The Reteaming Anti-Patterns
The three ways leaders reliably wreck a reorg: percentages, poof, and spreading stars.
- Difficulty
- Easy
- Time to result
- ~days to results
- Steps
- 4
- Confidence
- 93%
Helfand catalogues the recurring failure modes of team change. The three she names most sharply are the percentage anti-pattern (allocating humans as fractional resources across projects), the 'poof' anti-pattern (people appearing or vanishing from teams with no communication), and spreading high performers (breaking up a great team to seed its magic elsewhere). Each is a specific, avoidable mistake — and each is intuitive enough that leaders keep repeating it.
Origin
Drawn from Helfand's own waterfall-era experience with resource allocation, and from AppFolio CTO John Walker, who ran the spread-the-high-performers experiment himself and described the failure to her as a visceral learning.
Core principles
- 01Humans are not fractional resources; context switching has a real cost.
- 02Team performance is a property of the team, not a portable property of its members.
- 03Every reteaming pattern has 'balconies and basements' — an upside and a way to ruin it.
- 04Communication is not the garnish on a team change; its absence is itself the anti-pattern.
How to run it
- 1
Avoid the percentage anti-pattern
Do not allocate people across projects in fractions — 10% here, 5% there, 20% on that. The percentages don't add up in reality, and the context switching is brutal on the brain. Staff whole people to whole efforts.
Watch out This is a waterfall-era resource-allocation habit that keeps resurfacing whenever component-based teams are asked to serve multiple projects.
- 2
Avoid the 'poof' anti-pattern
Do not have people suddenly appear on a team, or suddenly disappear from one, with no communication whatsoever. Announce arrivals and departures, visualize hiring, and explain the reasoning.
Pro tip Visualizing open roles and upcoming hires kills this anti-pattern almost for free.
- 3
Do not spread your high performers
Resist the intuition that scattering members of a magic team across other teams will propagate the magic. It does not transfer — you just destroy the original team. The chemistry lives in the combination, not the individuals.
Pro tip If you want the magic elsewhere, study what conditions produced it — cadence, autonomy, shared purpose — and reproduce the conditions, not the roster.
Watch out This one feels like obviously good leadership. It is the trap.
- 4
Do not haphazardly rotate people
Switching should be tied to learning and fulfillment, with the person having a say. Rotating someone every two weeks with no agency in the choice is the basement version of the switching pattern.
In the wild
AppFolio had one team delivering at a cadence they loved, building great things, with a palpable energy. CTO John Walker's thought was: spread those people across the other teams and we'll have that energy everywhere. He ran the experiment.
→ It didn't work — the magic did not propagate, and by his own account he essentially destroyed the original team by splitting it up. Helfand calls it the story that inspired the anti-pattern.
Damon Valenzona described a high-performing team to Helfand as being like a band: riffing off each other, creating music together.
→ Frames why spreading members fails — you cannot get the music by distributing the musicians into other bands.
Common mistakes
Believing a great team is a bag of great individuals
High performance is emergent chemistry between specific people working in specific conditions. Redistribute the individuals and the emergent property does not come with them — it just stops existing.
Modeling people as capacity percentages
The allocation math looks clean on a spreadsheet and fails in practice: the percentages don't add up, and the constant context switching means neither effort gets a real person.
Silence around arrivals and departures
When people appear or vanish with no communication, the team invents a story — usually a worse one than the truth — and trust in every subsequent change erodes.
Is it for you?
Best for
Engineering and product leaders about to redistribute people, especially anyone tempted to break up their best team to 'spread the culture'.
Not ideal for
Nothing — but note these are diagnostics for what not to do, not a positive process on their own; pair with the five patterns.
From the transcript
“you'll work 10% on this one 5% on this one 20% on this one”
“so I call that the percentage anti pattern”
“spreading High performers this idea of we're going to spread the high performers across other teams”
“John felt like he essentially destroyed that when he split up split up that team”
From the episode
The art and wisdom of changing teams
Heidi Helfand (author of Dynamic Reteaming)