Tree of InfluenceRunaway scale
The tree grows too large for anyone to read.
Description
The tree keeps growing, to hundreds of nodes and a tangle of arrows, until nobody can see its overall shape. Zoomed out, it is visual noise. Zoomed in, nobody can tell where they are. Participants lose the thread, and the tree becomes a maintenance burden instead of a thinking tool.
Runaway scale usually creeps in rather than arriving at once: every session adds a few nodes, removal lags behind, and nobody steps back to reorganize.
Why it matters
A tree that cannot be read cannot guide anything. If people cannot trace a path from a problem to an outcome, or find what matters most, the tree has failed at its job.
The sessions that built it may still have been valuable: the conversations happened and people learned. But the diagram stops carrying that understanding forward. Instead of fading because it did its job, the tree is abandoned because nobody can use it.
How to spot it
- The zoom test. Zoom out until the whole tree fits on the screen. If you cannot read any node's text, the tree is too big for a single view.
- The path test. Tracing one problem to an outcome means following more than 8 to 10 arrows.
- What people say. "I can't find anything." "This is overwhelming." "Where do we even start?"
- The size. The largest observed tree that stayed readable had about 90 nodes and 100 connections, and only because related nodes were clustered in frames. A tree well beyond that size with no clear clusters is probably past the point of being read.
How to correct it
Before the session, set the scope. Agree on what is in and out of the tree: "We're mapping deployment problems, not the whole engineering organization." A clear boundary is the cheapest prevention.
Cluster related nodes into labeled frames. Group nodes by domain, such as "Deployment" or "Team dynamics." Frames give readers landmarks without changing the tree's meaning. They are a reading aid and need no legend entry.
Diagram as text: The onboarding tree organized into labeled frames
Nodes:
- Root / Problem: "Each team onboards differently" (in "Knowledge")
- Root / Problem: "Documentation is out of date" (in "Knowledge")
- Root / Problem: "Setting up a laptop takes a week" (in "Setup")
- Root / Problem: "Senior people have no time to mentor" (in "People")
- Root / Problem: "New hires are afraid to ask questions" (in "People")
- Action / Intervention: "Automate laptop setup" (in "Setup")
- Action / Intervention: "Assign each new hire a buddy" (in "People")
- Desired State / Outcome: "New hires ship something in their first month"
- Desired State / Outcome: "New hires feel welcome and confident"
Arrows:
- "Each team onboards differently" influences "Documentation is out of date"
- "Setting up a laptop takes a week" influences "Automate laptop setup"
- "Senior people have no time to mentor" influences "New hires are afraid to ask questions"
- "Senior people have no time to mentor" influences "Assign each new hire a buddy"
- "Documentation is out of date" influences "New hires ship something in their first month"
- "Automate laptop setup" influences "New hires ship something in their first month"
- "New hires are afraid to ask questions" influences "New hires feel welcome and confident"
- "Assign each new hire a buddy" influences "New hires feel welcome and confident"
Remove what does not belong. During refinement, the group may agree to remove nodes that turned out to be tangential or outside the scope. Not every contribution has to survive in the current tree, as long as removal is a group decision and the earlier version is saved.
Split the tree into views. Let the full tree be the map, and work in focused views:
- A blocked outcomes view shows only the nodes with a path to one outcome.
- A deep-dive sub-tree moves the detail of one dense area into its own tree.
- After voting, a priority view shows only the most-voted nodes and their immediate connections.
Add a level of hierarchy. Use a few higher-level problem nodes as organizing points, with the detailed nodes clustered beneath them.
Plan another session. If one area of the tree keeps growing, give it its own session and its own tree.
Very large trees are still being explored. If you have kept a tree of several hundred nodes useful, share how.