Automation Approach Finder
Which automation approach fits your process?
Answer six short questions about a process. Instead of just a verdict, you'll see live on an interactive map where your process sits in the decision space – and how close or clear-cut the call actually is.
The Automation Approach Finder maps a process onto four solution paths based on rule-basedness and standardization: classic process automation, an AI assistant on company knowledge, custom software with an AI component, or custom software with classic logic.
Your process
Your approach space
75/100
75/100
All four approaches at a glance
Your recommendation
Classic process automation
Workflow tools (Zapier, Make, n8n) or a targeted script connect your systems and run the process by fixed rules – no AI needed, quick to implement, and cheap to maintain.
Doesn't fit if the process needs reading comprehension, judgment calls, or case-by-case decisions that don't reduce to rules.
AI assistant on company knowledge (RAG)
A language model answers queries or supports decisions by retrieving from your own knowledge base – suited to judgment calls within a clearly bounded subject area.
Doesn't fit if your data foundation is incomplete, scattered, or poorly maintained – fix the data foundation first.
Custom software with an AI component
A tailor-made application that combines your specific logic with AI-driven judgment – for cases too company-specific for standard tools and too knowledge-intensive for pure rule automation.
Doesn't fit if the process still runs rarely or changes frequently – start with a small prototype, not a full build.
Custom software (classic logic)
A tailor-made system with clear, rule-based logic – for processes that run deterministically but are too company-specific or too deeply woven into your own systems for standard tools.
Doesn't fit if standard tools already cover the requirement – then custom software is unnecessary effort.
Want to talk through this assessment with us?
How this finder works
- Two axes decide the right approach: rule-basedness (from decision logic + data type) and standardization (from company-specificity + system landscape) – each averaged from two questions.
- The recommendation is whichever corner of the space sits closest to your answers. If an axis lands close to the midpoint, the call counts as close – then we also show you the neighboring approach.
- Volume and stability don't change the approach itself, but add honest caveats: low volume questions whether the effort pays off at all; an unstable process argues for a small prototype instead of an immediate full build.
- Every approach links to the matching deep-dive calculator from our tools lineup – the finder is deliberately the entry point, not the final word.
This finder gives you a well-grounded first read based on your answers, not a quote. The right solution always depends on details that only a conversation can clarify.
Frequently asked questions about the finder
Why a map instead of just a result?
Because the choice between the four approaches is rarely clear-cut – a process often sits close to a boundary. The map shows not just the result, but how close or clear-cut it is, and what would need to change for a different approach to fit.
What if two approaches are close to each other?
Look at both descriptions – often the difference is small in practice and both paths are defensible. An AI Opportunity Scan with us clarifies this in detail based on your actual process.
Can I check several processes at once?
The finder is designed per process – run it separately for each candidate process to get the right call for each. For a side-by-side comparison across several processes, our AI Opportunity Scan is the right next step.
Where does the logic behind the assessment come from?
From our hands-on experience with all four solution paths: classic process automation, RAG-based AI assistants, and custom software with and without an AI component. The two axes (rule-basedness, standardization) capture what makes the biggest practical difference.