Entries cover RAPID itself, the academic models it draws on, the metrics surfaced on results, and the source-tier conventions used in /sources. Every entry has a stable anchor (e.g. /glossary#tam) so other pages can deep-link.
A five-dimension maturity framework for GenAI investment - Readiness, Alignment, Portfolio, Impact, Diffusion - that scores an organisation against a 44-case enterprise dataset.
Each letter maps to a foundational management-research model: Readiness draws on UTAUT, Alignment on TAM, Portfolio on Markowitz, Impact on DeLone & McLean, Diffusion on Rogers and Kotter.
Data and technical readiness: data infrastructure, MLOps maturity, model-evaluation discipline, and AI/ML talent depth.
Strategic alignment: how well GenAI investments are tied to measurable business objectives with named executive sponsorship.
Portfolio balance: diversification across risk levels and time horizons; the discipline to kill or scale initiatives based on results.
Measurement maturity: KPI definition, baseline establishment, and ROI review processes for GenAI initiatives.
Organisational adoption: change management, user training, and cross-functional collaboration that turn a deployment into a behaviour change.
Buckets that summarise an overall RAPID score: Nascent (0-25%), Developing (26-50%), Established (51-75%), Advanced (76-100%).
Levels are coarse so cross-organisation comparison is meaningful at scale; per-dimension percentages give the precision needed for a roadmap.
A five-element checklist that predicts deployment success: business objectives, performance tracking, change management, iterative deployment, and a named executive sponsor.
Organisations with four or more elements see ~82% success in the dataset; zero to two drops to 23%.
Davis (1989). Predicts technology adoption from perceived usefulness and perceived ease of use. Strongest predictor of whether a tool gets used at all.
Venkatesh et al. (2003). Unifies eight technology-adoption models. Performance expectancy, effort expectancy, social influence, and facilitating conditions explain ~70% of adoption variance.
Kotter (1996). Eight sequential steps for organisational change: urgency, coalition, vision, communication, empowerment, wins, consolidation, anchoring.
DeLone & McLean (2003). Six interdependent dimensions of IS success: information quality, system quality, service quality, use, user satisfaction, and net benefits.
Rogers (2003). Innovations spread through social systems via adoption categories: innovators, early adopters, early majority, late majority, laggards.
Brynjolfsson & Hitt (1993). IT productivity gains require organisational change with a 3-5 year lag. AI productivity figures show the same pattern - per-task minute savings do not aggregate automatically into firm-level productivity.
Peer-reviewed journals, NBER working papers, government findings, and investigative-journalism cases with documented evidence trails.
Business press, corporate disclosures, and analyst reports with primary-source URLs but lower formal-vetting depth than Tier 1.
A specific cited case or academic citation that supports a claim in the framework. The Task Mapper labels every quantitative claim either anchored or illustrative.
Stable identifier (CASE-001 through CASE-045) for each entry in the 44-case dataset. Used in citations and cross-references throughout the site.
One of six categories where GenAI deployments tend to break down: data, technical, organisational, strategic, measurement, economic.
Each failure mode is mapped to specific cases in the dataset (e.g. CASE-010 Amazon hiring tool is an organisational + data failure).
One of six GenAI deployment archetypes: Automation, Generation, Analysis, Augmentation, Chatbots, Decision Support. Each has a verified success-rate band derived from the dataset.
Peffers et al. (2007). The development methodology behind RAPID: identify problem, define objectives, design artefact, demonstrate, evaluate, communicate.
Share of measurable cases (success + failure + partial) in a cohort that achieved their stated objective. Excludes Too Early outcomes.
Where an organisation's score sits relative to peers in the dataset. The 60th percentile means 60% of peers scored at or below this point.
Industry-specific projected return on GenAI investment, derived from thesis Chapter 4. Tech 18-24%, Financial Services 15-22%, Healthcare 12-18% are the most-studied bands.
Typical payback window for GenAI investments by industry. Ranges from 12-18 months (tech, retail) to 24-36 months (government, regulated sectors).
Share of GenAI proofs-of-concept that never reach production. Gartner forecasts 30% by end of 2025; BCG finds only 22% of organisations advance past POC at all.
Where a case sits on the maturity arc: Pilot/POC, Scaling, Production, Optimization, or Abandoned. The 44-case dataset is weighted toward Production and Optimization.
The curated, primary-source-linked corpus behind the framework. Each case has an organisation, period, industry, function, outcome, evidence statement, and source URL.
Maintained as a separate citable artefact so it can evolve independently of the website. See /sources for the searchable view.
Width of the band around an industry benchmark, reflecting sample size. Industries with fewer than 10 cases (Healthcare, Manufacturing, Government) carry wider intervals.
On the team-comparison view (/compare/team): a dimension where every participant scored below 40%. Likely the cheapest place for the team to improve together.
Want to cite the framework itself? Use the methodology page's Cite-this widget for BibTeX, APA, and MLA references. Each entry in /sources also has a Cite-this affordance for case-level references.