You get a confident answer to a question nobody asked. In August 2026 an owner came to Axis with a simple plan. Demolish one of two garages on a suburban lot. Build a second dwelling. Let it for extra income. The screen came back red. But the red band was not a verdict on the plan. It was a finding that the plan had been described as something it was not, and that almost nothing had been proven. Those are different problems, and only one of them is about the property.
The enquiry
A 490 sqm house block in the Newcastle region of New South Wales, owned outright, carrying a house and two double garages. The owner wanted to know one thing. Was it worth borrowing to demolish one garage and put up a second home for extra rent? Two options were on the table. The intake declared the deal a build-to-rent.
Where it came apart
Build-to-rent is a real archetype with hard requirements, and a 490 sqm lot proposing at most two tenancies does not meet any of them.
Scored against that archetype the case landed in the red band, and the two binding gates — planning and price — both failed. Not because the site is poor, but because no planning position and no feasibility of any kind had been supplied. Every component of the score carried C-grade confidence: the finding rested on what the intake asserted, not on a document, a quote or a council response.
Why the red band was not a “no”
This is the part that matters, and it is the part a scoring system will get wrong if nobody reads it properly. The report said so in terms: the red band should be read as wrongly framed and unevidenced, not as a verdict on the second-dwelling idea itself.
A model tells you how a deal performs against the tests for its kind. Point it at the wrong kind and it will still return a number, and the number will still look authoritative. That is the failure mode — not a wrong answer, but a precise answer to the wrong question. Naming the deal correctly is not administrative tidiness; it decides which tests apply, and therefore what the answer means.
What the owner was told to do instead
Three things, before any loan application, and none of them require Axis:
- Written planning confirmation — from the council or an independent town planner — that a second home is permissible on the lot. This was the owner’s own first concern and it was blank in the intake.
- A builder’s quote for the demolition and build. The investment was to be loan-funded, so the cost is not an assumption you can carry.
- Rent appraisals for both the existing house and the proposed dwelling under long-term letting — with any short-stay option treated separately, because it changes both the income profile and the regulatory position.
With those three in hand, the decision can be re-run as what it always was: a small-scale residential feasibility comparing two options. Until they exist, neither option can be ranked on proof, and any ranking would be invention.
Method, and what this does not do
The screen names the deal against the nine deal archetypes, scores it across five weighted dimensions, tests it at five gates, and grades the proof behind each component A, B or C. It is fast and cheap on purpose. Its job is to fail weak or unproven ideas early, before anyone spends money proving them.
It is not a valuation, and it is not formal financial, legal, tax or planning advice. It cannot verify a document nobody supplied. Where the proof is absent it says so and grades itself down, rather than filling the gap with an assumption — which is why this case reads as untested rather than as rejected.
Details have been changed to protect the client’s identity: the address and the owner are not given. The finding, the score, the gates and the recommendation are exactly as delivered.
The transferable lesson
Before you model a deal, name it. If the name is wrong, each number that follows answers a question you did not ask. And it does so with a confidence it has not earned.
Frequently asked questions
Longer answers live on our FAQ page.
Does a red score mean the deal is bad?
Not necessarily. A red band can mean the deal is weak, or it can mean the deal has been described in a way the proof does not support. In this case it meant the second thing. The proposal was measured against a model it was never going to fit, and almost no proof had been supplied. Red meant wrongly framed and untested, not disproven.
What is a deal archetype and why does it matter?
An archetype is the kind of deal you are really doing. Income you buy. Income you make. Value you create by permitting or building. Or land held for the option. It matters because each kind is judged on different tests. Score a small residential proposal against an institutional build-to-rent model and each test misfires, because the tests were written for a different thing.
What does a C-grade confidence grade mean?
Axis grades the proof behind each component of a score. C means the finding rests on what the client asserted rather than on a document, a quote or a written planning position. A page of C grades is not a judgement about the asset. It says the decision has not yet been proven.
If you would like a deal named and screened before you commit to it, a Quick Scan can be ordered through the intake page.