TOOL ROLLOUT
A tool rollout is more than ordering a licence. It is about the right selection, the legal basis, the connection to your systems, the permissions, the first productive use cases and the handover to your team. We do this in six steps, typically within four to six weeks.
HOW WE WORK
Each step stands on its own and puts the next one within reach. You can join in where it fits: at the selection, at the integration, or at the training during rollout.
Which tool fits a company depends on the existing software, the focus of use cases and the privacy requirements. We compare ChatGPT, Claude, Gemini, Microsoft 365 Copilot and Langdock along these three axes. The result is a reasoned recommendation, not a matter of taste.
For each tool we clarify the deployment. EU hosting (Azure OpenAI EU, AWS Bedrock Frankfurt, Vertex AI EU, EU Data Boundary). Data Processing Agreement under Art. 28 GDPR. Training opt-out. Where needed, a Data Protection Impact Assessment. EU AI Act obligations documented.
Outlook, CRM, ERP, Aareon Wodis Sigma, DATEV, Salesforce, HubSpot, Propstack, onOffice. AI tools are only productive when they connect to the systems where the work happens anyway. We build the interfaces, via official APIs.
For Microsoft 365 Copilot in particular, the permission structure in SharePoint is critical. We review it before the rollout and clean up where needed. For every other platform we define roles and access, so that data only goes where it belongs.
Tool rollout and the first training go together. We start with a workshop for the departments in which the first productive use cases are worked through on real data. The in-house prompt and template library grows from there.
At the end we hand over to the internal owners. Documentation of the configuration, ownership for maintenance, update routines. You can run things yourself or pull us in on specific points. No multi-year lock-in.
EXAMPLE PLATFORMS
These are just the five most common examples. We are vendor-neutral and roll out every serious AI tool. Which one is used in a specific mandate falls out of the selection workshop.
Each platform has its own detail page covering privacy, use cases and practical examples.
OpenAI
Broadly applicable, large tool ecosystem, Custom GPTs for in-house templates.
In detailAnthropic
Strong on long documents, legal review, Computer Use for autonomous platform maintenance.
In detailVideo, audio, image and deep Workspace integration. NotebookLM as an in-house knowledge base.
In detailMicrosoft
AI directly in Outlook, Excel, Teams, SharePoint. Shortest path for Microsoft 365 shops.
In detailBerlin (DE)
GDPR-compliant workspace platform from Berlin. Custom assistants, knowledge bases, EU hosting native.
In detailPRIVACY AND LAW
Which legal basis applies, in which region processing takes place, what is in the Data Processing Agreement. We clarify these questions not at the end, but before the tool is activated.
CONNECTION TO YOUR SYSTEMS
AI in the browser is a toy. AI in the CRM, in the Outlook inbox or in the Excel workbook is a tool. We build the connections that make it possible.
TIMELINE
Typical timeline for a tool rollout at a mid-sized company. Often faster for smaller shops. For enterprise units with complex IT, accordingly longer.
Capture of requirements, comparison of candidate tools, recommendation with reasoning.
Data Processing Agreement, EU hosting configured, licence ordered via the company.
API connections to existing systems, permission model set up, pilot group defined.
First pilot group trained, workflows tested with real data, feedback incorporated.
Gradual extension to the departments, documentation, handover to internal owners.
In the intro call we clarify the software stack, the focus of use cases and the legal situation. The right tool selection falls out of that.
