Team Resilience: The Best Score in the Survey Had a Single Point of Failure
What is described here happened inside one of the biggest units of a worldwide research and advisory group, as part of an organisational change programme. Nine product and technology teams made up the unit, which generated hundreds of millions of dollars in revenue every year, with a goal to double that. We built an evidence base first, scoring the organisation on twenty-five dimensions and running a qualitative pass to catch whatever those scores left out.
The analysis started with the map.
Start with the map
Correlation, not causation: every edge is a measured statistical association (p<0.05) from this engagement, not an asserted cause. Edges below |r|=0.35 are omitted for legibility; the rest are drawn thicker and more opaque the stronger they are. The map stays focused on Risk management: hover any dimension to preview its own connections against it, and use the strength slider to keep only its strongest links.
The map opens centred on risk and obstacle removal: how blockers get raised and cleared. By the words, this is the healthiest dimension in the whole study, and the map shows why it matters that it stays healthy. Its strongest neighbour is smooth flow of work: unblocking capacity is flow capacity. Whatever this capability runs on, everything downstream of it runs on the same thing.
What the numbers said
Healthy on the scores: the median comes in at 8 out of 10, and only seven respondents out of sixty-one rated it 4 or below. Nothing in the scores asks for a page. The reason this page exists is an asterisk the respondents wrote into their own comments.
What the words said
The two dark blobs not lining up is the chart.
Only five coded mentions land here, and by sentiment they are the healthiest in the network. The capability is real, and the leadership interviews describe a genuine readiness posture:
“We always have contractors on stand by, and remote support is also accounted for. We have a matrix that helps fulfill all the roles.”
Then the survey adds its own asterisk, in two consecutive units:
“I think oftentimes the people who resolve blockers tend to be spread thin.”
“So their ability to resolve said blocker is dependent on them having the time to look into it.”
The best capability in the study runs on a handful of overloaded individuals. Heroics with a matrix is still heroics: it works until the day the hero is busy.
Where it lived
Nine teams, anonymised, best to worst. Thin rows render wider and flatter; less data looks uncertain, not falsely precise.
Two teams sit at the bottom, down to 4.5, while seven of the nine carry medians of 7 or better, reaching 9. Even the study’s best capability has an address where it doesn’t reach, a plausible fit with the asterisk: person-bound capacity runs out somewhere, and it runs out at the same teams every time.
What unblocking turned out to be entangled with
The measured neighbourhood:
- Smooth flow of work (r = .71), the strongest pull
- One team (r = .66)
- Release processes (r = .65)
Flow at .71 is the highest correlation in this dimension’s neighbourhood, and it reads both ways: clearing blockers keeps work moving, and a stall anywhere becomes a blocker for someone. Which is exactly why the respondents’ own wishes point past the heroics, toward a system:
“We could implement more strategies and practices to improve our predictive capacity and detect problems before they occur.”
“I just wish we do take risk factor into consideration while evolving, that’s all.”
A capability that lives in individuals is a score that can change the day they’re unavailable. A coherent account of why the healthiest dimension still earned an asterisk. Correlations and comments are the evidence; cause is not proved.
What happened next
The systematization was already starting inside the same corpus: the transition from person-bound firefighting to instrumented practice:
“As the engineering team transitions to SRE, one to two team members will focus on monitoring. This will facilitate knowledge sharing and skill development to fully adopt SRE practices.”
That is the direction the asterisk asks for: move the capability from people who happen to have time into practices that don’t care who’s busy. The check that travels: name the three people who cleared your team’s last three blockers. If it’s the same name three times, your best score has a single point of failure too.
A note on the data
A real engagement; an unnamed client, and it stays that way. Team names and identifying details are removed, and quotes are lightly edited for anonymity. The numbers and charts on this page compute directly against the underlying assessment data each time it renders, roughly 2,000 data points pulled from both survey streams, with nothing asserted. One organisation, correlations not causes.
Read Next
- ROAM: the four dispositions as four commitments
‘Owned’ means a name and a review date: the systematized version of a capability this client ran on whoever had time.
- Flow metrics: system-level diagnostics
‘Detect problems before they occur’ is a wish until it’s instrumented: this is what the instrumented version looks like.