Sequencing Around a Forecast You Do Not Trust

You cannot programme around British weather in the sense of predicting it. You can programme around it in the sense of never having only one thing your crew can do on a given day. That distinction is the whole of it, and it is a sequencing decision made weeks earlier, not a decision made at seven in the morning in a wet van.

The cost is the cascade, not the day

A rained-off day on its own is a day. What makes it expensive is what it does to the trades behind it. The plasterer moves to next week, but next week they are on someone else's job, so it is the week after; the decorator was booked around the plasterer; the handover date was booked around the decorator. One day becomes a fortnight through nobody's fault.

Which is why the useful question about a wet day is never "what did today cost". It is "what is the first thing downstream that cannot move".

Sequencing, in practice

The trades that stop are the obvious ones, roofing, external render, brickwork below a certain temperature, groundworks in standing water, anything craned. The trades that quietly stop are the ones people forget:

• Anything with a drying or curing time. A screed does not care that it stopped raining; it cares about humidity and temperature for as long as it takes. Follow-on finishes have to be programmed off the drying time, not off the pour.
• Anything that needs the building watertight. Plasterboard delivered into an open shell is a delivery you will do twice.
• Anything requiring a crane or a MEWP, which has a wind limit well below the point at which the rest of the site stops.

The counterpart is the list of work that can be pulled forward into a wet day: internal first fix, ceilings, joinery, anything in a completed and roofed area. A programme with no such work banked has nothing to offer on a Wednesday in November.

What a forecast is actually for

A fortnight's outlook is not a schedule. It is a reason to ask a question two weeks before the question becomes urgent, is the roof on before the wet spell, or after it. Confidence falls off sharply with distance, so treat the near days as a plan and the far days as a prompt.

Two weeks out, you have time to bring a delivery forward, book a trade a week earlier, or tell a client the handover is at risk while it is still information rather than an apology.

Tell people the moment it moves

The failure that costs the most is not the weather. It is the two days between the decision to move a trade and the trade finding out. In that gap they take other work, and the slot you protected is gone.

Whoever is affected by a move should be told the day it is made, and told what their new date is rather than that their old one is cancelled. A subbie who is told "Tuesday is off" and nothing else has to ring you to find out anything useful, which is a phone call you have now created for yourself.

What the contract gives you back

Standard forms treat exceptionally adverse weather as a relevant event for time, not usually as a matter giving an entitlement to money. Read that sentence twice, because it is the one that surprises people: an extension of time protects you from damages for late completion. It does not, on its own, pay for the prelims you burned while the site was idle.

Two consequences. First, the notice provisions matter; entitlement can be lost by not telling anybody in the way and within the time the contract requires. Second, "exceptionally" is doing real work in that phrase, rain in a British February is not exceptional, and the argument is generally made against records of what is normal for that place at that time of year. Which means the site records you keep during the wet week are the claim. Check your own contract, and take advice before relying on any of this.

Where BuiltUp fits

The scheduler builds a Gantt from the priced scope, sequences the trades in dependency order, and avoids weekends and bank holidays. It checks the forecast for the site postcode. The Met Office in the UK, and flags the days that carry weather risk against outdoor work. On Pro and above, a background check runs against every live programme and proposes a move when the forecast turns, a delivery slips or an upstream trade has not finished; you accept or reject it, and accepting notifies the subbies affected.

It proposes. It does not decide, and it does not know that the client's access is only Tuesdays. Scheduling is available on Pro and Ultimate.

The short version

Bank some indoor work for the bad days. Programme off drying times rather than off finish dates. Look a fortnight ahead and move the big outdoor items rather than the small ones. Tell people the day it changes. Keep the site records, because the records are the extension of time.