What AI workflow implementation looks like in practice, for small Swiss service businesses. No abstractions, no vendor case studies, no benchmarks from companies with a hundred people. The work described here is ours.
Most small businesses know exactly which tasks waste their time. What they don't have is someone who will map those tasks and build the system that eliminates them.
Read the pieceMost automation projects fail because they start in the wrong place. Follow-up sequences, document assembly, reporting, onboarding, scheduling: here is where to start and why.
Read the pieceWe quoted four weeks. It took eleven. A data quality problem we did not catch early enough, and what we learned about scoping engagements for businesses that have never documented their own processes.
Read the accountA four-lawyer firm in Lugano was spending 9.2 hours of partner-level time per month on client intake. The cost was invisible because no one had counted it. Here is what the Clarity Scan found, what we built, and what the result looked like five weeks later.
Read the case studyThe most painful task and the most expensive task are almost never the same one. The reason owners consistently get this wrong. And what a structured diagnostic finds that intuition misses.
Read the pieceThree workflows, one spreadsheet, CHF 30,020 per year in recoverable capacity. The maths is simple once you do it. The hard part is making the cost visible in the first place.
Read the analysisA month-by-month account of what happened after the diagnostic. Including what broke, how it was caught, and what the receptionist said in month five that didn't appear in any spreadsheet.
Read the accountMost businesses start by asking which tools to buy. That is the wrong question. Here is what a structured audit finds that intuition misses.
Read the pieceAccounting practices lose 10 to 18 hours per week to manual data handling and reporting cycles. Here is what AI can reliably automate today.
Read the pieceThe complete cost structure: from CHF 490 diagnostic (introductory), CHF 3,000–8,000 per sprint, CHF 500–1,500 monthly maintenance. With the return figures that make the numbers make sense.
Read the analysisLaw firms are cautious about automation for good reason. But caution is often applied to the wrong workflows. What can be automated safely in a Swiss legal practice.
Read the pieceThe most common mistake with AI automation is starting before the conditions for success are in place. Read this before you commit to anything.
Read the piecePaper intake forms, 45 minutes a day of reminder calls, and a monthly billing reconciliation that consumed a Saturday afternoon. What the Clarity Scan found and what changed.
Read the case study38% of inbound inquiries were receiving zero follow-up. Not because agents didn't care — because there was no system. What the numbers looked like eight weeks after the Sprint.
Read the case studyCHF 14,400 per year in worked hours that were never invoiced. The time was spent. It was simply not captured. What happens when you compare the calendar against the invoices.
Read the case studyThe Clarity Scan takes 7 business days. A Sprint takes 6 weeks. Total time from you across both: around 3.5 hours. Specific answers to every calendar question before starting.
Read the articleA 3-person firm losing 4 hours/week per person loses CHF 103,680 per year in capacity. Small teams are often better candidates than large ones. Here is the arithmetic.
Read the articleThree patterns account for most failed automation projects. None of them are about the technology. All of them are about the order of operations.
Read the articleMost professionals only measure one of the three components. Time cost, error cost, and opportunity cost. The full formula consistently produces a number 2-3x higher than the estimate.
Read the articleAn honest comparison of four routes to automation, including when MEIKAI is not the right fit. Cost, timeline, risk, and maintenance for each option.
Read the comparisonThe tool question comes up in almost every Clarity Scan conversation. The honest answer depends on your technical profile, budget, and workflow count.
Read the comparison