Process automation is one of the most talked-about topics in business technology right now, and also one of the most misapplied. The organisations that get the most value from automation are not the ones that automate the most processes. They are the ones that automate the right processes, in a way that their team can actually maintain. This article sets out how we approach automation prioritisation with clients in Gifu and Aichi.
Start by measuring the actual time cost ¶
Before deciding whether to automate a process, measure how much time it actually consumes. This sounds obvious, but most organisations have only a rough sense of where staff time goes. We ask clients to track specific manual processes for two weeks before our first automation workshop. The results are often surprising: processes that feel burdensome turn out to take less time than expected, while others that nobody complains about are consuming significant hours across multiple departments.
Prioritise processes that are stable, not processes that are painful ¶
The most painful manual processes are often the ones that change frequently. Automating a process that changes every few months means rebuilding the automation every few months, which is expensive and demoralising. The best candidates for automation are processes that are stable, well-understood, and executed the same way every time. Data entry from one system to another, report generation from fixed data sources, and routine notification workflows are good starting points.
Build for maintainability, not sophistication ¶
The most common automation failure we see is a system that works perfectly until the person who built it leaves. When we design automations for clients, we use tools that the client's existing team can understand and modify without specialist support. We document every automation in plain language, and we include a half-day training session as part of every Automation Sprint engagement. The goal is that the automation is an asset the organisation owns, not a dependency on an external specialist.
Process automation done well frees up staff time for work that actually requires human judgment. Done poorly, it creates fragile systems that break at inconvenient moments and that nobody knows how to fix. The difference is usually in the prioritisation and the documentation, not in the sophistication of the technology.