"Scan: how to read a team before you design anything"

Before we design anything, we read the team. We call it a scan, and it is the step most team building skips entirely. The booking comes first, the team's actual state is guessed at, and the guess is usually the manager's guess, which is one person's view of a system they sit inside. Then an activity is chosen to fit the guess, and the whole day is built on a foundation nobody checked.

A scan replaces the guess with a reading. It is the first move in the way we work, and it is the one everything else depends on — the design, the delivery, and the proof all rest on getting this right. It is worth being concrete about what a scan is, what it reads, why it so reliably overturns the expected answer, and why it is the honest step that makes everything after it meaningful. This article is the companion to diagnostic-first design: that one argues for reading before designing; this one is about what the reading actually is.

Reading, not asking

The obvious way to find out what a team needs is to ask the person who runs it. That is a useful input and a partial one. A manager sees the team from a single vantage point, and it is the vantage point with the most power in the room, which is exactly the position from which certain problems are invisible. People behave differently in front of the person who decides their pay and promotion. The manager sees the version of the team that is performing for the manager.

There is nothing dishonest in this; it is simply how power distorts information. The things a team is most careful about — the disagreement it does not voice, the decision it quietly reopens, the person who has stopped contributing — are precisely the things it is most careful about in front of its leader. So the leader's account, however sincere, is systematically missing the parts that matter most. A scan exists because the most important information about a team is usually the information its manager is least positioned to see.

The views a scan recovers

Reading the team as a whole recovers the vantage points the manager's view leaves out, and those are often the decisive ones. The person who stopped speaking up in meetings knows something the manager does not — they know exactly when it stopped feeling safe to speak, and why. The two functions that quietly route around each other know something the manager does not — they know where the collaboration actually breaks, in the handoffs no status report captures. The new joiner who still feels like a guest four months in knows something the manager does not — they know the difference between being onboarded and being included, and they are living in the gap.

None of these people are likely to volunteer what they know, and certainly not in a room, and certainly not to the person with the most power in it. A scan is built to read the team across all of these angles, so the picture is not the view from the top but the view from inside. The whole point is to weight the quiet signals appropriately — because the quiet signals are usually where the real problem is hiding, and the loud ones are usually where it is not.

What we read for

We read a team across eight dimensions: trust, communication, alignment, collaboration, decision-making, energy, belonging, and leadership. Those are described in full in the eight dimensions of a healthy team, and they are the frame that turns a vague sense of "something is off" into a specific, nameable finding. We are not going to describe how the reading is taken, because that is the craft and it stays in the work. What matters here is the shape of what comes back.

A team is rarely weak everywhere or strong everywhere. It is usually strong on several dimensions and quietly hollow on one or two, and the hollow ones are doing most of the damage while the strong ones hide it. A team can communicate constantly and still not trust each other. It can be full of energy and completely unaligned. It can be warm and close and unable to make a decision that stays made. The scan is what separates the dimension that looks like the problem from the dimension that is the problem — and those are different far more often than anyone expects going in.

Symptom and source

The reason this separation matters is that dimensions disguise themselves as one another. What presents as a communication problem is often a trust problem underneath — people are not failing to talk, they are choosing not to say the risky thing, which is a safety issue, not a skills issue. What presents as a low-energy team is sometimes a belonging problem — the people who look disengaged are the people who have quietly concluded they are not really part of things. What presents as a team that cannot decide is often a team that does not trust itself enough to commit, because a decision is an exposure and exposure feels unsafe.

A scan reads for the source, not the symptom. This is the difference between a reading and a survey. A survey collects the symptoms — it asks people how they feel and reports the answers. A scan interprets them, looking underneath the reported feeling to the dimension actually generating it. Treat the symptom and you get a pleasant day that changes nothing, because the source is still there, still generating the same symptom. Find the source and a single well-aimed experience can shift several symptoms at once, because they shared a root.

The answer is almost never what was expected

Here is the pattern we see most. A leader comes in certain of the diagnosis. The team needs more fun. The team needs to bond. The team is low on energy and needs a lift. Sometimes that is right, and the scan confirms it. Often the scan finds something underneath it that changes the whole design.

The team that "needs to bond" turns out to trust each other fine and simply cannot make a decision without it being relitigated a week later — that is not a bonding problem, it is a decision-making and commitment problem, and a bonding day would leave it exactly where it started. The team that "needs energy" turns out to be perfectly capable and quietly excluded from anything that matters — that is not an energy problem, it is a belonging one, and a high-energy day would spend energy the team did not have to spare on a problem it did not have. The team that "needs communication training" turns out to communicate fine about everything except the one thing that matters, which it avoids because avoiding it is safer — that is trust, and no amount of communication technique addresses it.

If you design for the expected problem, you build a good day for a problem the team did not have. If you design for the real one, you have a chance of changing something. That gap is the entire reason the scan exists, and it is why we will not skip it even when a client arrives certain they already know the answer. The certainty is exactly the thing worth checking.

Not captured by the loudest voice

One discipline inside a scan deserves naming, because it is where amateur readings go wrong. A team has loud voices and quiet ones, and the loud voices are not necessarily describing the real problem — they are describing the problem that is easiest to say out loud. The person who complains most vocally about process may be comfortable complaining about process precisely because it is safe, while the actual issue, the one about trust or leadership, goes unsaid by everyone because it is not safe. A reading that simply amplifies the loudest complaint will confidently target the wrong dimension.

The same is true in reverse: a confident, cohesive-seeming team can have a silent problem that no one will raise because raising it would break the very cohesion everyone is protecting. Reading a team well means weighting what is said against who is not saying anything, and against what the pattern of the whole group implies rather than what any one member asserts. This is why a scan is interpretation, not tallying. The goal is the truest picture of the system, which is rarely the average of the opinions in it.

Two scales: the team and the organisation

A scan reads at two levels, because a team does not exist in isolation. There is the health of the team inside itself — how these particular people work together. And there is the health of the organisation around it — whether the team is aligned to a goal the company actually made clear, whether it trusts the teams it depends on, whether belonging to this squad is the same as belonging to the company. These can diverge sharply. A team can be strong internally and starved of direction from above, so its excellent internal alignment is pointed at the wrong target. A team can be individually healthy and yet isolated, trusting itself and no one across the seams.

Reading both is what stops a scan from prescribing a team-level fix for an organisation-level problem. If a team is struggling because the company above it has never made the priorities clear, a team offsite will not fix it, and diagnosing that early saves everyone a wasted day. This distinction becomes acute in large and distributed structures, which is why we treat it at length in GCCs: one company, two continents. The scan is where the two scales get told apart.

Why this is the honest step

A scan also does something uncomfortable and useful: it makes the goal explicit before anyone commits to an activity. Once you have read the team, you know what you are trying to move, which means you can be held to whether you moved it. That is the opposite of the usual arrangement, where the goal stays vague enough that any pleasant day counts as success. Vagueness protects the vendor and the buyer alike — if you never said what you were trying to change, no one can say you failed to change it.

We would rather start with a clear reading and a clear target, because it is the only way the measurement afterwards means anything. You cannot prove you fixed a problem you never named. The scan is what names it, which is what makes the follow-up readings — the ones we take at Day 14, 30 and 60 — interpretable. Without a baseline reading there is nothing to compare the later ones against, and "did it work" collapses back into "did people enjoy it." The scan is the baseline. It is the reason the whole method can make a claim at all, rather than just a nice afternoon and a hopeful feeling. This is the through-line of why measurement changes the conversation: it starts here, with a reading that sets the target.

What a scan is not

Because "reading the team" can sound heavier than it is, it helps to say what a scan is not. It is not surveillance — it is not about monitoring individuals or building files on people, and a scan that made a team feel watched would corrupt the very thing it was trying to read. It is not a personality test; we are reading how the group works together, not sorting people into types. It is not a long consulting engagement — the reading is proportionate to the work, sized to inform a design, not to produce a report that sits in a drawer. And it is not a grade — the point is never to score the team or rank it against others, only to find the specific dimension that is costing this specific team right now.

Kept in proportion and aimed at the system rather than at individuals, a scan is simply the practice of looking before acting. It feels unusual only because so much of the field acts without looking.

The team that looks fine

The hardest reading is not the visibly struggling team. It is the team that looks fine. A group can present as healthy — meetings run smoothly, no one is complaining, delivery is on time — and carry a fault line that no surface indicator reveals. Everything is agreeable precisely because the team has quietly agreed not to touch the thing that would cause friction. Smoothness can be a sign of health or a sign of avoidance, and from the outside the two look identical. A satisfaction survey would give this team high marks and miss the fault entirely, because the fault is exactly what no one is willing to put in a survey.

A scan is built to catch this, because it reads for the source rather than the surface. The absence of visible conflict is itself a datum: is this a team safe enough to disagree openly, or a team that has stopped disagreeing because disagreement stopped being safe? Those produce the same calm surface and could not be more different underneath. Reading well means being suspicious of a picture that is too clean, and knowing that on some teams the most important finding is the thing everyone is being careful not to say. The scan's job is to find that thing before a well-meant "celebrate the healthy team" day papers over a crack that is quietly widening.

Reading a team you cannot sit in a room with

Distance makes a scan harder and more necessary at once. A distributed or remote team gives off fewer of the ambient signals a co-located one does — you cannot read the pause before someone speaks, the glance across the room, the person who goes quiet. The information is thinner and easier to miss, which is exactly why guessing is more dangerous for these teams and a deliberate reading matters more. The manager's view, already partial, is partial in new ways when half the team is on another continent and only ever seen through a screen.

Reading across distance means compensating for what the medium hides — weighting the signals that do travel, noticing who never turns on a camera or never speaks first, attending to the seams between locations where belonging and trust most often thin out. A distributed team can look perfectly functional in its structured meetings and be quietly fragmenting in all the unstructured space a co-located team takes for granted. The scan is where that fragmentation gets seen before it becomes attrition. It is also why designing for these teams without a reading is close to guaranteed to miss: the very things that are wrong are the things distance hides, and only a reading built to look for them will find them.

A worked example

A leader asks for a bonding day. The team has grown fast, people barely know each other, and the obvious read is that they need time together to gel. Reasonable. But the scan finds something else: the team knows each other perfectly well and gets on fine socially. What is thin is not connection but clarity — no one is quite sure who owns what, so work collides and stalls, and the friction people feel gets misread as "we do not know each other well enough." It is not a belonging gap. It is an alignment-and-roles gap wearing a belonging costume.

A bonding day would have been pleasant and useless, sending an already-friendly team off to become slightly friendlier while the real problem — unclear ownership — carried on generating collisions the following week. The scan redirects the design toward getting the team genuinely aligned on who does what and how decisions get made, which is the thing actually costing them. Same team, same budget, and the difference between a warm day that changed nothing and a day that removed the friction. The only thing that produced that difference was reading the team before believing the brief.

Who does the reading

A scan is only as good as the person interpreting it, which is worth saying plainly. Reading a team is a skill, not a form — it is the ability to hear what a group is not saying, to weight a quiet signal against a loud one, and to resist the confident wrong answer the brief arrived with. This is why the reading and the facilitation are not separate crafts in our work; the person who reads the team and the person who designs and runs the day share the same judgement, described in what we look for in a facilitator. A reading handed to someone who did not take it, to be run mechanically, loses most of what made it valuable — the texture, the things that did not fit the categories, the sense of what this particular team actually needs. The scan is not a document produced and filed. It is a live understanding carried into the design and the room.

Everything downstream rests on this

Everything after the scan — the design, the delivery, the proof — rests on this first step. The design can only be as good as the reading it follows; aim at the wrong dimension and even a beautifully run day misses. The delivery can only reinforce what the design targeted. And the proof only exists because the scan named a target to measure against. Get the scan right and the rest of the method has something true to work from. Get it wrong, or skip it, and everything downstream is built on a guess, however confidently the guess was held.

Read the team. Find the dimension that is actually costing it. Design for that, and nothing else. It is a simple discipline, and it is the one that separates a team experience that changes something from one that merely fills a day. The scan is not a preliminary to the work. It is the beginning of it.