An electrical and generator service company · About $22M revenue; 3 branches
A Novare engagement. The client's name is withheld.
Why don’t our numbers match until the 20th?
The answer 4 systems, copied and checked by hand. We replaced the spreadsheets with a tool built in 6 weeks.
- 98 hrsof manual reporting saved a month
- 20th → 5thday the books close
- 8 weeksearly warning on a cash dip
A Novare engagement. The client's name is withheld.
The question
Every month, the controller spent about 110 hours exporting, pasting and checking. The numbers were not final until around the 20th, and by then the month was old news. Cash planning was a guess.
- Jobs, payroll, inventory and the books lived in four places.
- One broken formula could change the whole month.
- Nobody could say where cash would be in two months.
What we looked at
We timed every step of the month-end process. Most of the 110 hours went to moving data, not reading it.
- Over a third of the time was copying and pasting exports.
- Another third was checking that the copies matched.
- The rest was building and fixing the owner’s report.
Labels are generic to keep the client anonymous.
What we built
Connect the systems
Jobs, payroll, inventory and the books pulled automatically, every night.
Checks before publishing
Every number ties to the books. If one doesn’t, the report waits.
The owner’s report, built in
Same layout every month, ready by the 5th.
A 13-week cash forecast
Built on the same data and updated every week.
- RevenueJobs system ties to the booksPass
- PayrollHours tie to payroll registerPass
- InventoryWarehouse count vs. booksCheck $4,100
- Owner packHeld until all checks passReady
The answer and the money
Manual work fell from about 110 hours a month to about 12. The books close by the 5th. The first cash forecast showed a dip below the company’s minimum in week 8, driven by a seasonal generator inventory build. Collections were pulled forward and a truck purchase moved a quarter, so cash never crossed the line.
| Before | After | Change | |
|---|---|---|---|
| Manual reporting hours a month | About 110 | About 12 | 98 hours saved |
| Day the books close | Around the 20th | By the 5th | About 15 days sooner |
| Cash visibility | A guess | 13-week forecast, updated weekly | 8 weeks’ early warning on a cash dip |
Labels are generic to keep the client anonymous.
The controller now spends month-end reading the numbers instead of building them.
The work behind this story
Systems involved: The jobs system, payroll, inventory and the books.
A Novare engagement. The client's name is withheld. Results vary by business. One company's results are not a forecast for another.
Running into a version of this?
Tell us what you’re seeing. We’ll tell you how we’d find the answer and what it would take.