Cashflow planning across multiple sites without going crazy
Three sites, three cash positions, one company account. A practical planning rhythm for Uzbek directors managing more than one project at a time.
Running one construction site is a cashflow problem. Running three is a different problem — it is a cashflow problem where the solution to one site can accidentally starve another. Uzbek directors managing multiple projects almost always discover this the hard way: the big Friday payment from site A goes to the urgent supplier of site B, and on Monday site C cannot pay the brigade. Here is a planning rhythm that stops that pattern.
Rule zero — one ledger per site, one ledger for the company
The foundation is this: every site has its own cash position (projected inflow, projected outflow, running balance), and the company has one overlay ledger that shows the net across all sites plus the company-level flows (rent, salaries, taxes). Three site ledgers + one company ledger. Nothing else.
Most firms we talk to have zero ledgers — they have memory and a bank balance. Memory is not a ledger.
The 13-week rolling forecast
For each site, forecast cash 13 weeks out. Why 13 and not 4? Because construction inflows lag — a milestone payment from a client at week 6 is real money, but brigade payments at week 2 are more real. A 4-week view sees the pain; it does not see the relief. Thirteen weeks sees both.
The 13-week forecast is updated every Monday. It has three rows:
- Expected inflow by week — client milestone payments, approved invoices, known refunds.
- Expected outflow by week — supplier POs due, brigade payroll, subcontractor milestones, known fees.
- Running balance — start-of-week cash plus inflow minus outflow.
If the running balance goes negative in any week within the 13, that site has a problem. The director has 6–10 weeks to act, which is usually enough.
The company overlay
On top of the three site forecasts, one company-level view that does two things:
Identifies cross-site rescue opportunities. Site A has surplus cash in weeks 5–7, site B has deficit in weeks 6–8, and the transfer is a one-line approval. Without the overlay, this transfer happens in panic on week 8. With the overlay, it is planned in week 2.
Protects company-level obligations. Tax payments, salaries, rent — these cannot be delayed, and they cannot be moved between sites. The overlay keeps a weekly reserve for these and shows which site contributions fund them.
The Monday cashflow meeting
One hour every Monday morning. Director, finance lead, and the three site prorabs join by video or in person. Agenda:
- Update the week just closed. What actually came in, what actually went out. This is where the forecast learns.
- Walk through the next 4 weeks. Week by week, site by site. Any red cells get flagged and someone gets owner for the fix.
- Review weeks 5–13 at a higher level. Trends, large upcoming milestones, supplier renegotiations worth prioritizing.
One hour is plenty if the ledgers were maintained during the week. If the meeting runs over, the meeting is not the problem — the ledgers are.
The three cross-site failure modes
The rescue spiral. Site A is fine, rescues site B, now site A is not fine, and the rescue has just moved the problem. The overlay forecast prevents this by showing that site A's surplus is temporary — it gets consumed by weeks 9–11, at which point site B's rescue becomes a company-level crisis.
The invisible milestone gap. A client's scheduled payment is not actually scheduled — it is a hope. "They usually pay in the middle of the month." Without an actual date in the forecast, the hope becomes a number, and the number becomes a plan. Confirmed payment dates should be boldfaced; hoped-for dates should be visually different.
The subcontractor bunching. Three subcontractors have milestones falling in the same week. Individually each is fine; together, it is a cash cliff. Overlay forecasting catches this; single-site forecasting does not.
Why Excel is enough, until it is not
You can run this rhythm in Excel for six months. Three tabs per site (inflow, outflow, balance) and one company tab. It will work. What breaks first is two things: manual transfers of data between sites and the company tab, and the time it takes to actually enter purchase orders and invoices into three separate ledgers.
At that point — usually when the third site comes online or the company passes ~15 billion soum in annual turnover — you move to software. Not before. Excel is underrated; it is also exactly where most Uzbek construction firms should start.
The outcome after 90 days
Three visible changes. First, fewer emergency transfers between sites. Second, the director's Friday stops being a cash stress test. Third, supplier relationships improve because the firm can commit to payment dates it actually hits. That third effect pays for the discipline on its own.