Where our data comes from
This page lists every source we use for places, photos, recipes and events, and how we keep the data current. How we set the vegan level of a place is explained in our methodology.
Last updated: 17 November 2026
Places
We list about 50,800 places. They come from these sources:
- OpenStreetMap (about 90% of places). Venues that OpenStreetMap contributors tagged as vegan or with vegan options. A script syncs with OpenStreetMap every week. This data is available under the Open Database License, and every place imported from it shows an OpenStreetMap credit.
- Community submissions. Signed-in users add places on the website and in the apps. Submissions from the apps are checked for duplicates and against the venue's own website before they are published. The place page credits the person who added it.
- Place owners. Owners can claim their listing for free and correct their details. A claim does not change the vegan level or the ranking.
- The VegGuide.org archive (about 1,600 places). VegGuide was a community vegan directory that closed in 2019. We imported places from its archive, and scripts check them like every other place.
- Smaller lists. Accreditation lists for animal sanctuaries, lists of vegan hotels and vegan company headquarters, local vegan guides and blogs, and places our team adds by hand from the venue's own website.
- Google Places, for finding and checking only. Scripts use Google Places to find possible new vegan places and to check whether listed places are still open. A new place is published only after it is confirmed on the venue's own website. We store only Google's place ID. A change that Google reports (closed, moved, new opening hours) goes to a review list; it is not copied into our listing.
How we keep places current
Most checks are done by scripts that run on a schedule. A script never deletes a place: it changes a label only when the evidence is clear, and puts everything else on a review list.
- Every day: a script opens the stored websites of a batch of places and looks for signs that a venue has closed, for example a dead domain, a parked domain or a "permanently closed" notice.
- Every day: a script checks whether places are still open on Google Places, within a fixed daily limit, and sends changes to the review list.
- Every week: a script syncs with OpenStreetMap to pick up new places, changes and closures.
- Every week: a script runs a data-quality audit, for example for duplicates and for labels that do not match their sources.
- Country checks: we go through all the fully vegan places in a country, open each venue's own website and menu, and confirm, change or archive the listing. Sweden, Canada, Portugal and Belgium are the most recent.
- Community reports: every place page has a report button (closed, wrong hours, wrong vegan level, duplicate). We check each report against the venue's own website, usually within a week, and write down what we found.
A closed place is archived, not deleted, with the reason recorded, so it can be restored if the closure turns out to be wrong.
Vegan levels
The vegan level of a place comes from its source (for example the OpenStreetMap tags) and is then checked by scripts and people against the venue's own website and menu. Each place page shows how it was checked: by our team, by the community, cross-referenced across several sources, or imported and not yet checked. Today about 1% of places are checked by a person on our team or confirmed by the community, about 10% are cross-referenced, and the rest are imports not yet checked. The details are in our methodology.
Descriptions
Most place descriptions are short texts built by a script from our own data: the vegan level, the cuisine, the type of place and the city, with a link to the venue's website for the current menu. Some descriptions come from the venue's own text in OpenStreetMap or on its website. Where an older free-text description cannot be checked against the venue, we replace it with a scripted one.
Photos
- Most photos come from the venue's own website.
- Other photos come from Wikimedia Commons (shown with the author and licence), from place owners and from users who upload them.
- Each photo shows its source where we know it. If you own a photo and want it credited differently or removed, email hello@plantspack.com and we will act on it.
Recipes
Recipes come only from creators whose whole catalogue is vegan. Before we add a new source, we check its recipe catalogue and its about page. If a source starts to publish non-vegan recipes, we stop using it. Every recipe credits the creator and links to the original.
Events
Events come from organiser websites, venue pages and local event listings. We publish an event only when the organiser or a local listing states the date for this year, and every event links to that source. We check upcoming events again before they take place.
Tools and scanners
- Barcode scanner: product data comes from the open databases Open Food Facts and Open Beauty Facts.
- Menu scanner and ingredient-label scanner: these use AI to read the photo you take. The result is a guide, not a guarantee; ask the staff or check the packaging when you are not sure.
What we do not do
- We do not copy reviews or ratings from other websites, and we do not create reviews. Every review on PlantsPack is written by a user.
- We do not sell placement. Nobody can pay to appear, to rank higher or to get a better vegan level.
See something wrong? Use the report button on the place page, or email hello@plantspack.com.