Which cost-of-living tool should you trust?

Numbeo measures a shopping basket

Numbeo is the default, and for good reason: enormous coverage, granular line items, and constant updates from thousands of contributors. If the question is roughly how expensive daily life feels in a city, nothing free answers it better.

Its limits follow from its method. The data is self-reported, so a thin sample in a smaller city can skew a single line — and more importantly, a basket of prices cannot tell you what you would keep. Rent and groceries are the visible part of a move. Tax, out-of-pocket healthcare and what a local wage actually looks like are usually larger, and a price index does not model any of them.

Nomad List measures a community's opinion

Nomad List is a scene as much as a dataset. Its scores capture something real that no institutional source does — whether people like being somewhere, whether the cafés have wifi, whether other remote workers are around — and for choosing between two places you already know are affordable, that is genuinely useful.

It is also crowdsourced sentiment, which means it reflects who is answering. A city popular with one nationality on one budget will score well for that group and tell you little about anyone else. Treat it as a strong signal about atmosphere and a weak one about arithmetic.

Expatistan measures the price gap between two cities

Expatistan does one thing cleanly: it tells you that one city is some percentage more expensive than another, from crowd-submitted prices. As a first filter it is fast and clear.

The percentage is the whole output, though, and a percentage cannot survive contact with a real decision. It does not know your income, your tax position, whether you would be paying rent or a mortgage, or how much healthcare you would be buying yourself.

Crowdsourced and institutional data fail differently

It is worth being precise about why these tools disagree, because the instinct is to assume one of them is wrong. Crowdsourced prices are current and granular and biased by who bothers to submit them: they skew towards the neighbourhoods and the lifestyles of the people filling in the form, which in practice means expatriates and city centres.

Institutional series have the opposite shape. World Bank, OECD and Eurostat figures are methodologically consistent and comparable across borders, which is exactly what a crowd cannot give you — and they are also one to three years old, national rather than local, and silent on anything nobody has decided to measure.

Neither is the honest default. A price index will be closer than an institutional average on what a flat costs in Lisbon this month; an institutional series will be closer on what a Portuguese household actually earns, and no amount of crowd submissions will fix a national wage distribution. The trap is quoting one where the other belongs.

The question none of them answers

All three answer some version of what things cost. None answers what is left, and none answers whether you may legally be there at all — which is the pair of questions an actual move turns on.

That is the gap Geo-Parity was built for: what you keep after rent, everyday prices, out-of-pocket healthcare and income tax; how that compares with the local wage ladder rather than with your own country; which of the 33 nomad visas your income qualifies for, and which of them lead anywhere — only 6 offer a path to residency; and how many days you can stay before you become tax-resident.

It is also worth saying plainly what this site is worse at. Numbeo has far more cities. Nomad List knows things about atmosphere that no dataset holds. If your question is what a coffee costs in a specific neighbourhood, use Numbeo; it is better at that and it always will be.

More writing

• How many days do you actually overlap with a remote team?
• Why cost-of-living rankings ignore the weather — and what changes when they don't
• Why megabits tell you almost nothing about working remotely