The Lab
UX, AI & Product
Hands-on experiments and analysis around digital platforms.
Current Project
German Statutory Health Insurance Websites: What Do You Review When You Cannot Review Everything?
56,198 public pages from 84 German statutory health insurance web properties, AI-assisted and triaged. Which review patterns recur across insurers, and how editorial teams turn them into an order of work.
Read articleContent Governance for AI Search in Regulated Industries
Why regulated industries do not need another GEO tool, but an operating system for supportable claims, AI visibility, conversion and compliance.
Read articleInterpretation Stability: Keeping Boundaries Intact in AI Answers
A testable framework for claim fidelity: measure whether conditions, evidence levels and validity survive AI summarization and rephrasing.
Read articleWeekly AI Briefing Agent: What My Filter Really Needs
A working AI news filter, its prompt failures and three rules for relevance, sources and real tests instead of daily FOMO.
Read articleGemini Live in the Browser: A Voice AI Prototype in Three Hours
A browser demo with native audio and function calling. What the prototype shows—and what remains unresolved for costs, privacy, and real-world use in a medical practice.
Read articleSchema, llms.txt and Structured Outputs: What Each Is Good For
Schema.org helps search engines, llms.txt can guide agents and Structured Outputs enforce a format. Four layers that are often confused.
Read articleWhy Google Clicks Are Disappearing: AI Overviews, Zero-Click, and CTR
Pew, Ahrefs, and academic research measure fewer clicks when AI Overviews appear. What the figures actually show and which SEO metrics matter now.
Read articleHanzi Movie Method: How It Works (+ Free Script App)
Actor, Set, Room, Props, Script: how the Hanzi Movie Method turns Chinese characters into movie scenes. A free browser app writes the scripts — no sign-up.
Read articleGEO: What AI Overviews Really Change About SEO
GEO remains SEO, but adds a measurement layer: whether content is found, cited and represented accurately in AI Overviews and ChatGPT.
Read articleHumanizer (German): A Claude Skill for More Human AI Text
Humanizer (German) for Claude Code and Codex: edit 72 German AI-writing patterns without Python. Open source under MIT and CC BY-SA.
Read articleVerbalized Sampling: More Creative AI Responses Without Fine-Tuning
A prompt for more varied responses, with rough model probabilities, a cost trade-off, and a clear selection rule.
Read articleBad UX Endangers IT Security and Inclusion
A documented CAPTCHA case shows why abuse prevention, server-side validation and accessibility must be tested together.
Read articleContent Testing: Test Clarity, Measure Impact
A lean content-testing process for clearer interfaces, supportable conversion hypotheses and more reliable AI conversations.
Read articleAI Personas as Hypotheses: Making Contradictions Visible
How synthetic personas can produce usage hypotheses and test questions without replacing interviews, behavioral data, or real users.
Read articleDark Patterns in UX Design at Banks
An anonymized banking case shows how a reward, unclear purposes, and repeated prompts can distort consent.
Read article