Background
JESA operates food and agricultural processing facilities in Uganda. Like many commercial facilities, JESA was exploring solar energy as a way to reduce electricity costs and improve operational resilience against grid instability. The company had received a solar proposal and was preparing to move forward with the investment.
At the same time, management was aware that the facility's energy costs were higher than expected relative to production output — but without granular data, identifying the source of inefficiency was difficult.
The solar proposal on the table was built on estimated consumption — figures no one had independently measured. JESA was preparing to commit tens of millions on the strength of an assumption.
What We Did
Orijtech Energy deployed nineteen three-phase smart meters across the Busunju factory in a progressive programme running from May 2023 to February 2026. The meters cover the New Plant and Old Plant transformers and the major loads behind them: raw-milk, UHT and yoghurt processing; the ammonia chiller; boilers 1 and 2; the New Plant and Old Plant compressors; the Old Plant chillers; the effluent-treatment transformer; and the water-pumps transformer serving twelve submersible and two booster pumps. Each meter carries three current sensors, one per phase. The data was streamed to the Orijtech Energy dashboard, where load profiles were analysed by time of day, day of week, and process stage.
What the Data Showed
On the solar investment: The consumption data revealed that the actual load profile was materially different from the estimates that had been used to size the proposed solar system. Estimated consumption had been higher than actual consumption at the times when solar generation would be available. The proposed system was oversized for JESA's real energy needs.
Right-sizing the solar system based on verified metering data reduced the required investment by UGX 300 million — while still meeting the facility's actual daytime load requirements.
On raw milk processing: The metering revealed that the energy cost of processing raw milk varied significantly depending on when in the day the process ran. Running processing during peak tariff hours was substantially more expensive than running the same process during off-peak hours. The underlying process was unchanged — only the scheduling.
Outcomes
-
The solar system was re-specified using actual load data. The investment dropped by UGX 300 million, and the return-on-investment timeline improved accordingly. JESA proceeded with a correctly sized system rather than an oversized one that would have taken years longer to pay back.
-
Milk processing was rescheduled to off-peak tariff windows. This operational change — requiring no capital expenditure — delivers UGX 40 million in savings every year.
Total first-year value: UGX 340 million, with UGX 40 million recurring annually thereafter.
The point is not that solar was the wrong decision — it was the right one. It is that an investment is only as sound as the data beneath it. Measured consumption turned a decision made on estimates into one made on evidence, and surfaced a recurring saving no one had thought to look for.
"The monitoring showed us that our solar proposal was built on the wrong assumptions. The right data turned a good investment into a great one — and identified savings we hadn't even been looking for."