Run the Do-Nothing Baseline Test
Most automation business cases compare a proposed robot against an imagined worst case rather than against reality. The result is a project that looks compelling on paper and underperforms against the process it replaced.
The Bot Scout do-nothing baseline test requires three comparisons before any purchase: the current process measured honestly, the same process after low-cost redesign, and the proposed automated system. A robot that beats the first but loses to the second is an expensive way to avoid a layout change.
The alternatives worth pricing are unglamorous and frequently win: better part presentation, a fixture, a layout change, a shift pattern, or a purpose-built machine. The International Federation of Robotics counts reprogrammable multipurpose manipulators, and flexibility is exactly what you pay a premium for — so a task that will never change may not deserve one.
| Situation | Automation likely wins | Alternative that often wins | Deciding factor |
|---|---|---|---|
| Repetitive, injury-prone task | Yes | Ergonomic redesign | Whether the task is stable |
| High-mix, low-volume work | Sometimes | Flexible fixturing and training | Changeover time |
| Stable, very high volume | Sometimes | Fixed automation or a purpose-built machine | Flexibility genuinely needed |
| Unpredictable manual work | Rarely | Process redesign first | Whether the process is documented |
| Staffing shortage | Sometimes | Shift pattern or scheduling change | Whether the bottleneck is the task |
What Belongs in an Honest Business Case
Include end effectors, fixtures, integration, controls, safety assessment, training, spare parts, maintenance, and the production time lost during commissioning. The robot itself is usually the smaller half.
The second-year line is the one most often omitted: re-teaching after a part change, wear items, and retraining after staff turnover. The process view is in the robotics in manufacturing guide and the machine-type view in the manufacturing robots guide.
Safety is a per-application obligation. ISO 10218-1:2025 sets robot safety requirements, ISO/TS 15066:2016 supplements them for collaborative systems, and OSHA notes that many incidents occur during programming, setup, maintenance, and testing.
- Measure the current process before modelling the automated one.
- Price a low-cost redesign as a genuine competing option.
- Include commissioning downtime in the payback calculation.
- Budget year two: re-teaching, wear items, and retraining.
- Name the owner who restarts the system after a fault, per shift.
Sequencing Matters More Than Selection
The most common failure is not choosing the wrong robot; it is automating in the wrong order. A plant that automates its most visible bottleneck before stabilising the process feeding it has bought a faster way to produce the same problem.
The defensible sequence is to stabilise, then measure, then automate the step that is repetitive and documented, then reassess. Each completed step makes the next one cheaper and easier to justify.
Compare warehouse and distribution work separately in the warehouse automation guide, since throughput, traffic, and exceptions behave differently from a production cell.
Bottom Line
Robotic automation deserves the project only after the current process is measured and a low-cost redesign has been priced honestly. Sequence the work, and budget the complete system rather than the arm.
Price the low-cost redesign as a real competing option before approving any automation quote.
FAQs
When is robotic automation not worth it?
When the process is undocumented or unstable, when a low-cost redesign achieves most of the benefit, or when the task will never change and a purpose-built machine would be cheaper and faster.
What belongs in an automation business case?
End effectors, fixtures, integration, controls, safety assessment, training, spares, maintenance, and commissioning downtime, plus a second-year line for re-teaching and retraining.
Should a plant automate its biggest bottleneck first?
Not automatically. Automating a bottleneck fed by an unstable upstream process usually produces the same problem faster. Stabilise and measure first, then automate the documented repetitive step.
Is a cobot the safe option for a first automation project?
Safety is assessed per application, including the tool and the part. A collaborative arm carrying a sharp or hot end effector can present more risk than a guarded conventional cell.