Blog
What really works in support
Articles about automation, ticket load, privacy and the knowledge base behind it. Written for teams that have to decide — not for the search engine.
Support operations7 minMultiple brands, one support team: what to separate and share
One support team, several brands: when each brand needs its own tenant, what teams share across tenants, and which work doubles.
Embedding10 minEmbedding a support widget in single-page apps: route changes, Shadow DOM, load time
What a support widget has to manage in a single-page app: survive route changes, isolate its styles, not slow the page down. With the technical reasons behind it.
AI quality6 minTest questions for the AI agent before knowledge goes live
A fixed set of test questions for your AI agent, run before every knowledge base release, catches side effects before customers see them.
AI quality12 minWho is liable when AI gives wrong advice in customer support?
Who is liable when your support AI agent advises wrongly? Court rulings, the labelling duty under Art. 50 AI Act, and four approval tiers for daily work.
AI quality8 minPrompt injection in support chat: protecting your AI agent
How to protect your support AI agent against prompt injection and token exhaustion attacks. Real countermeasures instead of mere word filters.
Knowledge base5 minHelp Centre in AI answer engines: how to get cited
Make your Help Centre visible in AI answer engines. Use targeted robots.txt rules, llms.txt and structured FAQ schema to earn more citations.
Automation7 minAI in financial support: what is allowed and what is not
Find out which customer service requests banks and insurers may automate with AI without legal risk, and where the GDPR draws a line.
Privacy11 minDoes the support widget need consent? § 25 TDDDG
Does a support widget need the user's consent? Learn what § 25 TDDDG requires, which positions exist and which route carries the least risk.
AI quality6 minLabelling AI in support: what Article 50 requires
Article 50 of the AI Act requires clear notices whenever artificial intelligence is used, from August 2026. Here is how to implement the labelling duty.
Knowledge base8 minAnalysing handovers to humans: knowledge gap or outage?
AI handovers have two causes: missing knowledge or system outages. Here is how to separate the two metrics in practice and make the right decisions.
Knowledge base5 minOne knowledge base, seven languages: what breaks in upkeep
Keeping review cycles in sync across languages, separating binding answers from market-specific ones, and deciding what applies when the source and the question are in different languages.
Metrics6 minAI support: what usage-based costs really drive
Find out how usage-based billing works in AI support, which metrics count, and how clear limits keep your costs under control.
Support operations8 minStatus pages: communicate proactively, reduce tickets
A proactive status page absorbs support tickets. Learn what to communicate, how often to update and which language builds trust when it matters.
Support operations7 minSupport tools: what fragmented channels cost in time
Find out how fragmented support tools burden resolution time and first-contact resolution. An honest look at the toggle tax in customer service.
Privacy7 minChecking a DPA: the points where it goes wrong in practice
How to check a data processing agreement (DPA) efficiently for sub-service providers, rights to issue instructions and audit clauses, so you avoid high fines.
Support operations5 minSupport response times: benchmarks and realistic targets
Discover current benchmarks for first response and resolution times in customer service. That is how you optimise processes in B2B SaaS and e-commerce for the long term.
Metrics6 minFirst contact resolution rate: why teams measure the wrong KPI
Find out why deflection rate and transfer rate mislead you in support and how to make the real first contact resolution rate (FCR) measurable.
Automation5 minEscalation to a human: getting the handover moment right
Find out how to get the handover moment from chatbot to human agents right in support. That is how you avoid bot hell and frustrated customers.
AI quality7 minPreventing hallucinations in support: what helps technically
Find out how to prevent AI hallucinations in customer support at a technical level: from RAG architectures through confidence thresholds to source binding.
Knowledge base5 minBuilding a Help Centre: structure customers actually use
Learn how to build a Help Centre that customers really use, with logical categories, optimal searchability and continuous upkeep.
Privacy6 minGDPR-compliant AI in support: what really matters
Assess AI providers for customer service with confidence. How to make sure that the DPA, EU hosting and deletion periods stand up to GDPR requirements.
Tool selection6 minHelpdesk automation compared: chatbot, macro, AI agent
Helpdesk automation compared: when macros, chatbots or AI agents are the right choice for your customer service. Honest and practical.
Automation6 minAutomating customer service: which requests AI solves in 2026
Find out which support requests AI reliably automates in 2026 and where bots fail. A practical guide for SaaS and e-commerce, without the hype.
Support operations6 minReducing ticket volume: the levers that measurably work
Find out how to measurably reduce your support ticket volume with a solid knowledge base, AI self-service and removing the causes in the product.
Knowledge base5 minKnowledge base for AI support: structure and mistakes
A guide to building and maintaining a knowledge base for AI support, to avoid the typical data errors and secure precise answers.