Jakob Lange

05 Field · UX

Enterprise UX & Workflow Design

Dashboards, internal software and AI-assisted workflows that people use because they work — researched, accessible, measured.

Enterprise software is where UX debt is most expensive and least visible. A clumsy internal tool costs minutes per task, multiplied by hundreds of people, every day — and it quietly trains people to work around the system, which is where errors and security incidents begin. Designing for the people who run the organisation is not a nice-to-have; it is operational risk management.

Why this matters now

AI is being inserted into workflows at speed: copilots in ticketing systems, summarisers in case management, agents that pre-fill forms. Whether these help depends almost entirely on interaction design — how confidence is communicated, how a wrong suggestion is corrected, how responsibility is kept with the human. Poorly designed AI assistance produces either automation bias or rejection, and both are measurable in the numbers.

At the same time, the Barrierefreiheitsstärkungsgesetz has applied since June 2025 and extends accessibility obligations to many digital products and services, while public-sector and enterprise procurement increasingly requires WCAG conformance. Accessible internal software is also simply better software: clearer, faster, more robust.

How I approach it

I start in the field. Contextual inquiry — watching people do the actual work, on their actual screens, under their actual time pressure — reveals more in a day than a month of stakeholder workshops. Findings are turned into task models and prioritised pain points, then into information architecture and prototypes that are tested with the same people before anything is built.

For AI-assisted workflows I apply the human-in-the-loop evaluation methods from my research work: where does oversight belong, what does the interface need to show for a decision to be defensible, how does a person recover when the assistant is wrong.

Where it typically starts

A research sprint of two to three weeks on one tool or one workflow, ending with validated findings and a design direction. Larger engagements cover design, prototyping and the measurement that proves the change worked.

Questions I am usually asked

  • We built an internal tool and nobody uses it. What went wrong?
  • Our dashboards have two hundred KPIs. Which ones actually drive decisions?
  • How do we put an AI assistant into a workflow without people blindly trusting it — or ignoring it?
  • The accessibility law applies to us since 2025. What does that mean for our internal software?
  • How do we prove that a redesign saved time and reduced errors?

Typical deliverables

  • Research findings with task models and pain-point map
  • Information architecture and interaction design for the tool or dashboard
  • Clickable prototype validated with real users
  • Accessibility audit and remediation plan
  • Adoption and efficiency baseline with follow-up measurement

Contact

Start with a conversation.

No forms, no funnels. Write me a short note about your situation — I answer personally, usually within two working days.

Mon – Fri, 18:00 – 20:00 CET