Every site does itslightly differently.
Multi-site operations rarely have one process. They have a process per location, each locally sensible, none of them written down — and head office reconciling the difference by hand.
Open source projects
Public on github.com/JIGGAI
AI plugins
Published on npm, free to install
AI plugin installs
65,377 to date, per npm
AI monthly plugin installs
2,811 in August 2026, per npm
What we hear in the first hour.
Not a generic list. These are the ones that come up in businesses shaped like yours.
Head office spends two days a month turning site reports into one report.
The reconciliation is the process. It is usually the single most automatable job in the business and nobody owns it.
Every site swears they follow the same procedure. The numbers disagree.
They do follow it — with local workarounds for local constraints. The workarounds are the interesting part, and only a visit finds them.
Rolling anything out means visiting each site to explain it again.
Change cost scales with sites, which is why improvements stall at whichever location is furthest from head office.
One site is brilliant at something and nobody knows why.
Somebody there built a better process. It is worth more than most software, and it is currently trapped in one building.
We cannot tell which location is actually the problem.
Because the data arrives already averaged. Per-site visibility is usually a smaller job than people expect.
The work is never inside a site. It is in the seams between them.
Which is why we walk more than one of them. Three days on site buys the comparison a single visit can only guess at, and the three that follow are where the differences between sites turn into a reason rather than a list.
The software already in the building.
We do not arrive with a platform to sell you. We arrive expecting these, and the work is usually in the gaps between them.
The ERP nobody fully trusts
NetSuite, Epicor, Sage, Dynamics. Usually correct and usually bypassed for anything urgent.
Field or job management
ServiceTitan, Procore, Jobber, or a home-grown scheduler. Where the real day is planned.
The per-site spreadsheets
One per location, each subtly different, each maintained by one person. The de facto source of truth.
Rostering and payroll
Deputy, When I Work, ADP. Where headcount and hours actually reconcile, or fail to.
The group chats
WhatsApp or Teams per site. Where exceptions are handled, and therefore where the undocumented process lives.
Head office reporting
Power BI, Looker, or a monthly deck. Downstream of everything above, and only as good as the reconciliation feeding it.
Collapsing the monthly reconciliation.
Roughly what one entry in your report looks like — a real shape, with the numbers changed.
- What we saw
- Every site submitting a spreadsheet in its own format on its own schedule, then one person at head office normalising them into a single report before anyone could act on it.
- What it costs today
- A senior operations person spending part of every month on it, plus a reporting lag that left the figures stale by the time anyone made a decision from them.
- What we would build
- An agent that ingests each site submission in whatever format it arrives, normalises it against a defined schema, and flags the rows it is unsure about for a human rather than guessing. The person who did the reconciliation reviews exceptions instead of retyping everything.
- How you would know it worked
- Time spent on reconciliation, and the lag between period close and the report being circulated. If the exception rate stays high, the schema is wrong and we fix that rather than the tool.
The judgement about why a site is off. A variance is a conversation with a manager who knows their building, and an automated explanation would be a confident guess dressed as an answer. We automate producing the number, never interpreting it.
What coordination work costs you.
Three numbers you already half-know. Move them until they look like your business.
re-keying, chasing status, producing the same document again
on that work specifically, not their whole job
salary, tax, benefits, desk
A 75% capture rate. The other twenty-five per cent is judgement, exceptions, and not wanting to look greedy.
This is arithmetic, not a finding. It rests on three numbers you guessed. The assessment replaces all three with numbers we observed — and tells you which of those hours are actually worth automating.
See what a real finding looks like →What the six days look like.
The most common question we get is not about AI. It is what these people will actually do in my building.
- Day 1Walk the floorWhoever is on shift
We start where the work happens, not in a meeting room. Nobody prepares anything, and the first day is mostly watching.
- Day 2Sit with the people doing itOps, admin, finance
Conversations with the roles that touch the work most, and a real task followed end to end — including the parts that happen in a group chat.
- Day 3Systems, then a read-backWhoever holds the logins
What you run, what talks to what, and where a person is currently the integration. We tell you what we saw before we leave, while it is still cheap to correct.
- Days 4–6Research and discoveryOur desks, not yours
Away from your building. We cost the work we watched, model the alternatives, and test the shortlist against your own numbers rather than a framework.
- +1 weekOne recommendationPresented in person
The single change worth making first, specified precisely enough to build — with the ranked analysis behind it and the list of what we would not automate.
The questions this raises.
How many sites do you need to visit?
Two or three, chosen to differ — your best-run location and your most awkward one tell us more than a representative average would. That is what the three days on site are spent on.
Our sites are in different regions. Does that change things?
It changes travel, not method. If the sites are far enough apart to make visiting several impractical, we say so on the call and scope it differently rather than quietly sampling one and generalising.
Will this mean forcing every site onto one process?
Not necessarily, and often it should not. Some local variation is a sensible response to a local constraint. The report distinguishes between variation that costs you money and variation that is load-bearing.
Do you need access to our systems?
We need someone who can show us, not credentials of our own. We take an inventory of what runs where; we do not need admin access and we do not copy your data out of the building.
Start with half an hour.
We'll tell you whether an assessment would pay for itself in an operation like yours. Sometimes the answer is no.