For corporates
We set up your Center of Excellence for automation and applied AI: platform, governance, champions and a backlog that pays for itself. Then we hand over, on a date we plan together.
A small, permanent team inside your organisation that owns the platform (self-hosted n8n or Make, a model gateway), the rules (intake, risk classes, data protection, works council) and the people (trained champions in the departments). It turns scattered pilots into a managed portfolio of automations that run in production and are measured in hours and euros. coompanion builds it with you and steps back on purpose.
Why a Center of Excellence
Most corporates do not lack ideas or tools. They lack one place that owns automation end to end.
Every department has tried something with AI. Few things survived the demo, because nobody owned operations, monitoring and the hand-over to the people who do the work.
Sales automates in one platform, finance in another, operations with a freelancer. Licences multiply, nothing is reusable, and the same customer data flows through four systems nobody has mapped.
Legal, IT security and the works council see the automation when it is already built. The answer is no, and it is the right answer, because nobody designed the data flow with them.
A Center of Excellence is not another innovation lab. It is an operating unit: it runs the platform, sets the rules, trains the champions in the departments and keeps a backlog that is prioritised by hours saved.
What the CoE consists of
Every Center of Excellence we set up has the same five parts. The order matters: platform and rules first, then people, then volume.
Self-hosted n8n or Make Enterprise, a model gateway for Azure OpenAI, Mistral or local models, and one way of connecting systems. Everything your champions build runs here, logged and versioned.
Intake form, risk classes, approval paths and data rules agreed with legal, IT security and the works council before the first workflow goes live. Governance is what makes the no disappear.
We train two to three people per department to build and run their own automations. The CoE supports, reviews and maintains standards. Knowledge stays where the work is.
Every idea gets an hours-and-risk estimate. The CoE delivers the first processes with the champions, then shifts to coaching. Each process ships with a runbook and an owner.
A dashboard that shows runs, failures, hours saved and cost per process. The CoE reports to management quarterly, in the language of operations, not in demo videos.
How we build it
Who this is for
The CoE works when operations, IT and the departments sit at one table. These are the roles we usually start with.
Questions
A lab explores. A Center of Excellence operates. It owns a platform in production, rules that legal and the works council have signed, trained people in the departments and a backlog measured in hours. Labs produce demos; the CoE produces processes that run every day.
Usually self-hosted n8n or Make Enterprise, combined with a model gateway so you can switch between Azure OpenAI, Mistral or local models without rebuilding workflows. We recommend based on your IT landscape, data rules and existing licences, and we do not resell software.
Twelve weeks to the first processes in production and trained champions, then a scale phase of three to six months in which we coach rather than build. The hand-over date is agreed in the charter at the start, and it is in our interest to hit it.
From week one. They are interviewed in the readiness phase, co-author the risk classes and approval paths in phase one and receive a works council agreement template plus AI Act documentation. Nothing goes live that they have not seen.
The two-week readiness phase is a fixed price. Set-up and build are quoted after it, based on the processes you choose, and typically pay back within six months. Licences stay with you, we do not add a margin on them.
Thirty minutes with a founder. We look at your process landscape, your IT constraints and tell you honestly whether a Center of Excellence is the right move now.
Datenschutz
Wir verwenden Cookies zur Analyse und für Marketing. Notwendige Cookies sind immer aktiv.
