Ask a client's GIS team how they want a drone map delivered and the answer is usually the same: "Send a GeoTIFF." It is the default file for orthomosaics and elevation grids in most mapping workflows, yet many pilots treat it as just another image format. It isn't. A GeoTIFF carries its own location, and that is the whole reason engineers, surveyors and planners ask for it.
This guide explains what a GeoTIFF is, what is stored inside one, how it differs from a plain TIFF or a JPEG, and what to check before you hand one over. If you want the step-by-step for building a deliverables package in SkyView, that is covered in our post on exporting a GeoTIFF deliverables package locally. This one is about the format itself.
What is a GeoTIFF?
A GeoTIFF is a standard TIFF image with extra tags that describe where the image sits on the earth. The Open Geospatial Consortium (OGC) publishes the format as the OGC GeoTIFF Standard, version 1.1 (OGC document 19-008r4, checked Nov 5, 2026). According to that page, the format was first developed in the early 1990s, formalized as Revision 1.0 in 1995, and is now "used throughout the geospatial and earth science communities to share geographic image data."
In practice, that means two things travel together in one file:
- The pixels. Colors for an orthomosaic, or numbers such as elevations for a surface model.
- The georeferencing. Where the image is anchored, how much ground each pixel covers, and which coordinate reference system (CRS) those coordinates belong to.
Open a GeoTIFF in GIS software and it drops onto the map in the right place, at the right scale, with no manual alignment.
What's actually stored inside the file
Georeferencing tags
TIFF is a tag-based format, which is why it could be extended for mapping in the first place. GeoTIFF adds tags that tie pixel positions to map coordinates. The GDAL GeoTIFF driver documentation (checked Nov 5, 2026) lists the forms this can take: "one tiepoint and pixel size, a transformation matrix, or a list of GCPs." For a typical drone orthomosaic, the first form is the common one: a known corner coordinate plus a pixel size, from which every other pixel's position follows.
The coordinate reference system
A GeoTIFF also records which CRS its coordinates are in, typically through codes from the EPSG registry. That is what lets a GIS program know whether the numbers are UTM meters, state plane feet or latitude and longitude. If you need a refresher on how datums, projections and EPSG codes fit together, see our guide to coordinate systems for drone mapping. The short version: the file can only be as correct as the CRS it was written in, so confirm the client's system before you export.
Bands and data type
A GeoTIFF can hold one band or many. An RGB orthomosaic is usually three 8-bit bands, sometimes with a fourth for transparency. A digital surface model or terrain model is usually a single band of floating-point numbers, where each pixel value is an elevation rather than a color. That difference matters, which is why we cover the two surface types separately in DSM vs DTM.
Nodata, compression and overviews
Most drone maps have empty areas outside the flight boundary. A nodata value or mask tells software to treat those pixels as empty instead of black or zero elevation. Files can also be compressed (LZW and DEFLATE are common lossless options) and can carry internal overviews, which are lower-resolution copies that let software draw a zoomed-out view quickly.
GeoTIFF vs a plain TIFF or JPEG
A plain TIFF or JPEG of the same orthomosaic may look identical on screen. The difference is what the software knows about it.
- Plain TIFF: same image container, no location. It opens in a photo viewer, but a GIS program has no idea where to put it.
- JPEG: no location either, plus lossy compression and 8-bit color only. Fine for a preview or a report figure, not suitable for an elevation grid, which needs real numeric values.
- GeoTIFF: the image plus its position, scale and CRS, in a format that handles both color imagery and numeric grids.
This is why "just send me a JPEG" works for a marketing image and fails for an engineer who needs to overlay your map on a design file.
World files: the sidecar alternative
Before you deliver, it helps to know about world files. A world file is a small text file that sits next to an image (for TIFFs it usually ends in .tfw) and holds the pixel size and the map coordinates of the image's upper-left corner. The GDAL documentation notes that if no georeferencing is found inside the TIFF itself, it will check for an ESRI world file.
The catch is that a world file stores position and scale, not the coordinate system. Someone receiving a TIFF plus a .tfw still has to be told the CRS. Some CAD setups read world files more readily than embedded GeoTIFF tags, so if a client's drafting team says the image lands in the wrong place, ask whether their software is reading the tags or looking for a sidecar.
Cloud Optimized GeoTIFF, in general terms
You will also see the term Cloud Optimized GeoTIFF, or COG. A COG is still a valid GeoTIFF, but its internal layout is arranged with tiles and overviews so that software can read just the part of the image it needs, including over the web, instead of downloading the whole file. The GDAL COG driver page (checked Nov 5, 2026) is a good technical starting point. Whether a client needs a COG depends on how they host and view data; for a file that will be loaded into desktop GIS, an ordinary GeoTIFF is usually what they expect.
File size and large drone maps
High-resolution orthos get big quickly. A classic TIFF is limited to 4 GB, and the GDAL documentation describes BigTIFF as the evolution of the format that supports larger files. Tiling, lossless compression and internal overviews all help a large GeoTIFF open and pan smoothly. If a client reports that your ortho "freezes" their software, file layout is often worth checking before anything else.
A quick pre-delivery checklist
- Confirm the client's horizontal CRS and vertical datum in writing before you export.
- Open the file yourself in a GIS program, or run
gdalinfoon it, and confirm the CRS, pixel size and extent look right. - Check that nodata areas display as empty, not as a black frame or a pit of zero elevations.
- Keep elevation grids in a numeric data type with lossless compression.
- Note the CRS, datum, units and processing date in your delivery email.
A GeoTIFF that lands in the right place does not, on its own, prove the map is accurate. Positional accuracy comes from how the job was flown, controlled and checked, not from the file format.
Where GeoTIFFs come from in a drone workflow
The two most common GeoTIFFs a mapping pilot delivers are the orthomosaic and the elevation surfaces (DSM and DTM). Both come out of the photogrammetry processing step, which is where SkyView Desktop fits. SkyView processes RGB drone photos into orthomosaics and DSM/DTM surfaces on your own Windows PC, crops deliverables to the site boundary and exports georeferenced rasters, with no per-map processing fee. You can see what is included in each plan on the features page.
For clients who would rather look at the map than load a file, SkyView publishes it to a private Pilot Ledger project with shared client access. And once the files go out, the same Pilot Ledger account holds the quote, the job and the invoice, so delivery and billing don't live in separate tools.
Process and export on your own hardware
If your clients ask for georeferenced orthos and elevation grids, it makes sense to build them on the machine that already holds the imagery. See SkyView plans and start a 14-day trial.