liveBook · Manning

Human error

As explained in Think Like a CTO

Human error is an unintended mistake made by a person while performing a task or operating a process.

Human error is an unintended mistake made while carrying out a task. In an organization, it may expose a weakness in a manual workflow or contribute to a service failure. Reducing the opportunity for mistakes can help without requiring a wholesale change to how people work. section 2.3.3 section 7.2

When an error causes an incident, a manager should take responsibility rather than blame the person publicly. Avoid intimidation, insults, and immediate termination. First establish enough of the cause to stop the problem continuing; then plan how to prevent a repeat or limit its impact. section 7.2.1

Investigate the conditions around the mistake before choosing a remedy:

  • Distinguish a genuine mistake from a lack of foresight or planning. The five whys can guide questioning, though an incident may have several causes. section 7.2.2
  • Ask whether training was adequate, the case was unusual, distraction or stress played a part, or steps were missed. section 7.2.2
  • Choose proportionate safeguards, such as checks, software guardrails, retraining, or additional reviews. Record what happened, why, and how to rectify it. section 7.2.2

Small changes that prevent errors

A shared Google Sheet replaced emailed spreadsheet versions that people could accidentally update out of order, without disrupting a familiar workflow. In another incident, an operator deleted production web servers while believing a terminal window was connected to development; production windows were subsequently given a conspicuous color. section 2.3.3 section 7.2.2

How Think Like a CTO covers it

How deep it goes
Develops a response and investigation procedure, but stops short of implementing the suggested software guardrails.
Examples
Uses an emailed-spreadsheet version mix-up and an accidental production-server deletion from the wrong terminal window.
What it assumes
Assumes familiarity with workplace processes and operational incidents; does not teach technical setup from zero.
What you can do after
Practice/perform a calm incident response: contain the problem, investigate contributing factors, document rectification, and select proportionate safeguards.
How it connects
Links human error to business continuity, root cause analysis, the five whys, training, and software guardrails.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Think Like a CTO. The section links point to the parts of the book it draws on.