In geospatial work the hardware is the easy thing to sell, and the wrong thing to buy. The value was never in getting the pixels. It's in what has to happen after — and that is exactly the part TerraX is built around.
Satellite imagery is open and free — the Sentinel and Landsat archives have been for years — and the software to process it is commoditised. A drone is a weekend purchase. If a proposal is priced around access to imagery or airtime, it is charging you a premium for the part of the job that has already become cheap.
What "interpretation" actually involves
The scarce work sits between the raw scene and the conclusion, and it is a genuine scientific discipline, not a filter:
- Radiometric and atmospheric correction — turning raw sensor counts into physically meaningful surface reflectance, so that a number from one date can be compared to another at all.
- Feature engineering and selection — choosing the bands, indices and transforms that actually carry the signal, and discarding the ones that carry noise. In hyperspectral work this means reasoning about hundreds of narrow, correlated bands.
- Classification and modelling — methods such as random-forest ensembles that learn a defensible mapping from spectra to the thing you care about, with the model's assumptions stated.
- Validation and uncertainty — testing the result against independent ground data and reporting how far it can be trusted. Without this step, an output is an opinion in a raster.
This isn't hypothetical. My own doctoral and published work used hyperspectral remote sensing and machine-learning classification to detect foliar nutrient deficiencies in commercial forestry from reflectance alone — deficiencies invisible to the eye and, until then, reachable only by destructive leaf sampling. That is the far end of interpretation: pulling a defensible, quantified signal out of noisy spectra. It's the same discipline behind a rehabilitation trend, a closure volume, or an EUDR land-cover history — at a more practical scale.
Anyone can fly a site or download a scene. Far fewer people can tell you, defensibly, what it proves — and sign their name to it.
What you're actually buying
When a regulator, an auditor, a lender or a court asks "how do you know?", the drone doesn't answer. The interpretation does, and someone has to stand behind it. That is why TerraX sells the analysis and the accountable sign-off, subcontracts flying and other commodity inputs to vetted specialists, and works beneath prime consultancies rather than competing with them — the prime keeps the client; TerraX delivers the evidence layer that goes under the report.
So the useful question when you commission geospatial work is not "whose drone?" or "what resolution?" It is "who does the analysis, and whose name is on the conclusion?" That is what every one of our services is built to deliver. Buy the brain — the pixels come with it.