Slightly, and generally the wrong way. An EPC is a modelled assessment of a dwelling’s fabric and services, and adding glazed area to a roof increases the modelled heat loss. On a typical domestic rooflight the effect on a rating is small. What moves an EPC materially is insulation, heating and controls.
What an EPC actually measures
Nothing, in the sense of measurement. It is a model.
An assessor records the construction, the areas, the glazing, the heating system and the controls, and software calculates a modelled energy cost from standard assumptions about occupancy and weather.
It does not measure what a house actually uses, which is why two identical houses can use very different amounts and score the same.
How glazing enters the model
By area and by declared U-value. The assessor records the glazed area and its performance, and the model calculates the heat lost through it. More area, or worse performance, means more modelled loss.
A rooflight is glazing on a roof, and it is entered as such.
Why the effect is small
Because one rooflight is a small proportion of a building envelope.
A square metre or two of glazing against a house with hundreds of square metres of wall, roof, floor and windows is a modest change, and ratings move in bands.
It takes a substantial change to shift a band.
Where it stops being small
A large lantern, or several units, on a modest house.
Where the glazed area becomes a meaningful proportion of the roof, the modelled effect is no longer trivial, and on a small property it can matter.
Replacing an old unit improves it
The case that applies to most of our work, and it goes the right way.
Swapping a single skin dome or an early double glazed unit for a current one substantially improves the declared performance of that element, which the model reflects.
So a replacement generally nudges a rating upward rather than down.
Default assumptions
The part worth knowing, because it can cost points unnecessarily.
Where an assessor has no evidence of a glazing specification, the software applies a default assumption based on age and type, and that default is frequently worse than the actual product.
Which means a good unit with no paperwork is modelled as an ordinary one.
Keeping the paperwork
The practical consequence, and it costs nothing.
Retaining the specification and the declared U-value for any rooflight fitted means an assessor can record what is actually there rather than assuming.
On a well specified unit that is worth a small amount, and it is free.
What actually moves a rating
In rough order of effect on a typical house.
Loft insulation, then heating system and controls, then wall insulation, then floor insulation, then glazing.
Glazing is at the bottom of that list, and a single rooflight is a fraction of the glazing.
What we will not claim
That a rooflight improves an EPC, because on a new opening it does not.
People occasionally hope daylight counts for something in the assessment. Lighting is included, and the effect of daylight on it is not modelled in a way a rooflight benefits from.
Where daylight does count
Marginally, through the lighting assessment. The model considers the proportion of low energy lighting rather than how much daylight a room gets, so replacing lamps does more for the score than the rooflight above them.
Whether it matters for a sale
A fair question, and the honest answer is rarely.
A band boundary can matter to a buyer or a lender. A movement of a point or two within a band generally does not.
Where somebody is close to a boundary and about to sell, it is worth knowing before adding glazing rather than afterwards.
Rental property
Where it matters more, and it is worth flagging.
Minimum energy efficiency standards apply to rented property, and a landlord close to a threshold has a real interest in anything that moves a rating.
On a rental, the paperwork point above matters considerably more than on an owner occupied house.
Doing it during other work
Where the arithmetic can be made to work.
Where a roof is being insulated anyway, the improvement from the insulation comfortably outweighs the rooflight, and the pair together move a rating upward.
That is the honest way to combine them rather than claiming the rooflight is the improvement.
Overheating and the assessment
Worth separating, because it is a different check.
Overheating provisions for new dwellings are assessed separately from the energy rating, and a large area of roof glazing engages those rather than the EPC.
What that involves is on can a rooflight be too big for building regulations.
How long a certificate lasts
Ten years, and it does not update itself when a house changes.
Work done after an assessment is not reflected until a new assessment is commissioned, so a rooflight fitted this year sits outside a certificate issued last year entirely.
Which means the effect discussed here only appears at the next assessment, and by then other things will have changed too.
Getting a reassessment
Worth doing after substantial work and rarely worth doing after a rooflight.
A new assessment costs a modest fee and reflects everything done since the last one. After a re-roof with new insulation it is generally worth it; after one rooflight it is not.
What we supply
We supply the declared performance of every unit in writing so an assessor records what is actually installed, and we say plainly that a rooflight is bought for daylight rather than for a rating.
The whole picture is on the energy efficient rooflights page.
Thinking about this job? We survey across Worthing, West Sussex and the South Downs and give you a fixed price before any work starts, with a 10-year workmanship guarantee.