Methodology
How this site builds its content, what the classifications mean, and where the information comes from.
How Mappings Are Created
Each mapping starts with a real business task — something a person in a specific role actually does at work. The author identifies the task from direct professional experience, industry documentation, or practitioner interviews, then researches which AI technologies apply to that task and how they change the work.
This is a business practitioner interpretation, not academic research. The goal is to describe AI in the language that business people already use — not to translate business into technical jargon. Every mapping is written from the perspective of someone who does the work, not someone who builds the technology.
The site currently contains 612 mappings across 21 industries and 325 functions. Coverage varies by industry — some industries have deeper function-level detail than others.
Impact Classifications
Every mapping receives one of four impact classifications based on how AI changes the nature of the work:
Enhances
AI makes the existing task faster, more accurate, or more informed. The person still does the work. The workflow stays similar. The tools get better.
Automates
AI handles the routine portion of the task without human involvement. The person shifts to exceptions, oversight, and quality control. Volume work moves to machines; judgment work stays with people.
Transforms
AI changes the nature of the task itself. The workflow is fundamentally different. The skills required shift. The job title may stay the same, but what the person does day-to-day looks different.
Replaces (retired)
This classification was retired. In practice, even fully automated tasks require human oversight for edge cases and exceptions. All current mappings use Enhances, Automates, or Transforms.
These classifications reflect the author's professional judgment about the current and near-term state of AI capability. They are directional assessments, not predictions. The actual impact in any specific organization depends on the technology stack, data quality, regulatory environment, and organizational readiness.
Technology Matching
Technologies listed on each mapping represent the category of AI capability that applies — not specific vendor products. “NLP Classification” means the general capability of using natural language processing to classify text, not a specific vendor's product.
Technology names are standardized across mappings so the same capability is described consistently whether it appears in insurance underwriting or retail inventory management. The Technology Index shows all 2,355 technologies and which mappings reference them.
Editorial Process
This site was built using AI-assisted research and content generation with human editorial review. AI tools were used to scale content production, technology definitions, and structural formatting. Every mapping is reviewed for accuracy and industry relevance.
Core mappings, impact classifications, and action guidance reflect the author's professional judgment informed by 20 years of enterprise transformation experience across insurance, financial services, healthcare, and technology.
Where sources are referenced, they are directional references — not formal academic citations. The site prioritizes practical accuracy over academic rigor. If a mapping contains an error, the feedback widget on every mapping page allows readers to flag it for correction.
Corrections & Feedback
Every mapping page includes a feedback widget. Readers can flag errors, suggest improvements, or note when a mapping no longer reflects current practice. Feedback is reviewed by the author and incorporated into updates.
If you work in one of the industries covered and see something that doesn't match your experience, the feedback is genuinely valued — this site is built to be corrected by practitioners, not defended by its author.