“Management is not a science, not an art, it is a practice like medicine; you have a process of diagnosis, prognosis, and treatment” — Peter F. Drucker*

Generative AI changed the game, but made one issue unavoidable: the AI challenge is a data challenge. Unlike traditional analytics, foundation models such as those behind ChatGPT do not require causal models; they are largely correlation-based, so results depend on available data: no relevant data, no relevant AI result. For companies, missing context must be supplied, especially when it was not part of the model’s training data. For example, Retrieval-Augmented Generation, or RAG, does this by retrieving relevant data from internal systems or external sources and adding it to augment the prompt to improve specificity, grounding, and business usefulness. Before jumping to diagnosis or “medication” companies first need to establish symptoms: where does their AI-data readiness stand today? An index … based on “data as a product” … and “super-apps” … provides an efficient status-quo assessment to identify gaps, prioritize action, and move from AI ambition to AI results.

Data as a product
Raw data is not enough. AI needs refined, reusable data products. Much like food products need labels for ingredients and nutritional value, data products need metadata for information content, business context, meaning, quality, and source.
Data readiness is therefore not just a storage problem. It is first a productization problem, then an industrialization challenge. Companies need “data factories” that turn raw data into AI-ready products at scale, all under proper data governance.
See Data 3: Data products, digital twins – Compendium

Super-apps and agents are back
AI is changing applications from static tools into orchestrated systems. Super-apps integrate multiple services into one seamless experience, while agentic AI executes tasks, optimizes workflows, and coordinates automation across systems. These new app architectures can build on rich research in multi-agent systems and swarm computing. But their practical success depends on the same foundation: access to the right data across domains, companies, and value chains.
See SIM2: Agents (are back), Data 4: Super-apps – Compendium

AI-data readiness index: Concept and design
The AI-data gap is becoming the execution gap for digital transformation. An AI-Data Readiness Index is needed to help companies assess whether their data is usable, governed, discoverable, shareable, and relevant for AI applications. The index should cover the end-to-end data value chain: data sources, data products and factories, dataspaces for distribution, and application usage. It starts with business outcomes and works backward to the data capabilities required to create value. Results should show maturity from lagging to leading and identify practical improvement areas. The goal is not another abstract benchmark, but an executive decision tool to prioritize investments and move from AI ambition to AI results.
– For readiness index concept: Link
– For high-level design: Link
– For details, results, participation: Please contact us chris.langdon@cgu.edu

* Penta, A. 2017. Peter Drucker Speaks. A holographic interview – from Emmy Award winning Anthony Penta (Drucker Day 2017-11-16), Drucker School of Management, Claremont Graduate University, Claremont CA, link