds of self-awareness: internal (what you value and how you react) and external (how others see you). A mirror addresses both Eurich, HBR, 2018

Why does self-awareness matter for culture?

Because culture is built from individual habits repeated until they become the norm. Eurich's research found that only 10 to 15 per cent of people are self-aware in the sense of knowing how they come across. When more people can see the preferences they hold, the team can talk about differences plainly, and the organisation can see the pattern those preferences form.

Most people believe they are self-aware. In Eurich's research, one in seven were.

Eurich, Harvard Business Review, 2018
The Map: a trail of steps arriving at a destination.

The Map: what two people, or a team, can see together

A map is for the distance between people. Put two readings side by side and the differences that felt like awkwardness become explicable. The colleague who seems to slow everything down may prefer to think before deciding. The manager who decides in a heartbeat may prefer to act and adjust. Once both can see the other's preference, the conversation moves from "why are you like that?" to "how do we make this work?"

The evidence that fit between a person and their surroundings matters is unusually strong. Kristof-Brown, Zimmerman and Johnson pooled 172 studies in a 2005 meta-analysis and found that fit with the job, the group and the organisation each predicted satisfaction, commitment and the intention to stay. Fit with the group, the people you work beside every day, carried weight independently of the rest. That is the map's territory: the daily fit between two people, or among a team, measured as preference rather than guessed at from personality.

Two things follow from reading people as preferences rather than types. The first is that a difference between two people is information, never a problem to fix. The map says what each person does well and how the two work together; the decision about what to do with that belongs to them. The second is consent. A comparison appears only when both people agree to see each other's maps, and either can withdraw that agreement at any time. Whose preference to play to when two people differ is the subject of Play to my preference or yours?, and the answer from fifty years of reciprocity research is more generous than most people expect.

At team scale the same map shows where a team's preferences cluster, where the team divides, and the gaps that explain friction nobody had named. Google's Project Aristotle found that how a team works together mattered more than who was on the team, and psychological safety sat at the top of the list. Safety is easier to build when people can talk about their differences in neutral language, which is what a shared map supplies; the findings and their common misreadings are in Psychological safety: what Google actually found.

The Compass: the heads inside point the way, the last one sits outside the device, pointing north.

The Compass: what the organisation can see of itself

Now widen the lens to everyone. The claim that a personal survey can describe a culture rests on what culture is, so take the definition seriously.

"The way we do things around here" was coined by Marvin Bower of McKinsey in 1966 and became the working definition of organisational culture when Deal and Kennedy built their 1982 book around the phrase. Edgar Schein, the most cited scholar in the field, layered the idea: artefacts you can see, values people espouse, and beneath both the basic assumptions that guide behaviour largely outside awareness. Geert Hofstede called culture "the collective programming of the mind". Every one of those definitions locates culture in what people habitually do and assume, rather than in what the organisation says about itself. The longer treatment is in Organisational culture is language and behaviour.

The bridge from person to culture is Benjamin Schneider's 1987 paper in Personnel Psychology, one of the most cited in organisational psychology. The title is the argument: "The people make the place." Schneider's attraction-selection-attrition model says that people are attracted to organisations that suit them, organisations select people who suit them, and people who do not fit tend to leave. Over time the people who remain grow more alike, and their shared ways of working become the culture. Structure, process and climate are outcomes of who is there, not causes of who is there.

Can a survey of individuals really describe an organisation's culture?

The research says yes, provided the survey measures behaviour rather than opinion. Culture is defined as the way things are done in a place (Bower, 1966; Deal and Kennedy, 1982), and Schneider's attraction-selection-attrition work shows that the people who are drawn to, chosen by and remain in an organisation shape that place over time. Measure how those people prefer to work, aggregate the readings, and the shape of the culture appears.

1 in 5 People engaged at work worldwide, the lowest reading since 2020. The gap between the culture leaders describe and the one people experience. Gallup, State of the Global Workplace, 2026
200+ Companies in Kotter and Heskett's eleven-year study tying culture to economic performance. Cultures did well when they fitted the strategy and adapted. Kotter and Heskett, 1992
172 Studies pooled by Kristof-Brown and colleagues. Fit with the job, the group and the organisation each predicted satisfaction, commitment and staying. Kristof-Brown et al., Personnel Psychology, 2005

Put Schneider beside Bower and the compass follows. If the people make the place, and the place is the way those people do things, then measuring how the people prefer to do things, and aggregating the readings, describes the place. That is what a cultureMAP is: every person who has taken the survey appears as one anonymous point, the points assemble into fields, and the fields show where the organisation's centre of gravity sits, how widely people spread around the centre, and where the organisation divides into camps. No new questionnaire is issued. Nobody is asked to rate the culture. The culture is read from the same preferences that power each person's own mirror. How the fields are built, and why there are six of them, is explained in Why culture can be mapped and The six dimensions.

John Kotter and James Heskett followed more than two hundred companies over eleven years and found that culture had a measurable effect on economic performance, with one qualification that matters here: the cultures that did well were the ones that fitted the strategy and adapted as conditions changed. Strong was not the same as good. The same finding applies to a compass reading. A cultureMAP has no good or bad regions, only accurate ones. A concentration of people who prefer to act first suits a business that lives on speed and hinders one that lives on precision. The map shows the concentration; a person decides what the concentration means for the strategy. Kotter's later work on how culture changes is reviewed in Is Kotter's 8-step model still relevant?.

Structure, process and climate are outcomes of who is there, not causes of who is there. The people make the place.

Schneider, Personnel Psychology, 1987

The Compass in motion: Kotter's eight accelerators

A compass earns a place on the desk when the organisation wants to move. Here the reading meets John Kotter, whose 1995 analysis of why transformation efforts fail remains the most used change model in business. Kotter watched change die at eight recognisable points and turned each one into a step. In 2014 he revised the sequence into eight concurrent accelerators, running at once rather than in order, and that revision is the version Sariio uses. The history and the evidence for each accelerator are reviewed in Is Kotter's 8-step model still relevant?

70% Kotter's estimate of change efforts that fail, most often at the first accelerator, urgency Kotter, HBR, 1995
75% Of management who need to see the status quo as unacceptable before a change is safe to launch Kotter, HBR, 1995
8 in 10 Re-engineering efforts where the gains had dissolved within two years of victory being declared Kotter, HBR, 1995

Kotter's accelerators describe what leaders do. The cultureMAP describes who the people are. Put the two together and a question that used to be answered by instinct gets an answer from evidence: how ready is this culture for each of the eight?

Sariio's change readiness reading works like this. Each accelerator draws on two or three of the six cultureMAP dimensions, weighted by how directly each dimension bears on that accelerator. Urgency reads Direction and Action, because an organisation whose people prefer to keep options open and act only when sure warms to urgency more slowly than one that prefers to commit and move. A coalition reads Commitment and Impact. Communication reads Commitment and Direction. Removing obstacles and generating short-term wins both read Process and Action, in different proportions. Building on the change and making the change stick each read three dimensions, because consolidation asks the most of a culture.

8 Accelerators, each read against two or three of the six cultureMAP dimensions
5 Temperatures on the scale: Frozen, Thawing, Warming, Heating, Red Hot
0 New questions asked. The reading is computed from the same preference data as the cultureMAP

Each accelerator receives a temperature. Frozen means the culture's preferences pull against that accelerator; Red Hot means the culture is already inclined that way; Thawing, Warming and Heating sit between. The temperature is computed from the numbers in the culture snapshot by a fixed rule, and Sariio AI writes the reading around the number rather than choosing the number. That distinction matters for trust. Two readings of the same organisation, months apart, will show a temperature moving only when the distribution of preferences has moved. The reading is a thermometer, and a thermometer that changes reading on a whim is no use to anyone.

Urgency is readiness work, done before anything visible changes. A compass tells you how much of that work the culture has already done.

What a leadership team does with eight temperatures is the same thing a navigator does with a bearing: plan the route for the terrain in front of them. A culture that reads Frozen on urgency and Heating on short-term wins is telling the leadership team where the first month of effort belongs, and which accelerator will supply the early evidence of progress. The reading prescribes nothing. Which changes to attempt, and in what order, remains a decision for people. How readiness is assessed, and why readiness differs from the absence of resistance, is the subject of What is change readiness?, the companion piece to this one.

Why one source stands scrutiny

The objection is a fair one. A survey built for individuals is being asked to describe a pair, a team and an organisation. Why trust the same instrument at every level?

Three answers, each with a test attached.

The first is that the unit of measurement never changes. At every level the reading is a preference on a scale, drawn from the same statements answered by the same person about how that person behaves. A team reading is a set of those scales laid side by side. A culture reading is the same set aggregated and anonymised. No level introduces a new kind of data, so no level introduces a new kind of error. Engagement surveys, by contrast, ask each person to rate the organisation, which measures sentiment about the place rather than behaviour inside the place; two organisations can post the same score for opposite reasons, and the score alone cannot tell you which one you are running.

No level introduces a new kind of data, so no level introduces a new kind of error.

The second is that the aggregate is more stable than the individual. One person's reading moves with context, which is why the survey is re-taken. An organisation's distribution moves too, but slowly, because Schneider's attraction, selection and attrition operate over years. The compass is a reading of many mirrors, and the noise in any one of them averages out across hundreds. Anonymity does the rest: no individual is identifiable in a cultureMAP, so the reading carries no threat and invites candour.

The third is that the levels check each other. A mirror that says one thing and a team map that says another is a finding, not a flaw: the person may behave differently in this team than they prefer, which is exactly the kind of friction worth a conversation. A culture map that sits far from the strategy the leadership team describes is the same kind of finding at scale. Because all three are read from one source, the discrepancies are visible instead of being hidden between three different instruments that cannot be compared.

What the compass will not do

Does a Sariio Compass reading judge an organisation's culture?

No. A cultureMAP describes the distribution of preferences across an organisation and where the organisation divides into camps. There are no good or bad regions. Kotter and Heskett found that cultures perform when they fit the strategy and adapt as conditions change, so the same shape can suit one organisation and hinder another. A person reads the map and decides.

A compass gives a bearing. The compass does not choose the destination, and a cultureMAP does not either. The reading will not score the culture, rank the organisation against others, or prescribe a change. Preferences are orientations, and an orientation is only an asset or a liability in relation to where the organisation wants to go. That decision belongs to people, with the evidence beside them. Sariio takes no biometric input and analyses no voice, face or video; the whole reading rests on what each person said about how they prefer to work, in about 12 minutes, and chose to share.

Three instruments, one reading

Back to the opening question. Where does a culture come from? From people, and from what those people prefer, repeated until the preference becomes the norm. A mirror lets one person see their part in that. A map lets two people, or a team, see how their parts fit. A compass lets the organisation see the whole, and the direction the whole already travels. One survey, three distances, and the same evidence at each. Understanding costs nothing at the first level, and at the third level the understanding is what a leadership team needs before deciding where to steer.

Sources