Knowledge · Weather and climate
Weather is the one variable
that does not care how good you are.
Cost, labour and policy affect the economics of what gets built. Weather decides what physically gets done once a site is running. This hub covers why saturated ground stops a pour, how a lost week becomes a lost quarter, what the 2020 to 2026 wet run on the east coast coincided with in the public data, and what a builder can measure and evidence on their own jobs.
01 / Overview
The variable that does not discriminate
Nearly everything that shapes housing supply operates differently depending on who is running the business. A well-organised builder handles cost pressure, trade relationships, selections and regulatory complexity better than a poorly organised one, and over a year that difference shows up plainly in margin, in build times and in how many jobs get finished. Capability is visible in outcomes, which is the whole premise of running a business well.
Weather is effectively the exception. Outside a government-mandated shutdown, it is the one variable that does not reward capability at the moment it bites. A competent builder with a full order book, good trades and money in the bank still cannot pour a slab on saturated ground, and still cannot frame in sustained rain. The site stops. The best operator in the district and the worst are both standing in the same mud, looking at the same forecast, and neither of them has an advantage that day. That is an operator's argument rather than a published finding, but the mechanism underneath it is not controversial: the ground either carries the work or it does not.
Why it matters
Because a variable that stops everybody equally is still not carried equally. Capability comes back the moment the rain does stop, in whether the builder saw the delay coming, re-sequenced the trades that could still work, recorded the lost days as they happened, and priced the build period with an honest allowance in the first place. The weather is the same for everyone; the recovery, the record and the contract position are not. This hub is about the second half of that sentence, because the first half is not something any builder gets to manage.
02 / Framing
Economics and physical output are different problems
Policy levers, cost structures and labour availability affect the economics of what gets built. They determine whether a project stacks up, whether a client can borrow, whether a lot is worth developing and what a house costs to deliver. Weather determines something else entirely: what physically gets done once a site is already running, funded, approved and staffed. Both problems are real, and treating them as the same problem is how weather ends up filed as an excuse rather than as an input.
The distinction has a practical consequence for a builder. Economic pressure is something a business responds to with pricing, procurement and cost control, and those responses live in the contract model and the estimate. Physical constraint is something a business responds to with sequencing, allowances and evidence, and those responses live in the construction programme and the site record. A builder who applies the commercial toolkit to a physical constraint ends up discounting their way through a wet season, which fixes nothing.
It is also, in practice, a reasonable explanation for why weather has been underweighted in the public discussion of housing supply. Weather does not appear on a policy dashboard and does not respond to a policy lever, so there is no natural place for it in a debate organised around approvals, targets and completions. That is an observation about the shape of the conversation, not evidence that the effect is small.
03 / Mechanism
How a wet fortnight becomes a lost quarter
Seven steps from the rain gauge to the finish date. The important ones are in the middle, where the delay stops being about weather and starts being about other people’s calendars.
- 01
Rain falls on an open site
The rainfall record starts here and, in most arguments about weather, stops here too. What fell is the least useful number on the page, because the same 40mm lands very differently on a free-draining sandy lot and on a clay site with a cut and fill platform.
- 02
The ground takes the water and holds it
Clay and other reactive soils absorb water and lose bearing capacity. A platform that was compacted and level becomes soft, and the pad no longer behaves like the tested surface the engineer signed off. This is where weather stops being a rainfall figure and becomes a ground condition.
- 03
The pour cannot proceed
A slab needs a stable subgrade, a dry membrane bed and a surface a truck and a pump can reach without churning. Saturated ground fails all three. Pouring on it risks differential settlement under the slab, and no concreter or engineer will sign that off to save a week.
- 04
The site stays unusable after the rain stops
This is the part the weather record never shows. Waterlogged ground stays unworkable for days, sometimes weeks, after the last shower. In practice the days lost commonly exceed the days it rained, which is why a bureau printout understates a real delay.
- 05
The trade slot is lost, not deferred
Trades are booked by the week. The frame crew held the first week of August; when the slab slips, that crew fills the slot on another job and cannot come back until September. The delay is not the length of the rain, it is the length of the next available slot.
- 06
Every following trade re-queues behind it
Roofing waits on frame, lock-up waits on roof, fitout waits on lock-up. Each handover is re-negotiated against another builder’s calendar, and a fortnight at slab stage commonly emerges as ten to twelve weeks by the time fitout trades are needed.
- 07
The cost base keeps running the whole time
Supervision, site facilities, fencing, insurances and finance are consumed by duration, not by work done. On a fixed-price contract that consumption is not recoverable through the price, which is why the time question and the money question have to be handled at the same time.
The compounding is the whole point, and it is what separates persistent rain from a bad week. Ground saturated in early July means a slab that cannot be poured until late July. The frame trade booked for the first week of August has filled that slot elsewhere and cannot return until September, so roofing lands in October rather than August. By the time fitout trades are needed, a fortnight has become ten to twelve weeks, and on a fixed-price contract none of it comes back through the price. The sequencing side of this is covered in detail in wet weather delays and trade sequencing, and the mechanics of which delays actually move the finish date sit in critical path and dependencies.
04 / The record
What the east coast has been building through
Six documented markers from 2020 onward, each attributed to its source. Every figure here is point-in-time and should be confirmed against the current publication before it is relied on.
A triple La Nina, then another event
Between 2020 and 2023 eastern Australia ran through a rare triple La Nina, three consecutive years of above-average rainfall, on the Bureau of Meteorology’s ENSO record. A further La Nina formed in late 2024 and ran through early 2026.
Sydney’s wettest year since 1858
Observatory Hill recorded 2,530mm in 2022, 221 per cent of the long-run average and the highest annual total since records began in 1858 (Bureau of Meteorology, Sydney in 2022 Annual Climate Summary).
74 clear days in half a year
Across Sydney in the first six months of 2022, only 74 days were not affected by rainfall, on figures published by Newpoint Advisory in October 2022. That is a single-source figure and worth checking against the bureau record before it is quoted in a claim.
40 per cent of days against a 10 per cent budget
Brett Mason, chief executive of Built, told the Property Council of Australia in October 2022 that the contractor had lost 40 per cent of its construction days to wet weather, against a business that normally budgets for 10 per cent. That is a commercial contractor rather than a detached residential builder, so read the scale as indicative and the direction as the point.
The central bank said it out loud
The Reserve Bank of Australia named wet weather explicitly as a factor constraining construction output in its September 2023 Bulletin on recent trends in Australian productivity. That is a central bank observation, not an industry body making a case for its members.
More than six years without a sustained dry run
The operator observation, not a published statistic: most detached builders across eastern Australia have not had a settled dry stretch since 2019, and on the Mid North Coast and the rainfall-affected parts of NSW and Queensland the disruption has been unusually consistent.
Read together, these are not a claim that the last six years were the wettest on record everywhere. They are a claim that eastern Australia ran an unusually long sequence of above-average rainfall years, that the effect on construction days was large enough for a listed contractor to quantify publicly, and that a central bank thought it worth naming in a productivity Bulletin. The flood dimension of the same period, where remediation work absorbs the trades a region depends on for years after the water goes down, is covered separately in flood recovery and trade availability.
05 / Build times
What happened to the time it takes to build a house
The Australian Bureau of Statistics publishes average time from commencement to completion for new private houses, quarterly and by state, which makes this one of the few parts of the argument that can be checked rather than asserted. In September 2019, the last full quarter before the La Nina sequence began, the average new detached house took 2.2 quarters to complete, just under seven months. By June 2024 that had risen to 3.3 quarters, just under ten months. A 50 per cent increase in average build time, drawn directly from ABS data published in its Home Building Through the Pandemic article.
The pipeline figures give the same period its scale. At the peak in March 2023 there were 104,315 private new houses under construction nationally, against a pre-pandemic normal of roughly 70,000 (70,306 in June 2018, on the ABS series). Each of those houses was occupying a construction slot materially longer than it would have before 2020, which means the same number of active sites and the same trades delivered fewer completions per year. That is a statement about turnover, not about demand.
Weather did not cause that increase on its own, and no part of this page should be read as saying it did. The blowout had multiple causes running at once: materials lead times, a labour shortage, a run of builder insolvencies, and weather. What distinguishes weather is not that it was the largest contributor, which nobody has established, but that it ran continuously through the entire period, left a measurable trace in public data, and was named by the Reserve Bank in the September 2023 Bulletin while attracting almost none of the public attention the other causes did.
These are point-in-time figures and they will date. More recent ABS releases show a modest easing as supply chains stabilised, while build times remain materially above pre-pandemic norms. If you need a current number, pull the current quarter from the ABS release rather than quoting this page.
06 / Conditions
Weather is not only a wet-weather problem
Four conditions that constrain physical output on a residential site. The wet sequence from 2020 is the best documented, which is not the same as being the only one.
Wet
Saturated ground, unworkable platforms, cancelled pours and trades that cannot get plant onto site. The best-documented condition of the last six years, and the one most programmes carry an allowance for.
Extreme heat
Heat constrains concreting, because high ambient temperatures accelerate the set and raise the risk of plastic shrinkage cracking, and it constrains steel and roof work through surface temperatures and worker exposure limits. The usual answer is early starts and shorter days, which is capacity lost even when nobody records a lost day.
Prolonged dry
A dry cycle is not a free run. Reactive soils shrink and move as moisture leaves them, which creates footing and slab risk with its own delays, engineering input and cost. Builders coming out of a wet sequence commonly meet this on the same sites that were bogged two years earlier.
Bushfire conditions
Fire danger can shut a region entirely, close roads, stop deliveries and pull trades onto their own properties. It is short-lived compared with a wet season, and while it is running it is absolute.
The practical implication is that a dry cycle is a different set of constraints rather than an absence of constraints. Builders coming out of a long wet run commonly plan as though the problem has gone away, then meet reactive-soil movement, heat restrictions on concreting and a trade market that is rebalancing as flood recovery work draws down. What changes on a programme when the rain stops is the subject of the building through a dry cycle guide.
07 / Best practice
What a builder can actually measure
National averages are useful for understanding the period and useless for running a job. No client has ever accepted a delay on the basis that the ABS says houses take longer now. What is defensible is the record of what happened on this site, on these dates, with these trades, measured against the programme the job started with. That is the difference between a builder who can evidence a weather impact and a builder who can only describe one.
The operator observation worth stating plainly is that the days lost after the rain stops are almost always more numerous than the days it actually rained, and they are the days nobody writes down. Everyone remembers the week the site was under water. Almost nobody records the following nine days where the pad was still too soft for a pump and the excavator would have churned the platform, because it was not raining and it did not feel like a weather day. Those are the days that turn a legitimate claim into an argument, because the rainfall record shows a clear week and the builder is asking to be believed about the mud.
Four measurements carry most of the weight. Working days lost on the critical path, recorded on the day with what stopped and who was stood down. The unworkable days after the rain cleared, recorded the same way and for the same reason. The trade slot that was lost against the slot that was actually available next, which is the real length of the delay. And elapsed time per stage against the baseline programme, so a builder can see where their own build duration is going. Where the delay runs beyond the allowance inside the build period, that record is what an extension of time stands on, and the specific weather form of that claim is covered in inclement weather and extensions of time.
Where software fits the workflow
None of this is complicated; it is just easy to skip, because writing down a non-event on a day when nothing happened is nobody's priority. That makes it a workflow problem rather than a discipline problem. In VIABUILD the ViaSite site diary records site conditions and lost days against the job as part of the supervisor's normal day, with dated photos attached, and the construction schedule holds the baseline the lost days are measured against. The effect is that weather impact is evidenced per job rather than argued from memory when a variation or an extension of time is raised. The judgement about what to claim stays with the builder; what the software contributes is that the record already exists.
08 / Common mistakes
Where builders lose the weather argument
Six recurring errors, most of them made in the fortnight the delay happens rather than in the meeting where it is discussed.
Measuring rainfall instead of lost days
Millimetres in the gauge are not the impact. The number that matters is working days lost on the critical path, which includes the drying-out days after the rain stopped. Builders who record only the weather record consistently understate their own delay.
Treating weather as an excuse rather than a variable
An excuse is offered after the fact. A variable is measured, allowed for in the build period and evidenced when it exceeds the allowance. The same rain reads completely differently depending on which of those a builder did.
Blaming weather for the whole blowout
Materials lead times, labour shortage and insolvency all pushed build times out over the same period. A builder who attributes everything to rain loses credibility on the part that genuinely was rain, and stops looking at the causes they can actually influence.
Rebooking trades without rebasing the programme
Trades get shuffled by phone call, the programme stays on the wall showing the old dates, and the true finish date is only discovered at the far end. The re-sequence is the moment the programme has to be updated, not an alternative to updating it.
Pricing the fixed-price job on a good year
A build period set on a settled year’s productivity carries no allowance for the year that is actually coming. The allowance sits inside the build period, priced before signing, which is the only point at which a builder has any leverage over it.
Reconstructing the record afterwards
A bureau printout produced six months later proves it rained in the region. It does not prove the site was closed, which trades were stood down, or what the critical path lost. The diary entry written on the day proves all three.
09 / Practical example
Two builders, one wet July
Illustrative only, not a benchmark. Two detached builders on adjacent clay lots lose the same July. Both sites are bogged, both slabs slip, and neither builder had any way to prevent it. The first builder's supervisor logs each affected day as it passes: site closed, pour cancelled, pump stood down, and then, through the following week and a half, pad still unworkable, with photos of standing water on the platform. When the frame crew cannot return until September, that is logged too, along with the date the slot was offered and the date it was available.
The second builder does what most builders do, which is deal with it. Phone calls are made, trades are shuffled, the programme on the wall stays as it was, and the days are not written down because everyone on site already knows what happened. Both builders reach October about ten weeks behind. The first can put a dated record in front of the client showing exactly which days were lost, why the recovery took longer than the rain, and how the trade slot moved. The second has a rainfall printout showing it was wet in July, which the client also knows, and an explanation. Same weather, same ten weeks, entirely different conversation.
10 / FAQ
Common questions.
Because most variables that shape housing supply reward capability. A well-run builder handles cost pressure, trade relationships and regulatory complexity better than a poorly run one, and that difference shows up in outcomes. Weather is effectively the exception, outside a government-mandated shutdown. A competent builder with a full order book and money in the bank still cannot pour a slab on saturated ground or frame in sustained rain. In that moment capability and financial strength offer no advantage at all. That is an operator argument rather than a published finding, but the underlying mechanism is not in dispute: the ground either carries the work or it does not.
No, and nothing on this page should be read that way. The Australian Bureau of Statistics recorded average completion time for new private houses rising from 2.2 quarters in September 2019 to 3.3 quarters in June 2024, a 50 per cent increase, in its Home Building Through the Pandemic article. The causes of that increase were multiple: materials lead times, labour shortage, builder insolvency and weather. Weather is one documented contributor that ran continuously through the whole period and left a trace in public data, including the Reserve Bank naming it in the September 2023 Bulletin. It is a contributor, not the explanation.
No. Extreme heat constrains concreting and steel work, and commonly shortens the working day rather than stopping it, so the lost capacity is real but rarely recorded. Prolonged dry cycles shrink reactive soils, which creates footing and slab risk with its own delays and engineering costs. Bushfire conditions can shut a region entirely. The wet sequence from 2020 gets the attention because it was prolonged and well-documented, but a builder planning around weather is planning around all four conditions, not just rain.
Four things, all of them local to the job rather than national. First, working days lost against the programme, recorded on the day with what stopped and who was stood down. Second, the drying-out days after the rain cleared, which is where most of the underrecording happens. Third, the trade slot that was lost and the slot that was actually available next, because that gap is the real delay. Fourth, the elapsed time per stage against the baseline, which shows where a build duration is going without reference to a national average. All four are evidence when a variation or an extension of time is raised.
They are point-in-time and they will date. The ABS build-time series is published quarterly by state and the national picture has been easing modestly as supply chains stabilised, though build times remain materially above pre-pandemic norms. The pipeline and completion-time figures quoted here are drawn from ABS publications current at the time of writing, and anyone relying on them for a decision should pull the current quarter directly rather than quoting this page. The same applies to the climate outlook, which changes with every model run.
11 / Terms
Glossary for this topic
La Nina and El Nino (the two phases of the ENSO cycle, associated in eastern Australia with wetter and drier than average conditions respectively), reactive soil (clay-rich ground that swells when wet and shrinks when dry, moving footings and slabs), subgrade (the prepared ground a slab or pavement bears on), lost day (a working day on which critical work could not proceed, whether or not it rained that day), wet weather allowance (the anticipated weather already priced inside the build period), trade slot (the booked window a subcontractor holds for a job, which is lost rather than deferred when a job slips), build time (the ABS measure of elapsed time from commencement to completion), contemporaneous record (evidence created on the day of the event rather than reconstructed later). Definitions for the wider vocabulary live in the construction glossary.
Everything on this page ends up as a decision about the programme, which is its own subject; the next reference is construction scheduling.
12 / Keep reading
Related knowledge and guides
13 / Further reading
Primary sources
- Bureau of Meteorology, ENSO history , the record of past La Nina and El Nino events, including the 2020 to 2023 triple La Nina, and the reference point for any claim about a wet or dry sequence.
- Australian Bureau of Statistics, Home Building Through the Pandemic , the source of the completion-time and pipeline figures quoted on this page. Check the current quarterly release before relying on any number here.
- Reserve Bank of Australia, Recent Trends in Australian Productivity, Bulletin September 2023 , where wet weather is named explicitly as a constraint on construction output.
- Property Council of Australia, October 2022, for the Built chief executive's account of losing 40 per cent of construction days to wet weather against a 10 per cent budget.
- Your own site diaries and programme baselines, which are the only source that says anything about your jobs. Everything above describes the period; only your record describes your exposure to it.
You cannot change the weather. You can be the builder who measured it.
VIABUILD keeps the site diary, the dated photos and the baseline programme on one understanding of the job, so lost days are evidenced per job as they happen instead of argued from memory when the claim is raised.