Your source of truth for European neighbourhoods

You can view a house in an hour. You cannot view what it is like to live there. That gap is where most of the risk in a move sits, and it is the part almost no listing site tries to answer.

The problem gets worse the moment you cross a border. Sweden publishes school results as a meritvärde out of 340. France publishes them as an IVAL indicator against an expected rate. Denmark publishes crime by municipality and nothing finer. None of these convert into each other, so a neighbourhood score from one national portal cannot honestly be compared with a score from another. Most people moving within Europe are doing exactly that comparison anyway, with nothing to do it on.

Apraiz Neighborhood Insights is our answer: a profile of one district, written the same way in every country, so that the answer you get in Stockholm and the answer you get in Bordeaux are actually the same question answered twice.

What a source of truth has to mean

It is an easy phrase and a hard standard. For us it comes down to three rules that override everything else, including a deadline:

1. Never publish a figure without a source and a date. Both render on the page, next to the number, not in a footnote nobody opens.
2. Never invent a number. Not an estimate dressed as a measurement, not a national average standing in for a street.
3. Where the data does not exist, say so and leave the measure out. An honest gap builds more trust than a confident guess, and a bar drawn over nothing is worse than no bar at all.

That third rule costs us something on almost every article, and we keep it anyway. If a district has no secondary school inside it, it carries no schools score. If a country publishes crime only city-wide, no district in that city gets a crime figure of its own, because printing one city number three times would imply three measurements where there is one.

What we measure

Six headline measures, calibrated to Europe rather than to the United States, using the same plain names everywhere:

Cost of living is the one measure that is never inverted. A high bar means expensive, not good value. Inverting it would quietly turn a fact into an opinion about whether expensive is bad, which depends entirely on who is reading.

The score that leaves cost out on purpose

Each profile carries an Apraiz Living Score out of 100. It is a weighted mean of the five experience measures above, and it excludes cost deliberately.

Blending what life is like with what it costs produces a number that cannot be acted on. A quiet, green, well connected district that is expensive and a cramped, noisy, cheap one can land on the same blended figure, and that figure tells you nothing about either. So we show the experience as one number, the cost beside it, and let you decide which trade you are willing to make. Where a measure is missing, the score is recalculated across the ones we do have rather than treating the gap as a zero.

Why the boundary matters more than the number

This is the part that separates real neighbourhood data from a plausible-looking table, and it is where most published figures quietly fall over.

A price, an income or a school result is only true for the exact area it was measured inside. Change the boundary and you change the number, often by a lot. So we do not take a statistical area off the shelf and assume it matches the district people actually mean. We test it: each landmark that defines a neighbourhood is checked against the published boundary before the district is built from it.

Two examples of that check earning its keep. In Bordeaux, the national statistical grouping puts the Chartrons together with the Grand Parc, a smart riverside quarter and a 1960s social housing estate in one unit. An average across those two describes neither, so we built the district from the smaller areas that actually contain the Chartrons market, its church and the antiques street, and left the rest out.

In Paris, the same check told us the Marais as commonly drawn excluded the place des Vosges, which every resident would tell you is the middle of it. We widened the district to the six quartiers that match how the area is lived in, then regenerated every figure inside the new boundary. The price pool nearly doubled and the quality of life score moved, because the reason it had been low was that no park sat inside the old outline, and that stopped being true.

If a figure cannot be traced to a boundary you can see on a map, it is a number about somewhere, not about here.

Where the numbers come from

Public, official sources, per country, rather than one aggregator applied everywhere. It is more work and it is the only way the figures survive being checked:

• Prices paid, from national transaction registers rather than asking prices. In France that is DVF, which records every recorded sale.
• Income, from national statistical offices at the finest area they publish.
• Schools, from the education ministry's own results, pooled across the cohort rather than averaged over schools, so one small private school cannot outvote a large public one.
• Crime, from the national police or interior ministry statistics, at the finest grain each country publishes and never finer.
• Rents, separating what is advertised from what is actually paid, because in several markets those are very different numbers.
• Amenities, counted inside the district boundary by position rather than within a radius drawn around a point.

Each profile closes with its own methodology note listing the sources it used and the date each was read, plus how often that figure gets refreshed. Prices and rents quarterly, transport and amenities twice a year, schools and crime annually or whenever the source publishes.

What we do not know yet

A source of truth that only tells you what it is good at is a brochure. So, plainly:

• Several countries publish crime only city-wide. In those cities no district carries a crime figure, and the city page carries the rates instead.
• Many central districts have no school inside them, so they carry no schools score at all rather than borrowing one from a neighbouring area.
• Rent actually paid is unavailable in some markets, where only advertised rent is published. Where that is the case, the page says which one it is showing.
• Our own computed walking model is still being built. Today's walking scores are a documented editorial rubric, and they will be replaced by the computed model when it is ready. The page will say when that happens.

We keep a running count of every gap of this kind, per article, so they stay a worklist rather than something quietly forgotten.

How the profiles are published

Coverage runs across seven European markets to begin with, Sweden, Norway, Denmark, Finland, Estonia, Latvia and Lithuania, with Spain, Portugal, Italy, France, Greece, Switzerland and Croatia written and following behind. Profiles publish on a weekday cadence rather than in one batch, which keeps each one reviewable by a person before it goes out.

Every profile carries the same sections in the same order, so once you have read one you know where to find the answer in all of them: first impressions and history, who lives there, getting around, daily life, local places worth knowing, schools, green space, what it costs, weather, the honest downsides, and crime and safety.

How to actually use them

1. Shortlist on the experience score, then look at cost separately. They answer different questions and you will make a better trade seeing them apart.
2. Read the honest downsides section first if you are deciding between two areas. It is the fastest way to find the thing you would have discovered in month six.
3. Check what each figure covers. A city-wide crime rate and a district one are different claims, and we label which you are looking at.
4. Treat a missing measure as information. It usually means the area is small, central, or in a country that does not publish that data, and all three are worth knowing.

If you are weighing up a specific home rather than an area, the neighbourhood is only half the question. An independent valuation of the property itself is the other half, and it is the half an agent has an interest in.