Where to start with net-zero retrofit on an ageing university campus
Most university estates were never planned as a single thing, they simply accumulated over time. A Victorian core, a run of 1960s concrete teaching blocks, a 1990s library, a science building finished last year, and somewhere underneath it all a boiler house that has been patched rather than replaced since the 1980s. Every one of those buildings now sits under the same institutional net zero target, with the same deadline attached.
The sector has spent several years discussing the barriers holding education estates back. Funding, building complexity, competing priorities and shifting policy are all genuine constraints. However, for an estates team looking at forty buildings of wildly different vintages the more immediate question, narrower and less often answered, is where do you actually begin.
What follows is a sequence rather than a strategy. It assumes limited capital, occupied buildings and no clean slate.
Start by knowing what you actually have
Most retrofit programmes that stall do so because they were planned on incomplete information. The first year of work is rarely construction, and more about measurement.
A surprising number of campuses still meter at the site boundary or by cluster, which makes individual building performance invisible. Until consumption can be attributed to a building, any prioritisation exercise is guesswork dressed up as analysis. Sub-metering is comparatively cheap and it is the foundation everything else sits on.
Once the data exists, it must be normalised. Absolute consumption tells you which buildings are big. Intensity, expressed as kWh per square metre and kgCO2e per square metre, tells you which are inefficient. Adjust for hours of use and function, because a 24-hour research facility and a lecture block that runs 30 weeks a year are not comparable on any raw measure.
Alongside the energy picture, you need a condition picture: the age and remaining life of boilers, air handling units, controls and glazing. Thermographic surveys and air permeability testing are worth commissioning on a sample of buildings, because they show where fabric is genuinely failing rather than where the drawings suggest it should be.
Then comes the time to benchmark outwards. The Association of University Directors of Estates publishes annual sector data drawn from institutional returns, and its 2025 report records a third consecutive year of property costs rising faster than inflation, with energy the chief driver, alongside a 3.4 per cent increase in buildings rated in the poorest condition band. Knowing where your estate sits against that trend makes the internal case considerably easier to argue.
Decide which buildings go first
A worst-performers league table is a starting point, not an answer. Sequencing needs four filters applied together, with intensity being the obvious first sort, allowing you to tell which buildings waste the most energy for their size and use.
Occupancy provides the corrective element, with a poorly performing building used intensively offering the potential to repay work faster than an equally poor building that stands half empty for much of the year. Heavily used stock earns its place near the front of the queue.
Caption: A poorly performing building in constant use repays retrofit faster than one that sits largely unused.
Estate strategy decides whether a building has a future at all. There is little sense spending capital on a building earmarked for disposal, rationalisation or a change of use. Plenty of institutions have done exactly that because the energy team and the space planning team were working from different documents.
Alignment with planned works is the cheapest filter to act on. A building already scheduled for refurbishment, roof replacement or a compliance-driven upgrade is dramatically cheaper to decarbonise at the same time than as a standalone job. So, the retrofit programme should be read against the maintenance plan before anything is fixed.
Reduce demand before you size anything
Fabric first is repeated so often it has started to sound like a slogan. The engineering reason is simple. Heat loss determines plant size, and plant size determines both capital cost and running cost for the next two decades.
Install a heat pump into a leaky, single-glazed building and you buy a larger unit than you need, run it at higher flow temperatures than it likes, and lock in poor efficiency. Improve insulation, glazing and airtightness first and the same building may need substantially less installed capacity, with emitters that work at lower temperatures.
The same logic applies to quick wins. LED replacement, pipework insulation and retuning building management system schedules to match the actual timetable are genuinely useful, while releasing cash and building internal credibility for larger asks. The risk is that they get treated as the programme rather than the opening move, and that some of them foreclose better options later. Refurbishing a gas boiler for a marginal efficiency gain is a classic trap. Once the money is spent, nobody will sanction replacing it with a heat pump for years.
Design around the buildings you have
For non-domestic buildings, PAS 2038 sets out a recognised process covering assessment, design, installation and post-completion evaluation. The complication on a mixed-age campus is the older stock, where standardised performance targets often are not achievable. Historic England’s guidance on how energy efficiency standards apply to listed and traditionally constructed buildings is worth reading before anyone commits to an estate-wide target, since Part L carries exemptions where work would unacceptably alter a building’s character, and the advised approach weighs moisture behaviour and long-term fabric health over hitting a single number.
The practical caution is that fabric and services should not be assessed in separate exercises a year apart. Thermal modelling that tests fabric options and plant options against each other, before either is committed, is where the money is saved. Where an in-house team lacks that modelling capacity, specialist retrofit design support is usually cheaper than the oversized system it prevents. Done properly, the exercise ends in a concept design that sets out the carbon saving and the capital cost of each option side by side, which tends to be the document that survives scrutiny from finance when the business case reaches committee.
Heat, ventilation and controls
Moving off gas is the headline change and the hardest one. Full electrification suits some buildings immediately. Others, particularly listed or heavily constrained ones, are better served initially by hybrid arrangements that let a heat pump carry the base load while retaining backup for peak conditions, with the fossil element removed later as fabric improves.
Caption: Air source heat pumps retrofitted to an existing building, part of the shift away from gas heating.
Ventilation deserves more attention than it usually receives. Older teaching space is frequently naturally ventilated with almost no control, which means heat leaves the building through open windows in January. Mechanical ventilation with heat recovery addresses that, and it also gives the indoor air quality monitoring that many institutions now want for other reasons.
Controls are where the cheapest carbon usually sits. Time schedules that no longer match the timetable, zones heated for occupancy that ended three years ago, and setpoints nobody has reviewed since commissioning are common across the sector. A controls audit costs very little and often pays back within a year.
Funding, phasing and the academic year
The funding position has changed and any plan written before 2025 needs revisiting. Government has confirmed no further investment in the Public Sector Decarbonisation Scheme beyond currently awarded projects. Phase 3c funding ran to March 2026 and Phase 4 continues to March 2028, so live projects are unaffected, but there is no confirmed successor scheme to apply to. Institutions planning now should assume their own capital, invest-to-save arrangements or energy performance contracts rather than waiting for a grant round that may not arrive.
Phasing is then constrained by the calendar rather than the budget. The summer window is short, already contested by the maintenance backlog, and shrinking as summer schools and conference income grow. Residential stock offers longer vacation access than teaching space, which makes halls a sensible early target. Work in occupied buildings needs planning around teaching patterns rather than around contractor convenience, and that conversation belongs at the design stage, not on site.
Most institutions decarbonise across many small phases, which is why single-phase projects stand out. Where a college or university has electrified an entire campus heating system in one move, the emissions saving is immediate and the sequencing problem disappears. That approach demands funding and a delivery window most estates cannot assemble easily. However, it is worth testing against your own phasing assumptions before defaulting to increments.
Three things to do this year
If the programme has not started, these are the first moves.
Get building-level metering in place and produce a normalised performance table for the whole estate.
Take one representative building from your most common archetype through a full assessment and design exercise, so you have real costs rather than benchmarks.
Write down the sequencing rule the institution will follow, agreed with finance and space planning, before the first contractor is appointed.
None of that is construction. All of it determines whether the construction, when it comes, is money well spent.