One month-end close, run start to finish.
What you approve each month looks exactly like this.
Monthly processes
In run order. Open a row for its files, what it does, and what it made.
How this works
The short version
Five processes, one month, and five is just where the demo stops. A school's engine may carry as many workflows as its month-end needs. Every row starts from the raw files your team already pulls, runs the month-end the way your school runs it, and automates the manual work in between. What comes back is the finished task, drafted and stopped for your review.
Why this exists
Every month a school's finance team closes the books. Same work, same order, on a deadline, and it has to be right, because an auditor, an authorizer and a board all read what comes out of it.
- Code the card charges
- Reconcile the bank
- Post the entries the month always needs
- Get each budget holder their numbers
- Report to the board
Almost all of it is done by hand, and the information needed is scattered: some in the accounting system, some on a card statement, some in a purchase order log, some in an export one person knows how to run. So the same picture gets rebuilt from nothing, every month.
This is the other way. You walk us through how your close actually works, decision by decision, and we build that into software that runs it. The repetitive part stops being anyone's job, and the numbers come back the same way every month instead of depending on who did them and how late it was.
What the statuses mean
- Ready. Everything it needs is in the folder.
- Running. It is working. The result appears in the row when it finishes.
- Run complete. It finished and left nothing for a person to decide.
- Needs your review. It finished and left items for a person to decide. That is the system working, not failing.
Drafts only. Nothing posts to QuickBooks, nothing emails anyone, and nothing moves money. Each process reads its files, writes drafts, and stops.
Sample school, real engine. Every file these processes hand back was written by the same engine a client gets, run once on anonymized sample data. Run replays that run. Your own files are not processed on this page.
The card process on this board is the one we measured at a real school: 10 to 12 hours by hand down to about 30 minutes of review. One process, one real school. Not a projection.
If it repeats, it can run here
The same engine runs the rest of a school's month.
The close is the product.
Every charter school finance team runs the same loop: statements come in, reconciliations wait, card charges need receipts, entries need posting, the board report needs building, and somewhere in there the actual month happens. When the team is two people, or one, the loop wins.
The Command Center takes that loop and automates the manual work inside it. It starts from the raw files your team already begins each task with, and the automations draft the reconciliations, code the card charges, prepare the recurring entries, and write first-draft variance explanations from transaction detail instead of memory. Everything lands in one place with a single review queue. Your team approves what needs a human and owns the close, with a sign-off log behind every number.
No new systems. No credentials handed over. It runs on the files your school already produces from QuickBooks, Bill.com, your bank, and your card program.
Files in. Engine runs. Report out.
Your files in
Bank statements, card statements, QuickBooks exports. The same files your team already pulls, dropped in a shared folder. About 15 minutes.
Your engine runs
Built once for your school: your chart of accounts, your coding rules, your formats. It matches, codes, and drafts everything that repeats. Nobody learns new software.
Your outputs land
Reconciliations, coded card charges, journal entries with support, variance drafts, the board report. Everything lands in one review queue for your approval.
Nothing changes for your team. Same files, same systems. The work just stops taking the month.
The close is the start.
Program analysis, recurring reports, anything that turns data into a deliverable can be added to the same engine. If it repeats, it can run here.
Start with the Map. Keep the manual either way.
The Messy Middle Map
We map your close end to end and deliver your Close Operating Manual: every task, owner, source file, and deadline. Valuable even if we stop here, because your close finally exists outside one person's head. One piece of it comes back automated before we finish.
The Map fee credits toward the build if you sign within 60 days.
The build
We build the matching, the coding, and the report drafting on your real data, then run one full close in parallel with your current process to prove it ties. The range moves with how many workflows you want running.
Monthly automation maintenance
We keep the automations running. When a bank changes an export format, a platform changes how it generates a report, or anything in the code breaks, we fix it before it reaches your close. Changes on your side are covered the same way: a new grant, a new program, a new campus, or a restructured chart of accounts, and we update the automations to match. Your team works a short review queue and signs off on what goes to the board, with a log behind every number. Like the build, the number moves with workflow count, not school size: a school running two or three automations sits near the bottom of the range.
Entries never post without a named approval, and drawdowns, board communication, and accounting judgment calls stay human. It goes in the agreement, in writing.
FY27 just started. Your first close of the year sets the tone for all twelve.
Working with schools today: monthly program cost analysis, school-leader budgets, and automated financial workflows for a 3-school charter network.
Send this sample close to your inbox
Get the link to share with your board chair or walk through later. We follow up once, like a person, not a drip campaign.