Address to coordinates

Reading the result table

Each row shows your original text, the latitude, the longitude, and the label Photon matched. Always read the label. A query like "Main Street" matches the first Main Street Photon ranks highest, which may be in another state or country. Adding the city, state, and ZIP code fixes most wrong matches.

Coordinates are in decimal degrees on WGS84, the datum used by GPS and web maps, with south latitudes and west longitudes negative.

How to use it

1. Replace the sample lines in the text box with your addresses or place names, one per line.
2. Press Geocode up to 25. Lines after the 25th are ignored.
3. Wait for the table. Lines are geocoded one after another, so a long list takes a few seconds.
4. Check the label column against what you meant, then copy the latitude and longitude columns. Matched points are pinned on the map, which zooms to fit them.

How geocoding works here

Photon is an open-source search engine for OpenStreetMap data. The tool sends one line at a time, waits for the answer, and keeps the top result. Queries run in order so the public server is not hit with a burst of parallel requests.

The result is only as good as OpenStreetMap's address data. In many US cities house numbers are mapped, and the point lands on the building. Where they are not, Photon returns the street or the town, and the point can be far from the door.

Why the limit is 25

The public Photon instance is a shared, free service. Bulk jobs of thousands of rows belong on a self-hosted geocoder or a paid service. Twenty-five lines cover the common case of a short list of offices, stops, or venues. Run a second batch for more.

Once you have coordinates, paste them into CSV to map to plot them with their labels, or into the distance matrix to compare every pair.

Coordinate formats

The table uses decimal degrees. For degrees, minutes, and seconds, UTM, geohash, or plus codes, paste any row into the GPS coordinate converter. For example, Denver's capitol at 39.739, -104.985 is 39°44'20.40"N 104°59'6.00"W in degrees, minutes, and seconds.

Getting cleaner matches

Put one complete address per line, in the order you would write it on an envelope. Remove unit and suite numbers, which geocoders often cannot match. For landmarks, add the city. If a line fails, try the street and town without the house number to at least place it on the right street.