The problem on a cold terrace

The Bananzia pilot began with a complaint that is common in older UK housing: the rooms at the end of the terrace never felt warm, even when the boiler ran for hours. Residents had already replaced windows and topped up loft insulation, but gas bills kept rising. The real problem was not the boiler. It was the order of work. Any new heat source would have been sized to compensate for draughts, gaps around pipework and uninsulated party walls, locking in higher running costs from day one.

Why the heat loss was easy to miss

The cause was a mix of small defects and one big assumption. Thermal surveys found cold bridges at window reveals, open cavities behind kitchen units and unsealed penetrations where cables and pipes left the building. None of these defects looked dramatic on a site visit. Together they added up to a steady background loss that made the existing heating system work harder than its rated output suggested. The assumption was that a heat pump would fix the problem simply by being more efficient. It would not. A heat pump moving heat into a leaky envelope still pays for every unnecessary kilowatt-hour.

The solution: fabric first, then heat pumps

The bananzia approach reversed the usual sequence. Instead of quoting for a replacement boiler or heat pump first, the team mapped air leakage and insulation gaps, then priced a staged retrofit that residents could live through. The work moved from easy wins to major measures, with a second heat-loss survey before any plant was ordered.

1. Seal and insulate first. Loft hatches, pipe penetrations and suspended floors were treated before heating equipment was specified.
2. Measure again. A second survey confirmed the reduced heat demand, so the heat pump would not be oversized.
3. Size to the house, not the street. Each home received a room-by-room heat-loss calculation rather than a generic estimate.
4. Commission with residents present. Controls were set with the household, not left at factory defaults.

The takeaway is practical: a heat pump performs best when the building envelope has already stopped wasting heat.

What the first winter showed

The first monitored winter gave a clearer result than any showroom claim. In the homes that completed fabric work, the heat pumps ran at lower flow temperatures and cycled less often. Residents reported steadier room temperatures and fewer draughts. The most useful figure was not a single efficiency score but the reduced peak demand: the homes asked less of the grid at the coldest hours because the building held heat longer. That matters for streets where several homes might switch to low-carbon heating at once.

Lessons for other streets

Three lessons stand out from the Bananzia pilot. First, sequence is a design decision, not an administrative detail. Second, air sealing and insulation are not extras to be postponed until after the heat pump is installed. Third, monitoring has to continue through at least one heating season, because commissioning mistakes and underperformance show up in real weather, not in a handover folder.

• Start with the envelope. Fix draughts and cold bridges before sizing new heating equipment.
• Use measured demand. A second survey after fabric work prevents oversizing and protects running costs.
• Keep residents in the loop. Clear sequencing reduces disruption and builds trust in the upgrade.

For other terraces, the Bananzia example is less about a single technology and more about discipline: reduce the load, verify the reduction, then choose the heat source.