We start from 1,077,114 BER certificates, and say what kind each number is

SEAI publishes every BER ever issued for an existing home. The file we use has 1,441,089 rows. We keep 1,077,114: existing homes with a valid rating and a floor area of 35 to 400 m². 601,544 of them were assessed in 2019 or later. Certificates run from 18 January 2008 to 6 September 2026.

• Certificates, not homes. The file has no house identifier. A home rated in 2016 and again after an upgrade in 2024 is two rows, so our counts are counts of ratings.
• Rated homes aren’t typical homes. A rating is usually done when a house is sold, let or upgraded. Homes that have stayed in one family for decades are under-counted, and those most need the work.
• A rating is a standard calculation, not a meter reading. Assessors record what they can see; where they can’t, DEAP’s default values apply, and those tend to be pessimistic about old construction.
• The method changed twice. DEAP lowered electricity’s primary energy factor from 2.42 to 2.08 in 2018–19, and to 1.753 on 1 February 2023. Both improved electric-heated homes’ ratings with no work done, so trends across those dates are part method, part work.
• 2026 is partial. The file runs to 6 September 2026 and thins out from 2026-06: we never draw a partial year as a fall.
• One file per number. The site builds from the 8 Sep 2026 file. A few figures come from our research copy of a newer file (26 September 2026), and say so beside the number.

Every rating on this site is shown on the eight-grade scale: A0, A, B, C, D, E, F, G. The bands, in kWh of primary energy per m² a year: A0 up to 42, A up to 75, B up to 150, C up to 225, D up to 275, E up to 325, F up to 375, G over 375. A0 also needs no fossil fuel on site. Every certificate in the file predates the change, so we fold the old grades into the new letters: A1–A3 to A, B1–B3 to B, and so on.

For A, B and C the fold is exact. Below C the new scale moved the band edges, so we checked the fold against each certificate’s own figure: it agrees for 84.5% of 1,077,114 certificates, but only 71.7% of D, 35.2% of E and 3.7% of F. Most old Fs are a G now. No old certificate can be an A0, because the old scale never recorded the fossil-fuel test.

SEAI moved its grant target with the labels: the One Stop Shop requirement of a post-works B2 became a B, and SEAI says the improvement required is unchanged. B2 (125 or less) appears on this site where SEAI’s rules or the national targets still use it.

• Why not every certificate? A certificate done to claim a grant is a post-works rating, so it shows a home that has just been upgraded. Counting those made the typical home look already done: on the old cohorts, 60 of 442 county groups wrongly looked as if their cavities were filled. So “homes like yours” uses certificates for a sale or a private letting: 265,846 of them since 2019.
• One exclusion. A batch of 462 near-identical certificates for one holiday village in County Longford (2025) is left out: it would swing a small county on its own.
• Groups. We sort homes by four things you can answer in ten seconds: the kind of home, when it was built, how it is heated and its rating now. A heat pump has its own heating answer: SEAI’s file records it as electric, but its efficiency, over 100%, gives it away.
• Small groups. Where fewer than 100 homes match, the page stays out of search results. Where fewer than 30 match, we publish no number: we roll up to a wider group (fuel, then rating band, then age) and say so.
• “Don’t know your rating?” We use the rating most homes of that age and heating have that haven’t been upgraded, because someone who doesn’t know their rating has almost never had the work done. Homes with a heat pump are the exception: most really are at a B already.
• Heat pump readiness. 46.6% of certificates for a sale or letting since 2022 show a heat loss indicator of 2.3 or less, SEAI’s line for skipping the technical assessment. Only certificates that carry the figure count: SEAI prints it on every certificate from mid-2022. (Until 5 October we counted a blank as a fail; see Corrections.)
• County figures. A county’s ratings and heating are shares of its certificates for a sale or a letting, 2019 to 2026, placed by the county on the certificate: Dublin’s postal districts count as Dublin, and each city as its county. Its heat pump line counts only those since 2022 that carry a heat loss figure. Census figures count every home, not only those sold or let. Monaghan, Leitrim and Longford have under 3,000 certificates: fine for a share, noisy for anything finer. A county’s figure says nothing certain about one house in it.

• Grants come from SEAI’s own grant pages, terms and homeowner guide. seai.ie often turns an honest script away, so when it does we read the Internet Archive’s copy, say so, and link both. We never pose as a browser to get past it.
• SEAI’s words, word for word. Every rule the plan or a page quotes is stored with the page it came from and the exact words, and a test fails if those words aren’t in our dated copy of that page. A grant page and the plan can’t disagree either: before every release our checks compare each grant figure the grant pages print with the plan’s SEAI-quoted table.
• Costs are SEAI’s published median prices for upgrades done from January to June 2026 (23,412 upgrades; One Stop Shop jobs, 583 homes), one edition across the site. Pages show SEAI’s median for a complete set of windows; the route model still prices windows from each home’s window area (€490 a m²) and marks that cost as ours.
• Running costs use SEAI’s fuel cost comparison, edition of 1 July 2026: what each fuel costs per unit of heat.
• “Up to” means the most SEAI pays. A grant can’t exceed the cost of the work, and some parts (the boiler scrappage grant) only apply to some homes; pages say which.

For every home in a group we work out the cheapest set of upgrades that gets it to a B (150 kWh/m²/yr or less) with the grants on offer, then show the middle answer and the spread around it. It tells you what the work should cost a home like yours at last year’s middle prices with this year’s grants, not what your neighbour paid.

• Checked against SEAI’s own figures. Our heat loss for each home, worked out from scratch, lands within about 1% of the heat loss SEAI’s assessors recorded for the same homes (ratio 1.01, middle half 0.98–1.03, 223,587 homes).
• Physics. DEAP standard occupancy: a fixed heating pattern and household, not yours. Electricity’s primary energy factor is each certificate’s own: 2.08 before 1 February 2023, 1.753 from then on. A heat pump makes 3.5 units of heat per unit of power for space heating and 2.5 for hot water. Bills use SEAI’s delivered prices.
• Grants in the model. The heat pump grant is the 29 September amount: €10,500 for the heat pump, €2,000 for the heating system and €2,000 boiler scrappage for any home without a heat pump today; every grant is capped at its job’s cost. Before 5 October the model still used the earlier grant: see Corrections.
• Spread and confidence. We print the middle (median) and the quicker and slower quarters (p25, p75) of each group. Confidence is high when 300 or more homes match and the spread is narrow (p75 at most three times p25), medium at 100 or more, low below.
• Paybacks come from the 5 October re-run, with the current grant and SEAI’s fuel prices, and are always labelled Our model.
• Not modelled: floor insulation, doors, air-tightness and ventilation aren’t in the bundle, so the worst F and G homes look harder to fix than they are. In a shared building an air-to-water heat pump may be impossible; the model doesn’t know. Grid connection, planning, scaffolding access and asbestos aren’t priced.

• SEAI’s registers list 2,624 companies (home energy 2,117 · solar 528 · One Stop Shops 32), read 25 September 2026. The One Stop Shop and solar lists come from the files SEAI published on 8 September.
• “Say they work in”. Registers let a company tick whole provinces, so a county count is the companies that say they work there, not how many will take your job.
• Never a ranking. We don’t publish jobs done, reviews or ratings, and we don’t call anyone the best.
• Some lists are incomplete: RGII registered gas installers; Safe Electric Register of Electrical Contractors; OFTEC registered businesses publish only who opted in to their public search, and we say so where they appear.

• The targets. Ireland’s plan counts “B2-equivalent” upgrades: 120,000 by 2025 and 500,000 by 2030, plus 45,000 and then 400,000 heat pumps in existing homes. The Climate Change Advisory Council’s count of homes that reached B2 (84,270 by 2025) is a count of B2 homes, so it isn’t quite the same measure as the target. Both count only SEAI-supported work: local-authority retrofits and heat pumps fitted without a grant are outside them.
• Dáil answers. Ministers’ written and oral answers give waits, application counts and county splits SEAI doesn’t publish. We quote the answer’s own words and link it.
• CSO. The Census (April 2022) and CSO’s quarterly BER release. We say “in 2022” beside every Census number: it predates most of the heat pump boom.

How long a single grant takes between approval and payment; grants paid in each county; jobs done per installer; prices by region. Where we use a stand-in, the page says what it is (for grants by county, BERs done to claim a grant), and where there’s nothing, we say that.

How we catch changes. The fetcher keeps a dated copy of 105 source documents (SEAI’s pages, statutes, Department papers) and checks each quoted rule against it. Our dated copies start on 25 September 2026. When a page changes, a person reads it and, if a rule or an amount moved, updates the rule, the figures and the What’s changed log. A copy that changed on its own (page furniture, a new footer) is never logged.

Mistakes are fixed on the page the same day and dated on Corrections. A routine refresh isn’t a correction.

FixMyBER started because SEAI publishes the grant and installers quote the job, but nobody publishes the gap between them: what you actually pay, and how long it takes to come back on the bills. So we worked it out from the files SEAI already puts online.

• We are not SEAI, not a One Stop Shop and not an installer. SEAI hasn’t checked or endorsed anything here.
• We’re not selling anything, and there’s no lead fee in the quote form. If that ever changes, the form will say so before you fill it in.
• The BER file is Irish public sector information, and we’re required to say so. At the foot of every page: Independent — not affiliated with or endorsed by SEAI · Contains Irish Public Sector Data licensed under a Creative Commons Attribution 4.0 International (CC BY 4.0) licence
• data.gov.ie lists the BER Research Tool under CC BY 4.0; SEAI’s own terms for the file use narrower wording about personal, research and education use. We’ve asked SEAI which applies to a site like this, and we’ll publish the answer here.