Knowledge management (KM) emerged from a 1987 McKinsey internal study and went public at an Ernst & Young conference in Boston in 1993. [2] Tom Davenport offered the classic one-line definition the following year: 'Knowledge Management is the process of capturing, distributing, and effectively using knowledge.' [2] A few years later, the Gartner Group framed it more broadly as 'a discipline that promotes an integrated approach to identifying, capturing, evaluating, retrieving, and sharing all of an enterprise's information assets.' [2] APQC narrows the same idea down to its purpose: KM is 'the application of a structured process to help information and knowledge flow to the right people at the right time.' [3] IBM puts it more operationally: 'the process of identifying, organizing, storing and disseminating information within an organization.' [4] In practice, KM treats know-how as an asset: pulling it out of the people who have it, structuring it so others can find it, and keeping it current. [1] [4] In operations, that mostly means converting tacit, hands-on know-how into explicit, documented procedures, and making those procedures accessible at the moment of need (the workstation, the field, the shop floor) rather than buried in a folder. [1]
Key characteristics
- Distinguishes between tacit knowledge (in heads) and explicit knowledge (written or recorded), and works to convert one into the other. [1]
- Captures knowledge from people through interviews, recordings, templates, and lessons-learned databases. [1] [2]
- Organises and retrieves knowledge through search, taxonomy, expertise location, and point-of-use access. [2]
- Keeps knowledge current with named owners, review cycles, and feedback loops.
- Is judged on whether people use the knowledge, not on how much content exists. [1]
Example
Building a knowledge base for a 60-person field service team
A field service company employs 60 technicians who repair commercial refrigeration units. Senior techs know how to diagnose 200+ failure modes from sound, smell, and gauge readings. The company starts a KM program: each tech records a 2-minute video any time they encounter a tricky failure, narrating the diagnosis. Within a year, the library has 400 short procedures, tagged by equipment model and symptom. New techs in the field scan a QR code on the unit, type a symptom, and get a curated set of likely fixes. First-visit resolution rates climb 20 points.
How SOPX handles this
Most KM programs fail because writing is slow and search is bad. SOPX shortcuts both. Field workers and operators record short videos of what they did and why, the AI converts each recording into a structured, searchable procedure with steps and screenshots, and the whole library is organised, version-controlled, and translatable into 50+ languages. The bottleneck stops being 'we should write this down someday' and becomes 'pull out your phone and record it now'.
Related use case: Knowledge Retention →Frequently asked questions
What is the difference between knowledge management and document management?
What is the difference between tacit and explicit knowledge?
What are the components of a KM system?
How do you start a knowledge management program?
Who owns knowledge management in a company?
How do you keep a knowledge base from going stale?
Sources
Statements above draw on the references below. Numbers in the text link to the matching entry.
- [1]Knowledge managementWikipedia · Accessed 2026-04-28
- [2]What is KM? Knowledge Management ExplainedKMWorld · Accessed 2026-04-28
- [3]What Is Knowledge Management?APQC · Accessed 2026-04-28
- [4]What is knowledge management?IBM · Accessed 2026-04-28