Why one platform, not three companies
There are three obvious ways to build a solar company in Kenya. We chose the one that is harder up front and easier forever after.
There are three obvious ways to build a solar company in Kenya.
You can sell to one segment, say households only, and ignore the rest.
You can sell to many segments, but build separate operations for each: separate teams, separate billing, separate inventory.
Or you can sell to many segments and run a single operation underneath them all.
We chose the third.
The first option leaves money on the table and the customer underserved. A household that grows out of a 60 W system today will look elsewhere when it needs 800 W tomorrow.
The second option is what most companies do by default. It is also where margin disappears. Three teams means three sets of overhead. Three billing flows mean three places to lose customers. Three inventories mean three ways to be out of stock when it matters.
The third option is harder up front and easier forever after.
What "one operation" actually means
One service number. A customer with Spark, Current, or Grid calls the same line. The same dispatcher routes the same engineer.
One firmware. We ship updates to every product on the same release cadence. A bug in the battery monitor on Spark gets the same fix as on Grid, the same week.
One warehouse. Replacement panels, batteries, and controllers go to one place. Field engineers in Eldoret pull from the same stock as the team in Nairobi.
One telemetry layer. Every system reports state of health to the same store. We see a Grid in Kisumu and a Spark in Mombasa on the same dashboard.
One payment story. Cash, M-Pesa, or twelve-month instalments. The same accounts team handles all three.
Why this matters to the customer
Cost is one part of why this works. Operational coherence is the bigger part.
When the question "has anyone seen this before?" has a single answer, faults get fixed faster.
Customers feel it the most when something goes wrong. The right engineer arrives with the right part, because the system told us what was needed before they got in the van.
This is not novel. It is just disciplined.
It is also what makes us different from the solar company that sold and forgot.
More from the team
Operations
Five lessons from year one in distributed solar
A year and four thousand systems in, here are the lessons that still hold. Telemetry, dispatch, and the price of cheap.
Engineering
Inside the PreciSense edge module
Every Precifarm system has a small board on the charge controller. Here is what it does, why it costs twelve dollars, and what we got wrong.
