Just Over 600 Square Metres. The Usual Answer Is Two or Three Dwellings. This One Took Four.

Just Over 600 Square Metres. The Usual Answer Is Two or Three Dwellings. This One Took Four.

General information for South Australian landowners and small developers only. This is not planning, legal, engineering or valuation advice. This account is anonymised and simplified, and the fact that one proposal was approved is not an indication that any other will be — yield is not a function of site area, and every application is assessed on its own merits against the Planning and Design Code by the relevant authority, on the facts of the particular site and zone. What your own land can support is a question for a planning consultant who has looked at your title, your zone and your site. Confirm the current position for your own site before relying on anything here.

The answer you will get before anyone looks at the site

Describe a parcel a little over six hundred square metres in an established Adelaide suburb and most people will give you a yield before you have finished the sentence. Two. Maybe three if the frontage is generous.

That reflex is not ignorance. It is pattern-matching against what usually happens, and the pattern it matches is a real one. But the reflex is answering a question about site area, and site area is not what determines yield. What determines yield is whether every dwelling can be given its setbacks, its private open space, its daylight, its privacy and its vehicle access, all at once, on that specific piece of ground — and whether the land not doing any of those jobs has been minimised. Two sites of identical area can differ enormously on that test, which is why area-based rules of thumb are a starting point and nothing more. We have set out why in minimum block size and what it actually governs and in what block shape and orientation do to a development.

This one took four detached dwellings.

A fourth dwelling is not one more floor plan

The intuition that a fourth dwelling means drawing one more house is, in our experience, the most common misreading of infill.

Add a dwelling to a parcel this size and nothing stays where it was. Setbacks are renegotiated across the whole site. Private open space has to be found four times, in usable shapes, not as leftover strips. Daylight has to reach four sets of living areas. Privacy has to work in four directions at once, and on constrained sites that is where schemes most often come undone — a window that was unremarkable in a three-dwelling layout becomes an overlooking question in a four-dwelling one, which is the territory covered in second-storey overlooking and setbacks in SA and in coordinating privacy screening and daylight. And every one of those four dwellings needs vehicle access that works — getting in, reaching the dwelling, and getting back out onto the street — with what is actually required on a given site determined against the applicable Code provisions by the relevant authority.

So the fourth dwelling is not an addition. It is a simultaneous re-solve of five or six interacting requirements, any one of which can be the thing that makes it impossible.

One point needs saying plainly, because it is the difference between this being a case study and being a cautionary tale. A fourth dwelling is not worth pursuing by leaning on the thinnest available reading of a requirement. Whether a proposal is measured against deemed-to-satisfy criteria or assessed on its performance against the relevant outcomes depends on the pathway and is a matter for the authority — a distinction with real consequences, set out in deemed-to-satisfy or performance-assessed. But on either footing, a scheme that is only defensible on the most generous possible reading is a scheme carrying risk into assessment. The standard worth designing to is the one you would be comfortable having examined, which is the argument in raising the bar on residential infill in SA.

The breakthrough was in the driveway, not the floor plans

When a site like this is handed to a design team as "fit four", the work usually begins with the dwellings, because the dwellings are what the client is imagining. Plans get shrunk, then shrunk again, and the scheme fails in a way that looks like the site being too small.

CPM's judgement here was that the constraint was not the floor plans. It was vehicle movement — how cars got in, got to each dwelling, turned, and got out again. On a parcel this size, circulation is often the largest single consumer of land that produces nothing, and on this site it was: a driveway run that is longer than it needs to be, a turning area in the wrong place, or an access arrangement that forces every dwelling to sit off one spine can eat more ground than an entire dwelling footprint. And unlike a floor plan, it is not something you can compress your way out of, because vehicles have geometry and the crossover at the street has its own requirements, which we have described in the kerb crossover and the bond nobody budgets for.

So the design priorities were set in that order: solve the movement, then arrange the dwellings. Working to that brief, the design team reorganised the circulation and recovered the ground that traffic had been consuming — and each of the four dwellings was then resolved with its own setbacks, private open space, daylight and privacy, rather than with those things traded away to make the count.

That is the whole move, and it is worth being precise about who did what. CPM did not design the houses or draw the driveway; the design team did. What CPM contributed was the judgement about where the site's real constraint sat, and the decision to sequence the work against it — which is the difference between a brief that says "fit four" and a brief that says "the yield is in the circulation, start there".

Recovered, not squeezed

The framing that matters: the fourth dwelling was not squeezed in. It was recovered from ground that was being wasted.

That distinction is not rhetorical. A squeezed fourth dwelling shows up as four cramped homes, marginal open space and a scheme that is fragile at assessment. A recovered fourth dwelling shows up as four homes that each stand on their own terms, on a site that is simply using its land better. The unit count is the same; what you have actually built is not — and the difference is decided long before anyone is drawing elevations. How the market eventually reads that difference is a question for a valuer or a selling agent, not for us.

It is also the reason the yield question and the quality question should never be sequenced. An owner who settles the number first and then asks how good the homes can be has already made the trade. On this project the two were solved as one problem, and the four dwellings were approved — the decision itself being the authority's, made on the proposal's merits against the Code. What we can speak to is how the proposal was put together, not why it was determined the way it was.

Why this is not a formula

It needs saying directly, because this is exactly the article that gets misquoted at a barbecue.

Six hundred square metres does not mean four dwellings. It did on this site, on these facts, in this zone, with this frontage and this fall and this relationship to its neighbours, and the relevant authority approved it on its own merits against the Code. Change the frontage and the answer changes. Change the zone and the answer changes. Put a regulated tree or an easement on it and the answer changes a great deal — the developable area of a parcel can be much smaller than its title area, which is a whole problem of its own.

What generalises is the method, not the number:

Find the binding constraint before you commission a layout. On small infill sites it is frequently access and circulation rather than the dwellings — but it is site-specific, and the point is to identify it deliberately rather than discover it after three redraws.

Set the design priorities in the order the site demands, not in the order that is easiest to visualise.

Test the number and the quality together. A yield that only works if every home is compromised is a number, not a result.

If someone has told you your block is a two-lot site, that may well be right. It is worth knowing whether it is right because of the land or because of the assumption — which is the question subdivision and land division starts from. Start a conversation with us, or see how the role that asks these questions is described in what a development manager actually does.