30 to 50 % of enterprise AI initiatives fail to return what they should (Gartner: at least 30 % of proofs of concept abandoned). Idun Group structures your AI support engagement around value: analysis of the real usage of your AI tools from the logs you already have, strategic advice with a 6-month to 2-year roadmap prioritised by return on investment, then development of the use cases proven profitable and nothing else. A consultancy specialised in AI since late 2023, more than 40 AI diagnostics run in under a year, more than 10 large-account clients.
The primary cause is the absence of measurement: across the estates we analyse, 20 to 40 % of the AI licences paid for go unused, inherited projects pile up, and nobody knows the full cost of AI. A structured AI support engagement starts with visibility and with taking back control of AI usage: measure what exists, cost each use case, decide on facts.
1. Analysis of real AI usage: logs and exports from the platforms you administer, installing nothing and re-instrumenting nothing; a map across every tool, estimated return on investment per use case, the waste identified, the consolidated full cost of AI, a report ready for the steering committee. Interviews with the IT department and the business units are a mandatory step: the logs give the facts, the interviews give the intent. 2. Strategic advice and a 6-month to 2-year roadmap: prioritisation by return on investment, effort and adoption, licence arbitration, standardisation. 3. Downstream, on evidence: development of the use cases proven profitable (the code we write belongs to you) or training where the analysis shows an adoption gap. No pricing is published: every quote follows the analysis and is given in a meeting.
A first institutional conversation narrows a vague need, without measuring real usage or producing an actionable recommendation. Idun Group grounds every recommendation in your data: read-only, pseudonymisation and aggregation by design, an analysis environment inside the European Union by default, execution within your own perimeter if your data protection officer requires it. If the analysis does not prove a return on investment, we tell you so in writing and nothing is committed. No Idun product is recommended in the deliverables.
LVMH: design and operation of a generative AI agent platform in production. Randstad: migration of 100 applications in 6 months, against 12 months initially estimated. HynoTech: design and development of a marketplace.
