WeatherVane helps retail brands forecast weekly demand, understand why it moved, and know when to push or pause spend.
cross-brand training
WeatherVane helps retail brands make better weekly demand decisions.
It is a weekly operating workflow for the person deciding what to stock, when to spend, and whether a demand move came from weather, promotion, competition, or baseline demand.
The product does three things: forecast next week, explain the driver, and tell the operator whether to push, pause, or wait.
Brands still make high-stakes weekly calls with bad causal visibility.
They can usually see the result, but not the reason. So they mistake weather for marketing, promotion for baseline demand, and short-term noise for a lasting trend.
We already have benchmark proof, partner data, and a weekly product story.
A model trained on pooled data outperforms a single-brand baseline by +16% on held-out demand prediction. Tested on M5, the world’s largest retail forecasting benchmark: roughly 100,000 Walmart SKU-level time series.
For a $10M brand, a cleaner weekly read on demand means materially less overbuying and less wasted paid spend.
What the buyer gets every week.
SKU-level demand with uncertainty, refreshed weekly
What’s driving demand: weather, promos, baseline, competition
When to act and when to wait
This is valuable because the model has learned across brands, not because it produced another generic dashboard.
Why now: the modeling works, the data is reachable, and the pain is weekly. Moat: a pooled corpus that compounds over time.
Start with weather-sensitive retail where timing mistakes are frequent and expensive.
The first buyer is a marketing, demand, or inventory lead who already feels the cost of making a weekly call with bad signal. Legacy planning suites and generic forecast APIs solve different problems.
Business model: free design partners first, then software subscription for the weekly workflow, with later data/API upside only after the core product is proven.
Co-founded Homesick Candles (8-figure exit to WIN Brands). Owns NorSari and leads its marketing (44% CAGR over 8 years). Built WeatherVane’s causal ML architecture and benchmark proof. Solo founder today; first hires are ML/data and commercial.