Few retail categories carry as much SKU complexity as tires. The same customer need - "I need tires for my car" - fans out into dozens of combinations of size and brand, many of which only apply to a narrow set of vehicles. Layer seasonality on top of that - the predictable, hard surge in demand as winter approaches, and again at the spring changeover - and purchasing becomes a genuinely difficult forecasting problem, not a simple reorder-when-low calculation.

Before Fospertise, this tire shop was making purchasing decisions reactively - ordering based on recent sales patterns and experience rather than a forward-looking view of demand. That approach has a structural weakness in a seasonal, SKU-heavy business: by the time a stockout on a popular size or brand becomes visible, the seasonal surge that caused it is often already underway, and lead times mean the shop is playing catch-up right when customer demand is at its peak. At the same time, slower-moving sizes and brands were accumulating in inventory, tying up capital and shelf space without a clear, data-driven way to identify which SKUs actually needed trimming back.

The result was a familiar and costly pattern for seasonal retail: stockouts on exactly the items customers wanted most, right when they wanted them, alongside overstock sitting unsold in the categories that mattered least.

Generic inventory forecasting tools tend to work at an aggregate level - total units, total revenue - which misses the point for a tire business, where the real decision is which specific size-and-brand combination to stock ahead of a specific seasonal window. The shop needed a partner who could build forecasting at that level of granularity, grounded in an understanding of how tire demand actually behaves through the year in the Canadian market - and one willing to combine more than one forecasting technique rather than relying on a single model to carry the whole problem.

Fospertise's approach started with the purchasing and inventory team - understanding which stockouts had hurt the most, which SKUs habitually sat unsold, and how the shop's own seasonal patterns played out - before building the forecasting system around that reality.