Every property valuation is really an answer to one hard question: what would this specific home sell for today? Our instant valuation tool answers it for any Irish address in seconds, built entirely on real, public sold-price data. This post lifts the lid on how it works — not the secret sauce, but the ideas: where the data comes from, what we add to it, and the recent work that makes each estimate more trustworthy.
The foundation: the Property Price Register
Ireland is fortunate to have the Property Price Register (PPR): a public record of every residential property sale since 2010, with the address, the date and the actual price paid. That is gold for valuation. It isn't a listing site's asking prices or an estate agent's guide — it's what buyers actually paid, which is the only number that really matters. Hundreds of thousands of these transactions are the bedrock of everything we do.
But raw PPR data, on its own, only gets you so far. It tells you an address and a price — and little else. It doesn't tell you whether a sale was a one-bed apartment or a five-bed house. It has no coordinates, so you can't easily ask "what sold near here?". Addresses are written inconsistently, with abbreviations, punctuation and typos. And a headline figure from a single nearby sale can be wildly misleading. Turning that raw register into a reliable per-home estimate is the real work.
What we layer on top
We treat the PPR as a base and enrich it into something far more useful:
- Precise geography. Every sale is geocoded to a map coordinate, so we can find genuinely nearby comparables rather than relying on a town or postcode label.
- Clean, consistent addresses. We normalise every address to a standard form — expanding abbreviations, fixing punctuation and spacing — so "28 Slane Rd" and "28 Slane Road" are recognised as the same place.
- Property type. The PPR doesn't say whether a sale is a house or an apartment. We enrich sales with that classification, because — as we'll see — it changes everything.
- Bedroom counts. We add bedroom information where we can, so a valuation can lean on homes of a genuinely similar size, not just a similar location.
This enriched layer is what separates a useful estimate from a crude "average price in the area" figure. It's the difference between "homes around here sold for about X" and "a home like yours is worth about X."
Finding the right comparables
With that foundation in place, valuing a home follows the same logic a good estate agent would use — just at scale, and consistently. We gather recent, arm's-length sales near the property, widening the search until we have enough to be meaningful, and we bring every one of them to today's money using local price trends, so a 2023 sale is expressed in 2026 terms. Closer, more recent and more similar sales count for more.
The word doing the heavy lifting there is similar. One of the most important refinements we've made is to treat apartments and houses as separate markets. Valuing a two-bed apartment off the three-bed semis on the next street gives a badly inflated number — they simply aren't the same product. So the engine leans overwhelmingly on same-type sales and all but ignores the other type. Getting property type right is often the single biggest factor in a sensible apartment valuation.
Recent work: outliers and the size relationship
Two pieces of recent analysis have made our estimates noticeably more robust.
Setting aside local outliers. Occasionally a single unusual sale — a penthouse, a large multi-unit deal, a trophy home — lands right beside a cluster of ordinary sales. Left unchecked, one such price can hijack the whole estimate. We now detect when a nearby sale is a statistical outlier relative to its neighbours and stop it from dominating, so your valuation reflects the ordinary market around you, not the one exceptional deal.
Respecting how size drives price. Analysing thousands of sales confirms an intuitive truth: in the same location, a two-bed is worth more than a one-bed, and a three-bed more than a two-bed. That sounds obvious, but a naïve model that just averages nearby prices can break it — especially where there are only a handful of sales of a given size. We measure these size relationships directly from the data and build them into the engine, so a larger home is never valued below a smaller one next door, and a size with thin local data still gets a sensible, properly-ranked figure.
Honest by design
No statistical model is a substitute for a valuer standing in the room, and we don't pretend otherwise. That's why every estimate on HomeIQ comes with its working shown: the comparable sales we used, how far away and how recent they were, and a confidence level that reflects how much good data stood behind the number. A tight cluster of recent, same-type, same-size sales next door earns high confidence; a sparse, scattered pool earns a wider range and a note of caution. You get a number and the reasons for it.
Try it on your own home
The same engine that we've pointed at entire portfolios runs when you type in a single address. It starts from real sold prices, enriches them with location, type and size, chooses the right comparables, and shows you exactly what's behind the result. It's free, instant, and built on data you can check yourself.
Curious what your home is worth? Get a free instant valuation built from real comparable sales near you — with the comparables and confidence shown, and apartments valued off apartments. Value your property →
HomeIQ valuations are an indicative statistical model based on Property Price Register data and are not a valuation of record, investment advice, or a recommendation. For a formal valuation, consult a qualified valuer.
© 2026 HomeIQ. All rights reserved. The analysis and content in this article are the property of HomeIQ (homeiq.ie). You're welcome to share or cite it with a link back to the original; reproduction in full without permission is not permitted. Questions or permission requests? Contact us.