Start With Risk, Not Return
"What is the ROI on monitoring?" is the wrong first question, because it frames monitoring as a productive investment that generates output. Monitoring does not generate output. It reduces the probability and magnitude of failures that destroy output. The correct financial frame is risk reduction: what is the expected cost of unmonitored failures, and how much does monitoring reduce that cost?
Insurance is the useful analogy. Nobody asks "what is the ROI on business insurance?" - the answer is negative in the years you do not claim, and strongly positive in the year you do. Monitoring is the same structure: small ongoing cost, large infrequent benefit. The calculation is about expected value across scenarios, not simple payback.
What One Unplanned Stoppage Actually Costs
Take a modest manufacturing plant - 40 employees, a production line running one shift, making product worth R300,000 per day of production. An unplanned stoppage of 8 hours costs, at minimum: production loss of R37,500, labour still on clock (40 people × 8 hours × average loaded cost), maintenance contractor callout (R3,000–R8,000 for after-hours response), emergency parts at premium (30–50% markup over standard pricing), and post-incident overtime to recover the schedule.
A reasonable estimate for a single 8-hour unplanned stoppage in this size facility: R60,000–R150,000 depending on product, equipment, and time of day. A stoppage that escalates to physical equipment damage - bearing seized, hydraulic component failure, motor burnout - doubles or triples that estimate with replacement parts cost.
How often does an unmonitored facility of this size have an unplanned stoppage per year due to equipment condition? Industry data suggests once to four times per year for mechanical equipment without condition monitoring. The expected annual cost is R120,000–R600,000 just from stoppages that monitoring can prevent.
The iMining Case: What Was Prevented
The pressure monitoring system deployed at iMining (Pty) Ltd detected three pressure anomalies over a six-month period that would not have been visible to the previous inspection regime (manual checks every four hours). Each anomaly triggered an alert; each was investigated and resolved before it progressed to equipment failure.
The system cost - hardware, firmware development, dashboard, deployment - was recovered fully on the first prevented incident. The second and third anomalies represent pure risk reduction at zero additional cost. The system continues operating at 99.2% uptime.
The monitoring system cost is recovered with the first prevented downtime incident. Ongoing monitoring from that point forward is pure risk reduction at the marginal cost of power and connectivity.
Break-Even Analysis
A comprehensive IIoT monitoring system for a 40-person facility - covering pressure, temperature, and motor current across 15–20 monitoring points - typically costs R35,000–R80,000 for hardware and deployment. Annual operating cost (connectivity, power, maintenance) is R8,000–R15,000.
Break-even against a R100,000 prevented stoppage: the system pays for itself with the first incident it catches. Against a R50,000 incident: the first catch recovers roughly half the capital, the second recovers the rest. Year two onwards the system costs R8,000–R15,000 per year to operate against expected incident savings of R100,000–R400,000 per year - a strongly positive expected return.
The break-even question is not "will we recover the cost?" - that almost always happens within the first year. The question is "how quickly?" and the answer is usually "faster than expected."
Secondary Value That Does Not Show in the Primary Calculation
Condition monitoring generates continuous data that has value beyond preventing the incidents it was built to prevent. Maintenance intervals based on actual usage hours (motor run-time from current monitoring) replace calendar-based service schedules - the motor that runs 30% lighter than expected can go longer between services; the one running harder than designed needs earlier attention.
Temperature and environmental data creates automatic compliance records for HACCP audits, ISO processes, or contractual customer requirements - replacing manual log books with timestamped digital records that exist whether or not someone remembered to check.
Energy cost visibility from current monitoring often reveals equipment running outside its efficient operating band - a pump with a partially blocked impeller drawing 15% more current than its rated load, accumulating cost daily and degrading on a trajectory toward failure.
Presenting to Procurement
Procurement committees respond to numbers, not narratives. Three numbers work well: the estimated cost of one unplanned stoppage at this facility (use real numbers from a recent incident if one exists, or conservative estimates based on facility size and production value), the annualised expected frequency based on equipment age and maintenance history, and the monitoring system cost.
Frame the monitoring budget as a line item in the maintenance or risk budget, not the capital expenditure budget. Maintenance teams already understand that preventing failure is cheaper than repairing it - monitoring is the instrumentation that enables the prevention.
The hardest audience is a procurement committee that has never experienced a significant unplanned stoppage. The most effective approach with this audience: ask the maintenance manager or operations lead to estimate what the last major breakdown cost, then anchor the monitoring investment against that number.