The GlobeNomad visa dataset
Every long-stay and nomad visa record we publish, as one JSON file. Free to download, free to reuse — including commercially — under CC BY 4.0. Income requirements, fees, stay lengths, renewal and residency rules, tax treatment and family provisions, each one linked to the government page it came from and stamped with the date it was last checked.
Why this exists
Because we needed it and it didn't exist. Visa rules are published by dozens of governments in dozens of formats, most of them unstructured, several of them contradicting themselves, and a few describing programmes that closed years ago. Anyone building a tool, running a study or writing an article on where people can legally live and work has to assemble this by hand first.
So it's published. Not as a teaser for a paid tier — the whole set, the same records that power every country page on this site, regenerated from the same database. If you use it and it saves you a week, that's the point.
What's in each record
- Identity — country, region, programme name, and the page on this site
- Status — open, closed, announced but not yet open, or no nomad route at all
- Money — minimum income in USD, the original local-currency wording, and the application fee
- Duration — maximum stay in months, processing time, renewability and total years permitted
- Settling — whether it leads to permanent residency or citizenship, and after how long
- Tax — the residency-day threshold, whether foreign income is taxed, and any special regime
- Family — whether dependants are allowed and the income to add for a partner or child
- Applying — whether you must apply from abroad, by what method, and the requirement and step lists
- Provenance — the official government URL and the date the record was last verified
Read this before you render it
A null value means the government has not published that rule.
It never means "no". A record whose pr_path is null must not be displayed
as "No permanent residency path" — no source says that, and someone may plan a move
around it. Show unknown as unknown.
Every field is always present in every record. A value is either populated or explicitly
null, so you never have to work out whether a missing key means an unknown
answer or a dropped field. That distinction is the difference between a dataset you can
build on and one you have to double-check.
How to attribute
CC BY 4.0 asks for credit and a link. Paste this, or anything that names the dataset and links back:
Data: <a href="https://globenomad.com/data">GlobeNomad Visa Dataset</a>, CC BY 4.0That's the whole obligation. No registration, no API key, no rate limit, no email address to hand over, and nothing to renew.
What's deliberately not in it
- No affiliate or tracking links. This site is funded by affiliate referrals, and the dataset is kept clear of them on purpose. Data you can cite has to be data nobody paid to shape.
- No images. The country photography on this site is ours and is not covered by the CC BY 4.0 licence — only the data is.
- No scraped third-party content. Records are compiled from primary government sources, which is why each one can carry a link you can check yourself.
How it's kept current
Records are re-checked against their official source at least once a year and immediately when we learn a rule has changed, and the published file is regenerated from the same database that serves the country pages, so the two are never assembled from different sources. Regenerating the file is a manual step, so a correction can reach the site first — each record carries the date it was last checked and the repository carries the date the file was built, and those, not a promise here, are what say how current a copy is. Closed and never-launched programmes stay in the set with an honest status rather than being quietly deleted, because "this closed in 2024" is a fact people need and almost nobody publishes. The full method is on how we verify.
Corrections
If you've just been through one of these applications, you know the current rules better than any published page does. Send a correction through contact with the country and what you ran into, and it gets checked against the official source and fixed — in the dataset and on the site at the same time.
Frequently asked questions
Yes. The dataset is published under CC BY 4.0, which permits commercial use, redistribution and modification. The only condition is attribution: credit the GlobeNomad Visa Dataset and link to globenomad.com.
It means the government has not published that rule, so we have not recorded an answer. It never means "no". Rendering a null permanent-residency field as "No PR path" would state something no source says, on a subject where a wrong answer costs someone a rejected application.
Every record is re-checked against its source at least once a year, and immediately whenever we learn a rule has changed. The published file is regenerated from the same database that powers the country pages, so the two are never assembled from different sources — but regenerating it is a manual step, so a correction can reach the site before it reaches the file. Every record carries the date it was last checked, and the repository carries the date the file was built; those are the two dates to trust.
No. The export carries no affiliate or tracking URLs of any kind. The site is funded by affiliate referrals elsewhere, but the data is kept clear of them so it can be cited as a source rather than read as a placement.
A single UTF-8 JSON file with one object per programme — several countries run more than one long-stay route, so Thailand holds four records and a consumer grouping by country keys on country_slug, not slug. A header carries the licence, the attribution string and the newest verification date. Every field is always present — a value is either populated or explicitly null, so a consumer never has to tell a missing key apart from an unknown answer.
Prefer to read it as pages?
Every record in the file has a page with the sourcing and caveats written out — browse digital nomad visas by country, or compare them side by side in the visa comparison table. The visa finder is this same file queried by income, region and length of stay — a working example of what the data does before you download it.