Energy retrofit in Cyprus.Compare several quotes at once.

The problem is heat arriving, not heat leaving

Retrofit practice imported from northern Europe assumes a building losing warmth to a cold outside. Most of the measures follow from that: insulation first, airtightness next, then a more efficient heat source.

A Cypriot commercial building has the opposite problem for most of the year. Solar radiation arrives through glazing and roof, occupants and equipment add to it, and the plant spends its energy removing what has accumulated. Heating demand exists, particularly inland in winter, but it is not what the annual bill is mostly made of.

That inversion changes which measures matter and in what order. It does not make insulation irrelevant — insulation resists heat flow in both directions, and a roof that keeps warmth in during January is the same roof keeping it out in July. What it changes is priority: on a Cypriot building, stopping radiation before it enters is worth more than resisting conduction after it has.

The practical error to avoid is adopting a specification designed elsewhere. Glass selected for its insulating value alone, insulation applied without attention to solar gain, and plant sized against a heating-led model all produce buildings that underperform for reasons the specification cannot explain.

Measure before specifying

The single most useful step is establishing what the building actually does.

Half-hourly consumption across a full year shows the shape of demand: when the peaks occur, whether there is a baseload running overnight, how consumption tracks occupancy, and what happens on days the building is closed. That last one is frequently the most revealing. A building consuming substantially on a closed day is running plant nobody needs, and identifying that costs nothing beyond reading the data.

Sub-metering by system — cooling, lighting, catering, lifts, small power — converts a total into a diagnosis. Where sub-metering does not exist, a temporary survey can approximate it well enough to rank measures.

Thermal survey adds the envelope picture: where heat enters, where insulation is missing or has failed, and where a detail is performing worse than the element around it. Conducted at the right time of day it shows solar gain paths that a drawing does not.

The output of this stage is a ranked list rather than a wish list, and the ranking is specific to the building rather than to the building type.

Shading, and why it comes first

Radiation intercepted before it reaches the glass never becomes a cooling load. Radiation stopped after it has passed through the glass has already become one, and the plant has to remove it.

That is the whole argument for external shading over internal blinds, and it is why the measure that looks least technical is frequently the one with the best return on a Cypriot facade.

The forms available depend on the building and on what may be fixed to it. Fixed horizontal projections work on southern elevations where the summer sun is high. Vertical fins suit east and west, where the sun is low and comes in along the facade — which is also where the worst gain occurs and where a horizontal projection does least. External louvres and screens handle both at more cost. Planting is slower and cheaper and does not need a fixing.

Where a facade is being renovated anyway, shading shares the access and the scaffold with the rest of that work, which is the argument for deciding both together rather than treating energy as a separate later project.

The roof

On most Cypriot commercial buildings the roof is flat, dark, and receives more radiation for longer than any other surface. On older stock it is frequently the weakest part of the envelope.

Treatment takes two broad forms and they are not alternatives. Insulation reduces heat flow into the space beneath. A reflective or light-coloured finish reduces how much of the arriving radiation is absorbed in the first place. Applied together they address the same problem at two points, and on a building with occupied space directly below the roof the combination usually changes conditions noticeably rather than marginally.

Two constraints govern the work. The waterproofing has to be right, because a roof being insulated is a roof being opened, and a retrofit that solves an energy problem while creating a water one is a poor trade. And the structure has to carry whatever is added, which is a question for an engineer where the build-up is substantial.

Glazing, and the property that matters

Replacing single glazing removes a genuine weakness. What matters when choosing the replacement is which property is being bought.

Insulating performance resists conduction. Solar control governs how much radiation passes through. In a heating-dominated climate the first dominates the decision; in a cooling-dominated one the second frequently does. Glass selected on insulating value alone can admit solar gain that the plant then pays to remove.

Frames matter as much as glass on older buildings, where the frame may be both a thermal path and the source of the air leakage. Where windows are being replaced, the surrounding detail is the opportunity — an insulated reveal and a properly sealed perimeter are cheap while the opening is exposed and expensive afterwards.

Glazing is also where the shading conversation returns. External shading on a facade in direct sun usually outperforms any glass specification available, and the two together outperform either.

Plant, controls and commissioning

Controls are the cheapest measure and the most commonly omitted.

Plant running outside occupied hours, heating and cooling set points close enough to fight each other, zones conditioned to the same temperature regardless of use, and systems commissioned once at handover and never revisited — each of these consumes energy that no equipment replacement addresses, and each is correctable at a fraction of the cost of replacement.

Where plant does need replacing, the order matters. Load reduction first, plant sizing after. A chiller specified before shading, roof and glazing work is a chiller sized for the building as it was, and it will spend its life running at part load in a building that no longer needs its capacity. Part load is where efficiency is lost, so oversizing is not a harmless margin — it is a permanent penalty paid to avoid a sequencing decision.

Hot water deserves separate thought. Solar thermal is long established on the island and well suited to the resource, and for a building with substantial hot water demand — a hotel, a kitchen, changing facilities — it addresses a load that is otherwise served by the most expensive route available.

Occupancy, and the retrofit that is behavioural

Some of the largest differences in a commercial building come from how it is run rather than from what is installed.

Occupancy patterns determine when conditioning is needed, and a building conditioned to a fixed schedule that no longer matches its use is paying for comfort nobody is receiving. Seasonal buildings — hotels, resort retail — have the sharpest version of this, where plant serving a full house in August is serving a skeleton operation in November.

Metering makes this visible and it is the reason metering leads the process. A retrofit that installs measures without establishing a monitoring routine has bought a one-time improvement; one that leaves the client able to see consumption by system, month by month, has bought the ability to keep it.

Lighting, and the load that arrives twice

Lighting is the measure most often already done and most often done incompletely, and in a cooling-dominated building it counts twice.

Every watt of inefficient lighting is paid for once at the fitting and again at the chiller, because the heat it produces becomes cooling load in the same space. That doubling is why lighting upgrades tend to perform better in Cypriot commercial buildings than the fitting-level arithmetic suggests.

Where an upgrade has been done, it is worth checking what was left out. Back of house, plant rooms, stores, car parks and external lighting are commonly skipped because they are unglamorous, and external lighting in particular can run for long hours on a fixed timer that no longer matches the season.

Controls apply here as much as to plant. Daylight is abundant on this island and a perimeter zone lit at full output on a bright afternoon is lighting a room the sun has already lit. Occupancy sensing in intermittently used spaces — stores, corridors, sanitary accommodation — addresses hours rather than watts, which is frequently the larger number.

Verifying that it worked

A retrofit without verification is an expenditure with an assumption attached.

The comparison that matters is consumption before against consumption after, normalised for the things that changed independently: weather, occupancy, opening hours, and any change in what the building is used for. Without that normalisation a mild summer flatters the work and a hot one hides it.

The practical arrangement is to agree at the outset what will be compared, over what period, and using which data. Where a contractor's proposal carries a performance expectation, the method for testing it belongs in the contract rather than in a conversation afterwards.

This is also the point at which the monitoring routine either exists or does not. A building whose consumption is read monthly by somebody with responsibility for it keeps the gains; one where nobody looks tends to drift back as set points are adjusted, schedules are overridden for a one-off event and never restored, and equipment is replaced like-for-like without reference to the strategy.

Sequencing the programme

Measure, then reduce load, then right-size plant, then monitor.

Within that, group work by access rather than by measure. Facade shading with facade repair. Roof insulation with roof waterproofing. Glazing with any internal work affecting the same rooms. The marginal cost of doing two things from one scaffold is far below the cost of two visits, and that consideration frequently outranks the theoretical payback ranking of the individual measures.

For an operating building, the constraint is the disruption rather than the work. Envelope measures can usually proceed with the building in use; plant replacement generally cannot, which puts it into a shutdown window and makes it the item the programme is built around.