GIS Tools - Coordinate Precision Reducer

What Is Coordinate Precision Reduction?

Coordinate precision reduction trims the number of decimal places in the longitude and latitude values of a GIS file. The shape of the data is unchanged, but the file gets dramatically smaller because each coordinate becomes a shorter string. The Coordinate Precision Reducer on gis.tools is a free, browser-based tool that takes a GeoJSON file, lets you pick a target precision (typically 4 to 7 decimal places), and rewrites every coordinate accordingly. It's the simplest, safest way to shrink a vector file without changing its geometry.

Coordinate precision matters more than most people realise. A typical geocoder or GIS export emits coordinates with 12 to 15 decimal places — far more precision than is meaningful at any real-world scale. The 7th decimal place of a latitude is about 1 cm; the 10th is about 10 micrometers; the 14th is about 1 nanometer. Storing nanometer precision for a city boundary is wasted bytes. Reducing precision typically cuts file size by 30–60% with no visible change in the data.

Decimal Places vs Real-World Distance

At the equator, one degree of longitude is about 111 kilometers. Each additional decimal place divides that by 10:

0 decimals: 111 km
1 decimal: 11.1 km
2 decimals: 1.11 km
3 decimals: 111 m
4 decimals: 11.1 m
5 decimals: 1.11 m
6 decimals: 11.1 cm
7 decimals: 1.11 cm
8 decimals: 1.11 mm
9 decimals: 0.11 mm

For most web maps, 5–6 decimal places (1 m to 10 cm precision) is plenty. For cadastral or surveying applications, 7–8 may be needed. Beyond that, you're storing noise.

How the Reducer Works

The reducer parses the input GeoJSON, walks every coordinate in every geometry, and rounds each value to the target number of decimal places. The result is written back as a new GeoJSON file. The number of vertices is unchanged — only their precision is reduced. Properties are untouched.

Smart Defaults

The reducer suggests a default of 6 decimal places (about 10 cm precision), which is suitable for almost all web mapping use cases. You can override this for higher or lower precision.

Round vs Truncate

The reducer rounds (not truncates) coordinates, which preserves the centroid more accurately than truncation.

Practical Applications

Web Map Optimisation

Every byte saved on a GeoJSON file is a faster page load. Precision reduction is the lowest-effort optimisation available, often beating gzip compression by a factor of two.

Mobile and Offline Maps

Mobile apps storing offline data benefit dramatically from smaller files. A 5 MB precision-reduced file fits in a free-tier app bundle where a 12 MB original would not.

Email and Cloud Storage Limits

A file that hits Gmail's 25 MB attachment limit may slip under it after precision reduction.

CDN Egress Cost

For maps served at scale, every 30% reduction in payload directly reduces CDN bandwidth bills.

Data Anonymisation

Reducing precision can also serve as a privacy measure. Snapping individual house addresses to 100-meter precision (3 decimals) effectively anonymises them at the block level while preserving aggregate statistics.

Archive Storage

Long-term archives benefit from smaller files. Precision reduction is a one-shot improvement that compounds over millions of records.

API Responses

APIs returning GeoJSON can apply precision reduction at the response layer to keep payloads small.

Step-by-Step Workflow in gis.tools

Open the Coordinate Precision Reducer page.
Drag your .geojson file into the drop zone or paste GeoJSON text.
The tool reports the original file size and average coordinate precision.
Choose a target precision (4–8 decimal places typical).
Preview the size reduction and any visible geometric change (usually none).
Click Download Reduced GeoJSON.

Worked Example

A developer has a 14 MB GeoJSON of bike lane centrelines for a metropolitan area, exported from a city open-data portal. The coordinates have 14 decimal places — silly precision for street-level data. They drop the file in, set the target to 6 decimal places (10 cm precision), and the output is 5.8 MB — a 59% reduction. Page load time on their bike map app drops from 1.8 seconds to 0.7 seconds. The bike lanes are visually identical at any reasonable zoom level.

Common Pitfalls and Gotchas

Cumulative rounding errors — Repeated rounds at different precisions accumulate. Round once, not repeatedly.
Precision below your tolerance — Rounding to 3 decimals (~100 m) can break adjacent polygon boundaries that need to be coincident at the meter level.
Z and M values — Elevation and measure values may need different precision than X/Y; the reducer can handle them separately.
Survey-grade data — High-precision surveying data shouldn't be reduced below the survey's stated accuracy.
Snap-to-grid effects — Aggressive precision reduction can produce visible stair-stepping at high zoom levels.
Map projection sensitivity — In projected CRSs (like Web Mercator), 6 decimal places mean millimeters at the equator but tens of centimeters at high latitudes.
Topology breakage — Reducing precision below the spacing between adjacent vertices can collapse them, creating zero-length segments.

Tips for Best Results

Default to 6 decimal places unless you have a specific reason for more or less.
For street-level data, 5 decimals (~1 m) is usually enough.
For continental boundary data, 4 decimals (~11 m) often suffices.
Pair with the GeoJSON Simplifier for compounding savings.
Validate after reduction with the GeoJSON Validator & Fixer to catch any topology issues.
Check the file size before and after to confirm the reduction worked.
Keep the original at full precision as an archive.

Comparison with Other GIS Approaches

PostGIS has ST_SnapToGrid which serves a similar purpose for projected data. ogr2ogr supports -lco COORDINATE_PRECISION=6 when writing GeoJSON. Mapshaper has a precision setting in its CLI and web UI. The browser-based gis.tools reducer is the most accessible option for non-developers and the only one that requires no installation.

Performance Considerations

Precision reduction is O(n) on vertex count and very fast — a million vertices reduces in well under a second. The bottleneck is usually parsing and serialising the JSON, not the rounding itself.

Data Privacy and Browser-Based Processing

Everything runs in your browser tab. The original file is not uploaded, no telemetry is collected, and the reduced version is generated locally and downloaded directly to your machine.

Related GIS Concepts

Significant Figures — The number of meaningful digits in a measurement, set by the precision of the source instrument.

Snap to Grid — A related operation that rounds coordinates to a fixed grid (often used in PostGIS).

Quantisation — The same idea applied in TopoJSON, where coordinates are mapped to integer indices on a grid.

Floating-Point Precision — The hardware-level limit on how many decimal places a 64-bit float can represent (about 15–17).

Frequently Asked Questions

Will reducing precision change the visible shape of my data?

At typical web map zoom levels, no. At extreme zoom (street level or below), aggressive reduction may show stair-stepping.

What's the safest default precision?

6 decimal places — about 10 cm at the equator — is a good universal default.

Can I reduce precision differently for each feature?

This tool applies a single precision globally. For per-feature variation, use a script.

Does precision reduction affect attributes?

No — only geometry coordinates are changed.

Coordinate Precision Reducer

Precision Reference