In the age of Amazon, allocation and replenishment decisions can make or break retailers and distributors. Profit margins and customer loyalty depend upon making the most optimal tradeoffs every day. Due to many variables, however, getting the balance right can feel as unlikely as winning the lottery. Even replenishing fast-moving commodity items can be complex and highly variable when you try to forecast demand by SKU, by store, by day. When allocating new or one-time buys and factoring in promotions and other demand-shaping variables, the number of possible trade-offs are practically infinite.
Spreadsheets and legacy planning systems that rely on aggregated sales histories can’t handle this complexity, never mind cope with today’s multi-channel, seasonal and ‘fast-fashion’ planning challenges. But ToolsGroup applies advanced algorithms and AI to provide highly accurate, time-phased allocation and replenishment plans that achieve target service levels.
The complexity, time pressures and educated guesswork involved in merchandise planning invariably leads to either too much safety stock or unmet service levels. Our replenishment and allocation solutions take out the guesswork, recommending optimal ship-to quantities down to the store level. Forecasts are continually updated on a rolling, time-phased basis, factoring in changing demand signals (pricing, promotions, seasonality) and supply constraints (lead times, supplier minimums, capacity issues). The result: planners meet their service level targets without building excess inventory in the supply chain.
For items with a sales history, our replenishment solution sets optimal inventory levels for every location in your network – down to the store level – to achieve service targets. A time-phased forecast is updated on a daily, rolling basis in response to changing variables like promotions, seasonality and supply constraints. It uses an advanced statistical method that ensures the locations with the highest probability of stocking out get served first with inventory. All these capabilities ensure planners always stay on the ‘front foot’.
For one-time buys and new products that have no sales history, our allocation solution calculates optimal inventory levels using advanced statistical methods. “Clustering” adjusts the replenishment plan depending on store properties (store size/type, footfall, overall sales volume). “Supersession” analyzes the sales history of similar items in a category to help calculate optimal inventory levels. “Holdback” recommends reserve levels at the DC level to be allocated as-needed to stores. These advanced capabilities support retailers’ decisions to intelligently re-allocate stock, prevent unnecessary markdowns due to poor stock allocation, and plan alternative fulfillment (e.g., letting customers order online in store and have products shipped home).
Our AI capabilities improve allocation for new production introduction (NPI) by analyzing the attributes of a new item, comparing against previously launched products with similar attributes and predicting sales by location.
Our demand sensing capabilities enable planners to be even more proactive when planning items with short lifecycles. By factoring in point of sale data, our solution generates an early warning replenishment signal that identifies which items are ‘runners’ and which are ‘laggards’. This gives planners the advance warning they need to determine whether to hold back more (or less) inventory upstream.