The Communication Topography
Launch week, and someone who should have known does not. Support finds out about the pricing change from a customer. A partner team ships against an assumption you retired a month ago. Two people on your own team are still arguing a decision that was, as far as you were concerned, settled and announced. Nobody was negligent. The information existed. It was in the thread, the doc, the update you sent, the standup you all sat through. And still the gap was real.
The reflex is to treat this as a discipline problem. People should read more carefully, write clearer updates, keep the wiki current. So you add a channel, a weekly digest, a template. Sometimes that helps. Often it does not, because information gaps are not scattered randomly across a team. They follow the shape of the land. Some things travel easily because the terrain runs downhill toward them. Other things never arrive because the terrain routes around them, and no amount of pouring more water into the same channel changes where the ground slopes.
This quarter I am writing about seeing practice: noticing how work actually runs before adding structure to it. This piece is about learning to see the terrain your communication moves across, so the gaps stop surprising you.
The mistake: mistaking the org chart for the map
Ask most teams how information flows and they will describe the official plumbing: these are our channels, this is the standup, decisions go in this doc, the org chart says who reports to whom. That is a list of boxes and lines. It is not a map. A map tells you position and direction: what sits where, and which way things actually move between them.
The real terrain is different from the chart. It is who genuinely talks to whom, which meeting carries weight and which is theatre, the one person everyone quietly routes hard questions through, the two teams that never speak because their leads once clashed. Information gaps live exactly where the assumed map and the real terrain disagree. You sent the update along the line you drew on the chart. The ground sloped a different way, and the update pooled somewhere short of the people who needed it.
Information runs downhill; knowledge needs somewhere to pool
Before you can read the terrain, separate two things that get bundled under the word "communication".
There is information: portable, legible, easy to move. A ticket, an export, a decision written in a doc, a number on a dashboard. Information behaves like water. You can pour it into a channel and it will flow to wherever the channel leads.
Then there is knowledge: the shared sense of what the information means and why it matters now. Why we killed that feature. What the pricing change is really protecting. Which constraint moved and who felt it. Knowledge is not portable in the same way. It forms only where people share enough context to make sense of the information together, and it lives in the people who were there when it formed.
That distinction explains a gap that looks impossible on paper. A high-traffic channel can move an enormous amount of water and still leave the team parched of knowledge. Everyone technically received the message. Nobody built a shared understanding of it. You get coverage without comprehension: people can tell you that pricing changed and go blank on why. So there are really two kinds of gap, and they need different fixes. Either the information never reached someone, which is a routing problem, or it reached them and no knowledge formed, which is a problem of there being no place for it to pool.
Reading the terrain
A communication topography is a rough map of that landscape for one product team and the people around it. You are not drawing an accurate survey. You are sketching the features that decide where things flow and where they get stuck. Five features do most of the work.
SOURCE (high ground)
the decision, the customer call, the incident
|
v CHANNEL (watercourse: standup, the busy thread, the one-pager)
|
[ BASIN ] <- knowledge forms here: a real conversation, a shaped bet
| discussed live, a pairing session
|
+--> CATCHMENT: core team, in the same meetings (reliably gets wet)
|
+--> SINK: the archived doc, the 40-reply thread (pools, never re-read)
|
. . . RAIN SHADOW: support, a partner team, the new starter,
the person on leave (the terrain routes around them)- Sources are the high ground where context originates: a leadership decision, a customer call, an incident, a research finding. Everything downstream starts here.
- Channels are the watercourses, the paths information actually takes. The standup, the one busy thread, the weekly note, the corridor conversation. Channels move information well. They do not, on their own, build knowledge.
- Basins are the few places where knowledge actually pools: a recurring conversation where people make sense of things together, a shaped one-pager talked through live, a pairing session, a discovery review. Basins are where water sits long enough to change the ground. Most teams have far fewer of these than they think.
- Sinks look like basins but are not. Information flows in and pools, and no one ever returns to it. The archived decision doc. The thread with forty replies and a tick emoji. It felt like the message landed because it was written down. Nothing formed there.
- Rain shadows are the roles the terrain routes around by default. Support, a partner team, a downstream squad, the new starter, the person on leave. Not because anyone excluded them, but because they sit behind a ridge that the water does not naturally cross. Their gaps are the most predictable of all, which is exactly why they keep catching teams out.
The point of drawing this is that the gaps become predictable. If you can see that support sits in a rain shadow and that your only basin is a meeting three of the eight relevant people never attend, you already know where next month's "how did nobody know?" is going to come from.
Drawing one of your own
Do not map everything. That produces a mural nobody uses. Map around a single real surprise or a single decision that absolutely has to land.
Take the last time information failed to arrive. Trace it backwards along the terrain:
- Source. Where did this actually originate? Name the moment and the person, not the doc it eventually landed in.
- Channel. What path did it take from there? Be honest about the real one, not the one the process says it should have taken.
- Who got wet. Which people or roles reliably received it, and did they only receive the information, or did they build knowledge from it?
- Who stayed dry. Who needed it and did not get it, and which terrain feature explains that: a rain shadow, a sink that looked like a basin, a channel that stopped short?
Then ask the diagnostic question. Is the fix a channel or a basin? A channel fix moves the water: reroute the update, add the missing person to the source conversation, change who writes the note. A basin fix builds a place where knowledge can form: a short recurring conversation between the two functions that never sync, a live walk-through of the decision rather than a written announcement. Most teams reach for channel fixes because they are cheap and visible. Many real gaps are basin problems, and adding another channel just digs a parallel ditch that carries the same water past the same dry ground.
When not to reach for it
This map is a hypothesis, not a survey, so treat it like one. The fastest way to check it is not to admire your own diagram. It is to go to two people you have marked as "dry" and ask what they actually know. If they surprise you, redraw.
Be wary, too, of confusing more channels with better terrain. A team that responds to every gap by adding a channel ends up with a floodplain: water everywhere, nowhere for knowledge to settle. And do not map the whole organisation for its own sake. The topography earns its place when it is pointed at a real decision that has to reach real people by a real date, not when it becomes another artefact to keep current.
A line you can reuse
When the room reaches for another channel to fix a gap, try:
"Where does this actually need to land, and does the way we work route it there? Or are we pouring it into a channel and hoping the ground slopes the right way?"
It moves the conversation from broadcasting harder to seeing where the terrain already carries things, and where it never will.
Seeing practice in the landscape
I keep coming back, this quarter, to the idea that product work breaks down before the artefacts look wrong. Information gaps are a clean example. The doc existed, the channel was busy, the update went out, and the terrain quietly carried it past the people who needed it most.
Seeing practice here means looking past whether the message was sent and asking where it could travel, where it could pool, and where the ground was always going to route it away. Map the terrain once, around a gap that actually hurt, and you stop being surprised by the same dry valley every launch.
The question is not whether you communicated it. It is whether the shape of how your team works was ever going to carry it to the people who needed it.
