SAP consultancy · 130 consultants · SAP Partner of the Year 2024How Applium reached 90% AI adoption in 29 days.
Applium, a 130-person SAP consultancy, deployed Dust to augment its consultants and accelerate operational performance. With Idoption, they went from zero to 90% AI adoption in under a month — personalized AI agents for 70% of the team and direct integration into the SAP ecosystem.
Applium
Founded in 2002 by Bertrand Dalaison, Applium was built on one conviction: to bring genuinely innovative SAP services to clients who deserve better than standardized consulting.
Twenty-four years later, it's a 130-person firm with nine offices in France, a subsidiary in Portugal, €22M in revenue, and was named SAP Partner of the Year 2024. An average of 16 years of SAP experience per consultant. 77 client references spanning aerospace (ATR), food (Lactalis), retail (Boulanger), luxury (Chantelle, Boardriders) and the public sector (UGAP, Business France).
What makes Applium's AI adoption remarkable: this isn't a generalist firm testing productivity tooling. It's a firm of specialists giving itself the means to be even sharper.
What was holding them back, before Idoption.
When technical excellence is no longer enough
In SAP consulting, competition is relentless. Clients — large industrial groups, retailers, institutions — expect partners who are technically sharp, fast and up to date. For Applium, the question was never whether AI would change the profession. It was: would we lead that change or endure it?
We're convinced generative AI is non-negotiable to stay competitive in our sector. We wanted to remain a reliable, relevant, high-performing partner — and an attractive one for our own consultants.
The bottlenecks eating up bandwidth
Pre-sales & RFPs: handling complex RFPs meant reading enormous volumes, coordinating go/no-go decisions, and managing multi-author drafting. Every opportunity assessed consumed hours — sometimes entire days — for a team whose value is expertise, not admin.
Meeting minutes: inconsistent at best, skipped under pressure. Knowledge simply vanished.
Application maintenance (AMS) ticket management: spotting recurring patterns across hundreds of tickets required manual work from account managers. Surfacing systemic issues and advising proactively was hard.
What we shipped.
A pilot built to survive, not just to test
Applium came to Dust through a direct recommendation via the Idoption network. The decision demanded rigor: data security, GDPR compliance, independence from any single LLM. Dust met all three. Applium accelerated.
Approach co-designed with Idoption: a core team of 5 champions + ~15 people in a wider circle · a one-month timeline with real use cases from day 1 · 4–5 targeted use cases, each built to go straight to production. Key principle: nothing built during the pilot would be thrown away.
Once Dust convinced us on security, GDPR and data governance, we took an empirical approach. We wanted a powerful tool, independent of any single LLM, able to connect to our ecosystem. We tried it — the gains are phenomenal, the limits very modest by comparison.
The agents that went to production
Pre-sales & RFP agent: helps the team assess go/no-go decisions, process large tender documents, and coordinate drafting. It targets one of Applium's most time-consuming workflows head-on.
Personalized AI assistants: every consultant received their own agent configured for their context. 70% of the company is now equipped. Impact, in Joseph's words: “impossible to measure, but considerable.”
Automated meeting minutes: built on the Teams transcript integration (available since April 2026). The agent generates structured summaries from recordings. Simple, immediate, used where nothing was written down before.
A 4th use case — automated AMS ticket analysis — is running as an internal pilot, with a client-facing product view down the line.
What we measured.
97 of 120 connected in 29 days
Applium rolled out Dust in early April 2026. Twenty-nine days later: 97 of 120 users connected and signed up (~90%) · 60% regular users · 30–40% intensive users who rely on Dust for their daily work.
For a firm of seasoned SAP specialists — not natural early adopters — these numbers are remarkable.
There isn't a single important topic I work on today without AI. It's a real accelerator, a drastic change in the way I work.
Where the time saved shows
Meeting minutes: 20 min to 1 h saved per meeting — plus the systematic production of notes that simply weren't written before.
RFPs: a few hours to a few days saved per opportunity, letting the team assess a wider pipeline with more precision.
Personal assistants: deployed across the whole company to write, summarize and search — too broad to quantify individually, but omnipresent day to day.
What's next for Applium.
SAP integration and a new Level 0 product
Applium has already connected Dust to its internal SAP system — consultants now work directly in SAP with AI assistance. Next step: extend that integration to clients, after a full security / GDPR / governance mapping.
A new product line is emerging from the AMS ticket pilot: a subscription offer where the client receives an initial AI analysis of their tickets as a Level 0 response, with human escalation for Level 2–3.
Thirty hires planned in 2026. AI won't be an add-on for them — it'll be part of the job from day one.