Solar for businesses and institutions in Kenya: sizing, savings and payback by sector

A barbershop, a boarding school and a factory use power at very different times, and that timing, more than the bill, decides how much solar can save. This guide explains the method once, compares seven common types of building, and then gives each one only what is specific to it: its load pattern, a worked example and the mistake that most often spoils the case.

The method, explained once

1. Find the daytime load. Solar only saves money when its output is used, stored or credited. Start from twelve months of bills, then estimate the share used in daylight. A bill is an average that hides busy and empty days, so opening hours and timetables matter more than the monthly total.
2. Know your tariff. The value of each solar unit is the rate it replaces. Small commercial customers on SC3 pay about KES 28.82 per kWh all-in in the daytime and KES 17.32 at night. Commercial and industrial customers on CI1 pay about KES 22.09 in the daytime and KES 13.96 at night, plus a monthly demand charge. Off-grid sites running a generator pay far more: at the EPRA Nairobi diesel price of KES 227.76 per litre and about 0.45 litres per kWh, fuel alone costs about KES 100 per kWh. See what a KPLC unit really costs and how to read your KPLC bill.
3. Size for the load that is always there. On Kenya's average sun, each kW of panels produces about 3.83 kWh a day, falling to about 2.70 in July. Divide the daytime load you can count on by 3.83 for a first array size, then by 620 W for the panel count. Where reliability matters more than savings, size for July instead.
4. Check payback on the savings per kW. Payback is the installed price divided by the yearly saving, after maintenance. The key test is the savings per kW installed: each extra panel is worth less than the one before once its output starts to go unused. Our solar payback article covers panel ageing and inverter replacement, and comparing solar quotes shows how to check prices.

The seven sectors compared

Installed prices vary widely, so payback for most sectors depends on your quotes. The factory case shows the full calculation.

Sector by sector

Shops, salons and barbershops

Load profile. Open from about eight to seven, so most use falls in daylight. A barbershop or salon uses about 85% of its power while the sun is up; a minimart with fridges only about 55%, because the cold chain runs all night.

Main pitfall: an inverter too small for the dryers. Hair dryers, flat irons and steamers draw 1 to 2 kW each, so the inverter must run two or three at once, typically 3 to 5 kW, even when the array is small. For outages, about 3.3 kWh of lithium battery keeps a 1 kW essential load (lights, clippers, router, till and one dryer) running for 3 hours. If you rent, get the landlord's written permission first.

Hotels and lodges

Load profile. Energy around the clock, and busy and quiet seasons. A 30-room lodge on CI1 uses about 9,000 kWh a month in high season, about 296 kWh a day, of which hot water is about 35%.

Worked example. Take hot water off the bill first: a solar water heating system covering 75% of the 104 kWh a day of water heating avoids about 28,000 kWh a year, worth about KES 550,000 a year at the CI1 average of about KES 19.40 per kWh. Then size the panels for low season:

Main pitfall: sizing for the busiest month. The bigger array saves more in total, but each extra kW earns less because its output has nowhere to go in low season. Start near the low-season size, shift laundry and pool pumps into sunny hours, and consider a battery for the 18:00 to 22:00 dining peak, which on CI1 is still charged at the daytime rate. See time-of-use load shifting. Off-grid camps on diesel have a far stronger case; see diesel generator vs solar hybrid.

Health clinics

Load profile. Here power is about reliability first: a vaccine fridge that must never warm up, an oxygen concentrator and light for a delivery at 2 a.m. A typical rural dispensary uses about 7.2 kWh a day, about 2.4 kWh of it at night.

Worked example. Allowing one cloudy day of the vaccine fridge, security lighting and two hours of the oxygen concentrator on top of the night load gives about 5 kWh of usable storage, or 5.6 kWh of lithium at 90% depth of discharge. With about 8% storage losses the panels must supply about 7.4 kWh a day. Sized for July at 2.70 kWh per kW, that is about 2.7 kW, so five 620 W panels (3.1 kW), producing about 8.4 kWh a day in July and 11.9 kWh in a typical month. A 3 kW inverter with a good surge rating starts the water pump.

Main pitfall: sizing for the average day. A clinic that only works in an average month fails when it matters. Put the vaccine fridge, concentrator, delivery room lights and security lights on an essential circuit and agree a maintenance contract with response times. Where available, a WHO-prequalified solar direct drive vaccine fridge takes the most critical load off the battery entirely. The battery backup and storage guide covers autonomy in more depth, and the solar water pumping guide covers the borehole.

Schools

Load profile. Full term days, quiet weekends and holidays. A day school using 1,500 kWh a month, 80% in daylight, averages 40 kWh of daytime use a day on paper, but in practice uses about 70 kWh on about 190 busy days and only 10 kWh on the rest.

Worked example. A 10.5 kW array (17 panels) sized from the bill average produces about 14,700 kWh a year. The bill average says 92% of that is used, saving about KES 390,000 a year. Following the timetable, only 64% is used, saving about KES 271,000: the plain calculation overstates savings by more than 40%. A smaller 6.8 kW array of 11 panels uses about 70% of its output and saves about KES 28,000 a year per kW, against about KES 26,000 for the larger one.

Main pitfall: trusting the monthly average. Move flexible loads, such as pumping the borehole and running freezers, into weekends and holidays, and size any battery for evening prep, not for holidays. Boarding schools differ: with half their use in daylight, a 4,000 kWh a month boarding school might need a 21 kW array plus about 13 kWh of battery. StimaPulse prepares a school decision pack for boards and bursars.

Churches and mosques

Load profile. One busy day a week. The example uses about 50 kWh of daytime power on the main day and about 6 kWh on each other day, about 86 kWh a week.

Main pitfall: sizing for the main service. Doubling the panels adds only about KES 10,000 a year, and a battery cannot store a week of surplus for one day. Move pumping and laundry to weekdays, share the panels with a school, clinic or offices on the same compound, and size any battery for the length of the service as backup.

Apartment blocks (common areas)

Load profile. The common meter runs booster and borehole pumps, a lift, corridor and security lights, the gate and CCTV. A 24-unit block on SC3 uses about 1,500 kWh a month, only about 45% of it in daylight as most blocks run today.

Worked example. As the block runs today, about 22 kWh of daytime use supports 10 panels (6.2 kW), saving about KES 223,000 a year, about KES 770 per unit each month. Pumping water to the roof tanks during the day raises daytime use to about 30 kWh, supporting 13 panels (8.1 kW) and saving about KES 295,000, about KES 1,020 per unit. That free change alone adds about KES 70,000 a year.

Main pitfall: trying to solar the night lighting. Lighting is about 35% of the load and mostly at night. Fit LEDs and motion sensors first, which costs far less than a battery. Keep the lift on the grid and back up pumps, security lights, gate and CCTV on an essential circuit; see on-grid, off-grid or hybrid inverters.

Factories: the full worked case

Factories have the largest cases and the easiest to get wrong: the bill has three parts, energy, demand and power factor.

Step 1: fix the power factor. A 250 kVA peak at 0.85 is about 212 kW of real power. A capacitor bank lifting the power factor to 0.95 brings it to about 224 kVA, saving about 26 kVA x KES 1,100 = about KES 29,000 a month, or KES 347,000 a year. This is usually the cheapest kVA a factory can buy, so it comes before solar. See KPLC demand charges.

Step 2: size solar for the daytime base load. About 1,380 kWh a day falls in daylight. A 250 kW array (about 400 panels) sits below that and produces about 349,000 kWh a year. On the rest day most output has nowhere to go without export, so plan on using six days in seven: about 299,000 kWh, worth about KES 6.6 million a year at the CI1 day rate.

Step 3: trim the peaks solar misses. The monthly peak often comes at shift start or on a cloudy afternoon, so solar rarely lowers demand on its own. A battery supplying about 40 kW at peak moments cuts about 40 kVA, saving KES 44,000 a month or about KES 528,000 a year, provided its controller catches every peak. Staggering large motor starts achieves part of this for free.

Step 4: shift what solar cannot cover. Batch mixing, pumping to storage or pre-cooling can move into the 22:00 to 06:00 window, where energy costs about KES 8 less per kWh.

At an example installed price of KES 75,000 per kW (KES 18.75 million) and 1% a year for maintenance, the solar alone pays back in about three years. Before committing, log the load for at least two weeks and check the roof can take roughly 1,100 m² of panels.

Next steps

• Start with the free solar calculator, using your bill and a realistic daytime share.
• Compare ways to pay, including cash, loans and power purchase agreements, in paying for solar.
• Choose storage with lithium vs lead-acid battery cost and the battery backup guide.
• For farms and agribusiness, see solar for farms in Kenya.
• The commercial and industrial solar guide explains how StimaPulse models energy, demand and seasonal use together, and our solutions page covers decision packs for institutions.
• With your bills and load data, request an assessment for a business case your board or finance team can check.