"The team that does not learning is doomed to repeat the same mistakes... forever." (Note: literal translation of opening quote context)
Every technical team makes mistakes. In living systems that undergo constant changes, incidents are inevitable. The real difference between an amateur operation and elite engineering lies in what happens the morning after a mistake. Mediocre teams sweep the problem under the rug and hope nobody notices. Good teams hold a post-mortem meeting. Excellent teams turn every failure into institutional learning and automatic safeguards in the system.
The science behind evolving teams: Edmondson, Senge, and Dweck
For a long time, traditional managers believed that the best teams were those that registered almost no failures. In 1999, a landmark study by Professor Amy Edmondson from Harvard Business School, published in Administrative Science Quarterly, proved the exact opposite when investigating clinical and operational teams.
Edmondson discovered that top-performing teams reported more errors than weak teams, not because they made more mistakes, but because they possessed psychological safety: the shared belief that the environment is safe for interpersonal risk-taking, raising your hand when something goes wrong, and asking for help without suffering humiliation or retaliation. Where there is no psychological safety, failures continue happening every day, but they remain hidden until they cause an irreversible disaster.
This discovery connects two other milestones in management and behavior literature:
- In The Fifth Discipline (1990), Peter Senge from MIT demonstrated that so-called learning organizations are those that abandon the fragmented view of hunting down isolated culprits and adopt systems thinking, seeing how the process structure produces the results.
- In Mindset (2006), psychologist Carol Dweck from Stanford University showed that environments dominated by a "fixed mindset" treat any mistake as proof of permanent incompetence, leading people to lie to look smart; whereas growth mindset cultures treat technical stumbling as raw material to evolve engineering.
The organizational learning cycle in four steps
To transform these concepts into a practical technology routine, I structure team learning into four non-negotiable stages:
1. Failing without fear of speaking up
The first step is to ensure that whoever identified or committed a mistake immediately notifies the operation channel. If the professional is terrified of exposing a problem at the beginning, they will try to fix it alone in the dark, and what would take ten minutes to contain turns into a three-day crisis.
2. Analyzing without blaming (focus on the system)
A technical post-mortem is never a witch-hunt tribunal; it is an engineering investigation. The goal is not to ask "who pressed the wrong button?", but rather "why did our system allow a single manual command without validation to bring down the production environment?" Well-intentioned people fail when the process does not offer guardrails.
3. Documenting without forgetting
The learning discussed in the meeting cannot die in the memory of those who were on the video call. It needs to turn into a short, clear, and searchable record for anyone joining the team six months from now.
4. Applying without delay
It is no use collecting a beautiful wiki of incident reports if nothing changes the following Friday. Every real learning needs to generate concrete action in the system: a new automated test in the pipeline, a more accurate monitoring alert, a permission lock in the database, or the elimination of an error-prone manual step.
Simple practices that work in daily life
You do not need a million-dollar training budget to build a continuous learning culture:
- Regular and honest retrospectives: not just out of calendar obligation at the end of the sprint, but whenever a relevant delivery or incident ends.
- Shadowing (practical accompaniment): put junior or mid-level professionals to follow senior engineers during critical diagnostics and architecture changes.
- Brown bags and technical exchanges: short, informal meetings where someone from the team itself shares a real solution or lesson learned in the week within thirty minutes.
- Documentation as a natural habit: updating the operation guide is part of the definition of a "done task", never a punishment left for when there is spare time.
The knowledge test and competitive advantage
Take a stress test in your area today: if a key person on your team receives an offer and leaves the company tomorrow, how much critical knowledge of the operation goes away along with them through the door?
The size of this hole is your learning debt (known in engineering as the Bus Factor). And, exactly like technical debt in code, it charges expensive interest when least expected. Mature teams deliberately invest in knowledge redundancy: nobody should be the only human being in the company who knows how a vital service works.
In the long run, in a market where tools change every year, the company that learns the fastest wins. It is not necessarily the largest nor the one with the most accumulated capital: it is the most adaptable. And real adaptability starts with teams that treat shared knowledge as their greatest strategic asset.
Continue deepening: also read Categorization and Priority and Culture that Secures or Sabotages. Also learn about work in Executive Engineering Leadership and the Case Studies.
References and verifiable readings
- Edmondson, A. C. (1999). Psychological Safety and Learning Behavior in Work Teams. Administrative Science Quarterly, 44(2), 350-383.
- Senge, P. M. (1990 / 2006). The Fifth Discipline: The Art and Practice of the Learning Organization. Doubleday / Currency.
- Dweck, C. S. (2006). Mindset: The New Psychology of Success. Random House.
Good ideas keep the conversation going.
Share this article or publish the visual deck to your Stories and Feed.

Cesar Zanis
15+ years leading software architecture, database performance, FinOps, and applied AI across enterprise operations. Built and scaled an engineering organization from zero to 20+ developers.
Enjoyed this article?
I publish practical analyses on critical systems architecture, Cloud FinOps, data, AI, and engineering leadership directly on LinkedIn.
Continue Reading
Take this analysis further with data and sources
All visual cards of this article are ready above to share on your Stories, Feed, WhatsApp, or LinkedIn with full sources.



