Work — Equiterrain
Equiterrain is an independent land valuation platform for African markets, beginning with Cameroon. It combines title information, economic-gravity modeling, local benchmarks, and Bwendi location context to produce a structured estimate where comparable-sales data is difficult to obtain.
Across much of Africa, land is the largest asset most families ever own — and the one they have the least information about. Sellers undersell parcels passed down for generations. First-time buyers overpay because they have no benchmark. Lenders cannot extend credit against land they cannot reliably value. Speculation, manipulation, and missing data fill the gap.
Without a common evidence layer, families, buyers, and lenders must make high-value decisions with asymmetric information. The engineering problem is not simply digitizing an appraisal form; it is building a reproducible model from sparse and uneven inputs.
Three steps for the user: upload the title deed (French or English formats supported), confirm the parcel location, receive a full Equiterrain valuation with key drivers, a confidence score, and the underlying analysis.
Under the hood, the engine applies established economic-gravity models — Reilly's Law of Retail Gravitation and the Huff Probability Model — to quantify how proximity to urban centers, infrastructure, and economic activity shapes land value at any given coordinate. Equiterrain runs on top of Bwendi for location context, then layers title analysis and market benchmarks to produce a reproducible, auditable estimate.
The ambition is straightforward: become the reference land valuation platform for the African continent — one honest estimate at a time. Starting in Cameroon, expanding from there.
Equiterrain demonstrates how OSIH approaches African data systems: make uncertainty visible, combine multiple weak signals into an auditable model, and build around local economic structure instead of waiting for a complete imported dataset. The same approach can support regional pricing, asset intelligence, logistics planning, and other decisions under sparse data.