Forecast what sells.
Plan ahead.
ForecastEdge turns your sales history into a product-by-product demand forecast: what will sell next, in what quantity, and when to plan your next order.
1,240 units ± confidence band
Your next stock decision should not live in a spreadsheet.
Small product teams are balancing cash, lead times, and customer demand with incomplete information. ForecastEdge shows which products are likely to sell next and the quantity to plan for.
Cash tied up in slow movers
Order to the demand you actually expect, with a confidence range that makes your risk visible.
Stockouts on best sellers
See which products are trending toward a stockout early enough to reorder while demand is still there.
Safety stock by instinct
Replace “just in case” with a range that shows how much buffer is enough for the pattern you sell.
From sales data to a buy plan.
ForecastEdge turns messy order history into three decisions your team can act on this week.
Connect your data
Upload a sales export or connect your store. Your data stays yours throughout the process.
See what will sell
The model learns trend, seasonality, and recent demand to build a forward view for every SKU.
Plan what to buy next
Get product-level quantities, timing, and buffer ranges so you can make the right stock decision at the right time.
Useful now. Honest about uncertainty.
ForecastEdge is designed to help you make a better next decision, without pretending forecasts are magic.
Transaction revenue prediction
Best current model: CatBoost. Latest benchmark run delivered strong error metrics on transaction-level prediction and currently powers the MVP workflow.
Future revenue forecasting
Time-series performance is improving but not final yet. We are expanding features and evaluating with chronological and walk-forward validation.
Safer rollout approach
Phase 1: deploy transaction prediction. Phase 2: release forecasting as a labeled beta with confidence ranges and clear model limitations.
Built with an evidence-first modeling process
- Model benchmarking across Random Forest, XGBoost, CatBoost, and SARIMAX rather than assuming one algorithm is always best.
- Leakage-aware design separates transaction prediction from true future-revenue forecasting to avoid misleading accuracy claims.
- Chronological and walk-forward validation to simulate real forecasting behavior and avoid random-split bias.
- Transparent uncertainty reporting so decisions are made with risk context, not just a single number.
MVP transaction model currently leads our internal comparison set.
Future-revenue forecasting remains in pilot while we improve time-series feature engineering and validation.
Built around the stock decisions that often go wrong
Three common moments where a real demand forecast changes what you'd otherwise plan on instinct.
Pre-season stock planning
Know how much of each seasonal SKU to order before the season starts, with a confidence range that tells you how much buffer to actually carry.
Avoiding stockouts on top SKUs
Get an early warning when a fast-moving product is trending toward a stockout, in time to reorder before you lose sales.
First-order sizing for new SKUs
Estimate initial order quantities for new products using demand patterns from comparable items in your catalog.
How ForecastEdge helps, in practice
A few examples of the kind of decisions ForecastEdge is built to support.
Instead of guessing how much of a best-selling item to reorder, a confidence range shows how much buffer stock actually makes sense.
Ahead of a busy season, a per-SKU demand forecast replaces a rough spreadsheet estimate for how much stock to bring in.
Launching a new product, first-order quantity is estimated from demand patterns of comparable items already in the catalog.
Reducing overstock on slow-moving items frees up cash that would otherwise sit tied up in unsold inventory.
Pricing that grows with your Nigerian brand.
Start small, get a clear demand forecast, and upgrade when your catalog and sales channels grow.
For small catalogs replacing guesswork and spreadsheets.
- Up to 10 SKUs
- 1 forecast plan
- Shopify or CSV data import
- Email support
For growing brands selling across multiple channels.
- Unlimited SKUs
- Unlimited forecast plans
- Wholesale, retail, and B2B channels
- Shopify, CSV, and manual imports
- Priority support
For larger teams with complex catalogs or multiple warehouses.
- Unlimited forecasts
- Daily refresh
- Custom model tuning
- Dedicated analyst support
Why we built this
Most inventory tools are built by engineers who bolt on a basic reorder-point formula as an afterthought. ForecastEdge started from the opposite direction: real statistical forecasting methods first, wrapped in something a small e-commerce store can actually use without a data science degree.
We believe every small retailer deserves the same demand-forecasting rigor as a large enterprise — without the headcount that usually comes with it.
Ifeoma Maryanne Ndibe
Statistician who upskilled into data science, combining classical forecasting methods with modern machine learning to bring real demand-forecasting rigor to small e-commerce retailers who've never had access to it before.
Ready to stop guessing?
Tell us what you sell and where your data lives. We will show you the clearest path to forecasting your next products and quantities.