Last winter I sat in a Wellington villa with a couple in their sixties and watched one of them get up to fetch a jersey. The heat pump was running. The double glazing had been in for ten years. The ceiling insulation was up to current code. The house was still cold, still damp in the bathrooms, still cost a small fortune to keep above 18 degrees through July.

If that sounds like your house, you’re not alone. Most New Zealand homes were never designed to keep you warm in winter, dry in the bathrooms, and healthy in the bedrooms at a low energy cost. The Building Code sets the worst house you are legally allowed to build, and most designers have tried to be right at or slightly under the legal minimum (chasing minimum construction costs). The result is a housing stock that performs poorly on every metric that matters at the level of an actual person living in it: thermal comfort, indoor air quality, energy bills. A deep energy retrofit fixes all of these.

The vision and the model

Before you ring a builder, get clear on what you’re after. A deep energy retrofit isn’t a list of upgrades. It’s the path to a house that’s actually comfortable to live in warm in July, dry in the bathroom, fresh air in the bedrooms, a heat pump that does its job, an electricity bill that doesn’t make you wince. All at once.

Once you know what you want, the next step is the one most people skip. Model the house. Not the new windows. Not the new insulation. The house as it is, with its current walls, windows, roof, floors, the shading from the neighbour’s two-storey extension. The whole thing. Then use this model to build a plan for your deep energy retrofit.

A model is a digital twin. You enter the location, the orientation, the building elements, the airtightness, the ventilation, the local shading. The model calculates, in kWh per square metre per year, how much energy the home will need to heat and cool, and flags overheating risk. It does this for the house as it is now, and for the house after combinations of upgrades. You can see, on screen, what each upgrade gets you.

The plan and the model are one process

Here’s the bit I see get skipped more than anything else. A deep energy retrofit is not “do the windows in year one, see what happens, do the roof in year five.” A deep energy retrofit starts with the outcome, builds a model of the house, writes up a masterplan, and refines both as you go. You keep iterating between the model and the plan until the build sequence is obvious.

You can’t write the plan before you’ve modelled the building, and the model isn’t finished until the plan is. The two are integrated. The masterplan is the destination and the route. The model is the simulator that lets you check, on screen, whether the route actually gets you there. They’re not separate stages. They’re the same work.

Step-by-step still works, with each step aligned

Most New Zealanders can’t afford to do it all at once. Step-by-step is the realistic path, and the Deep Energy Retrofit Handbook for New Zealand argues it can be the better one  if every step is aligned with the final deep energy retrofit you’ve modelled.

If your windows are crapped out and you’re getting them replaced, you don’t replace them with whatever’s special. You replace them with windows that will work for the final retrofit. The flashings and air sealing are detailed to integrate with the wall insulation arriving in five or twenty years. The windows you install in Year 1 are part of the path, not a standalone project. That’s the difference between a series of jobs and a coherent deep energy retrofit.

The EnerPHit standard from the Passive House Institute offers two ways to check if the performance hits a target, and they sit inside the model. The component method is the per-element checklist minimum R-values for walls, roof, floor, windows. The energy demand method is the whole-house check if the modelled home land under a climate-dependent target. Use the component method as your baseline. Use the energy demand method to see which elements you can sensibly relax, and which you can’t. That’s what the model is for.

A family we worked with had a 1910s villa. The masterplan lines the work up with the natural rhythm of the building: windows in Year 1, floor insulation in Year 5, MVHR in Year 6, re-roof and reclad in Year 9, PV in Year 10. The detail that matters: the Year 1 window flashings are prepared for the wall insulation arriving eight years later. PHPP at each step confirmed the house was still on track for the EnerPHit target. Each step looked ordinary on its own. None of them made sense without the masterplan.

Without the model, you’re guessing

Here’s where it goes wrong. A deep energy retrofit done without a model is just a string of jobs. Each one looks sensible, add insulation, replace the windows, add a heat pump for heating and hot water. But the interactions are invisible. Add insulation where ventilation is the real problem and you produce a damp house worse than the one you started with. Replace the windows with mid-range double glazing and you might lock yourself out of ever reaching the performance you want. We’ve written a handbook to guide your design team through this process and help them address these risks: Deep Energy Retrofit Handbook for New Zealand which should be available soon.

Where to start this month

Write the vision. What does comfortable, healthy, low-energy look like in your house? Get the family on the same page. Keep it short  a page is enough.

Get a model. A Performance Snapshot  an entry-level PHPP run from $1,800 plus GST  will tell you what the modelled outcome looks like before you commit to the full design. The cost of the model is a fraction of the cost of getting it wrong on the building.

Once you have the vision and the model, the rest masterplan, sequence, builder briefs, consent lodgement  follows. The order is vision, then model, then plan. Stick to it and the rest comes together.

Step-by-step isn’t the second-best option. Done right, it’s the smart way to go it works with the natural rhythm of a building’s life rather than against it. Get the vision right, get the model right, and the rest follows.