Gap Analysis and Process Improvement
Gap analysis compares what a process or organization achieves today against an outcome it intends to reach, then names the specific differences that must be closed. Process improvement is the disciplined work of closing those differences through targeted changes, measured results, and repeated cycles.
itEngineering leadership and delivery management | OpenSkills.info
Course pathWalk it in order
Look it upDip in anytime
Go furtherLeaves this page
Don't Panic
Don't Panic - Gap Analysis and Process Improvement
Gap analysis is the awkward honesty of comparing what a process produces today with an outcome somebody actually committed to. Process improvement is the disciplined work of closing that difference in a way you can measure twice. One names the change; the other makes it. Skip either half and you get either a complaint or a random tweak.
A gap is not a mood. "Orders take too long" is a symptom. "Median order cycle time is 6.4 hours against a target of 2 hours" is a gap. The second sentence tells you what to investigate, what to measure, and when to stop celebrating.
Everything hangs on a loop with four blunt questions. Plan asks for the target, the current state, and the gap. Do makes one controlled change aimed at part of that gap. Study checks the result against a prediction. Act adopts, adjusts, or abandons, then starts again. Deming called it Plan-Do-Study-Act. ISO 9001, Six Sigma DMAIC, and Lean wear different hats over the same shape.
Define the target first or the gap is fiction. A usable target carries an outcome, a measure, a threshold, and a deadline. "Be better at incident response" fails that test on contact. Current state comes from evidence: logs, tickets, observation, interviews, customer feedback, and process mining when event data exists. The gap between the written procedure and what people actually do is itself a finding.
Then structure each gap so a decision can act: magnitude, impact, owner, evidence. Group by cause, not by symptom. Prioritize with Pareto, cost, risk, and feasibility, but do not let the chart replace root-cause work. Five Whys, Ishikawa, FMEA, and change analysis are hypothesis tools. A candidate cause that has not been tested against evidence is still a guess.
Read the Intro for the full loop and the vocabulary map. Use the Cheatsheet when you need the gap statement template and method table. The Practice Reference and Exercise walk one bounded improvement cycle without inventing a whole transformation program.
Where this skill leads
Relevant careers
See how this topic contributes to broader role-level skill maps.
Sources
- https://deming.org/explore/pdsa/
Supports
- PDSA (Plan-Do-Study-Act) as a systematic process for gaining knowledge and improving a product, process, or service through repeated cycles
- Deming's emphasis on Study over Check and on prediction as the basis for learning
- The Plan, Do, Study, Act phase definitions used in this course's loop
- Shewhart as the origin of the cycle that Deming adapted
- https://www.ibm.com/think/topics/business-process-management
Supports
- BPM as the discipline of discovering, modeling, analyzing, measuring, improving, and optimizing repeatable processes
- Six Sigma and Lean as examples of BPM methodologies
- Distinction between repeatable processes, individual tasks, and one-time projects
- The BPM lifecycle of design, model, execute, monitor, optimize
- https://www.ibm.com/think/topics/business-process-analysis
Supports
- As-is analysis before process improvement
- Use of process diagrams, records, formal and informal behavior, and end-to-end boundaries
- The gap between documented procedure and actual practice as a finding
- Modeling as evidence for improvement rather than automation approval
- https://www.ibm.com/think/topics/root-cause-analysis
Supports
- Root cause analysis as the quality management practice of searching for the underlying cause after a problem occurs
- Methods: 5 Whys, Ishikawa (fishbone), FMEA, Pareto chart, change analysis, causal factor tree, DMAIC
- 5 Whys as a prompt to keep asking why until reaching a cause you can act on
- FMEA scoring by severity, occurrence, and detectability
- Six Sigma as a data-driven process improvement methodology
- GE's corporate-wide adoption of Six Sigma in 1995
- Motorola's introduction of Six Sigma in 1986
- https://www.ibm.com/think/topics/process-mining
Supports
- Process mining applies algorithms to event log data to discover, validate, and improve workflows
- Three types of process mining: discovery, conformance, enhancement
- Wil van der Aalst as the academic credited with much of the research around process mining
- IEEE Process Mining Manifesto published in 2011
- Process mining's limitations: data quality, missing manual tasks, concept drift
- The market consolidation including SAP's acquisition of Signavio, Celonis, and UiPath's acquisition of ProcessGold
- https://www.ibm.com/think/topics/business-process-improvement
Supports
- Process improvement framing of targeted, measured changes to close specific gaps
- End-to-end process view rather than step-only optimization
- https://en.wikipedia.org/wiki/Gap_analysis
Supports
- Gap analysis as the comparison of actual performance with potential or desired performance
- Gap analysis at strategic or operational levels of an organization
- Gap analysis as a formal study of what a business is doing currently and where it wants to go
- Use of gap analysis to develop a better process by comparing existing and desired outcomes
- https://www.iso.org/standard/72894.html
Supports
- ISO 9001:2015 as the quality management systems standard
- Continual improvement of the quality management system as the ISO framing of the improvement loop
- First published in 1987
- https://asq.org/quality-resources/plan-do-study-act-continual-improvement
Supports
- Cross-reference description of the PDSA cycle from the quality management profession
- https://www.omg.org/spec/BPMN/2.0.2/
Supports
- BPMN 2.0.2 as the OMG standard for graphical business process models
- OMG stewardship of the notation used for as-is and to-be process models
- OMG adoption of BPMN 2.0 in 2011
- https://github.com/TheWoops/awesome-processmining
Supports
- Community-curated list of process mining libraries and tools
- PM4Py, Apromore, ProM, bupaR, and PAFnow as process mining resources for learners
- Fluxicon Process Mining Book as a free further resource
- ICPM as the International Conference on Process Mining