Why You Can Trust the Price on LandData Explorer

“LandData Explorer’s composite methodology” is not a marketing phrase, nor an internal trade secret. It’s a real research methodology that calculates the price per hectare on this site every day — validated against a dataset of real purchase-sale deals.

The methodology combines five methods, each optimized for its own segment of the data:

• Hybrid weighted (UAH 1–500K/ha) — the widest range, filtering out only the crudest outliers.
• Conservative weighted (UAH 15–150K/ha) — a narrower corridor focused on typical deals.
• Area-balanced (UAH 8–250K/ha, plots of 0.1–50 ha) — the market segment typical by size.
• Ultra-conservative (UAH 20–100K/ha) — the market price “core,” the narrowest and most stable range.
• Adaptive weighting — dynamically boosts the weight of whichever methods more deals fell under the criteria of, in a given sample.

The results of the first four methods are combined using fixed weighting coefficients, while the fifth — adaptive — recalculates those weights dynamically, based on how many deals actually met each method’s criteria. The final price is the average of the fixed-weighting and adaptive-weighting results. This isn’t a single formula but a small statistical system with a built-in safety net: if one method happens to catch an atypical sample, the other four compensate for it.

To see how much more stable a result this produces, the methodology’s author tested it against a dataset of 81,026 real purchase-sale deals from January–August 2025. On that same data, three traditional methods produced very different results: the arithmetic mean came to UAH 62.2K/ha, the median UAH 43.7K, and the weighted average UAH 40.4K. The spread between them is 54% from lowest to highest value. The composite methodology landed in the middle — UAH 50.3K — and that’s not a coincidence but the point of its design: no single method dominates the others, each one only adjusts the shared result.

An even more telling difference shows up in month-to-month stability. Over that same period, the arithmetic mean swung between UAH 54K and UAH 72.2K/ha depending on the month, while the composite methodology stayed within a tight UAH 48.1–52.6K corridor:

Source: State Geocadastre data, January–August 2025. Coefficient of variation: 54% for traditional methods vs. 17.2% for the composite methodology.

April is a telling case: the arithmetic mean for that month spiked to UAH 72.2K/ha due to a handful of atypical deals, while the composite methodology held at UAH 50.1K — essentially its usual level. Numerically, this stability shows up as the coefficient of variation: 54% for traditional methods vs. 17.2% for the composite methodology — meaning the result became roughly three times more stable.

It’s worth being upfront: the methodology systematically produces a result about 17.8% lower than the State Geocadastre’s official figures (the deviation ranged from -7.8% to -29.4% across different months). This isn’t an error but a deliberate methodological choice — the State Geocadastre’s official methodology doesn’t filter outliers the way described above, so it naturally produces a systematically higher result. Which of the two figures is “more correct” is a separate, debatable question that the study’s author openly acknowledges in the paper; our choice is a methodology resistant to anomalies, rather than one that mechanically repeats raw, unprocessed data.

For anyone using the calculator or analytics on LandData Explorer, this means one thing: the number you see isn’t the output of an opaque “black box,” nor a simple average over a raw sample. It’s a method that has been tested against real data, published, and open to scrutiny the same way any other piece of research is — not a commercial formula no one can verify.