About Fairyard

Fairyard was built to answer a deceptively simple question: what will this specific pen sell for next week?

MLA publishes after-the-fact averages by category and weight bracket. Useful, but backward-looking. They don't tell a producer what their 100 Dorper lambs at 23 kg fat-3 are likely to clear at Wagga next Thursday — and that's the question that matters when you're deciding whether to take an OTH offer or send the mob to auction.

We pull MLA NLRS data continuously, plus FX, weather, slaughter statistics, fuel costs, ENSO indicators and a few other macro signals. We train an ensemble of quantile models with Mondrian-stratified conformal calibration, and surface the result as a calibrated price band per head.

Every pen gets a confidence tier — HIGH, MEDIUM or LOW — based on sample size, staleness and bracket width. Sparse cells widen automatically. When the market is in regime transition we surface a warning so you know the model is in catch-up mode.

We're small. We're independent. We don't take money from processors or stock agents. We publish a weekly accuracy scorecard so you can see how close we got — win, lose or draw.

If you'd like to talk before you sign up, call or email. No sales pitch.