Forecast.Buy plan.Replenish.
AI inventory replenishment, demand forecasting and budget-driven buy planning for retailers.
Create your free account
Pick a module to start. Free and anonymous, no credit card required.
Platform: One platform, three modules
Know your cost in under 30 seconds
Enter your store count and SKU catalog size for an instant monthly and yearly estimate — no forms, no sales call required.
What is buffers.ai?
The Platform
buffers.ai is a real-time AI platform that automates demand forecasting, purchase planning, and inventory replenishment for retail and manufacturing enterprises, built from three focused modules: Forecasting for AI demand prediction, Replenishment for AI inventory replenishment, and Buy Plan for budget-driven, periodic purchase planning. It is used by global retail and manufacturing enterprises including H&M, Bath & Body Works, Victoria's Secret, P&G, and Toshiba.
Why It Matters
Inaccurate demand forecasts and slow replenishment cause two costly outcomes at once: stockouts that lose sales, and excess inventory that ties up working capital. buffers.aiaddresses both by improving forecast accuracy per SKU and automating the resulting purchase and replenishment decisions, so safety stock reflects real, current demand rather than a static rule.
How It Works
Forecasting backtests five algorithms (TimesFM, LightGBM, Prophet, Exponential Smoothing, and Moving Average) over a rolling six-month window, selects the closest-fitting champion model per SKU, and converts that forecast into a supply plan accounting for lead time, safety stock, minimum order quantities, budget, and open commitments. Replenishment then manages day-to-day store and channel inventory: restocking, assortment, new product introduction, promotional push, omnichannel balancing, and KPI reporting. For products with no sales history — or for franchisees and multi-banner retailers working to a periodic buying budget — Buy Plan generates category-and-month purchase quantities directly from that budget, independent of the forecasting pipeline above.
Our Solutions
Three focused modules: Forecasting for backtest-driven demand prediction, Buy Plan for budget-driven, periodic purchase planning, and Replenishment for day-to-day store operations
AI Demand Forecasting (Forecasting)
The forecasting engine: backtests five algorithms to find a champion model per SKU, then turns that forecast into a supply plan.
Backtest & Champion Selection
Race 5 forecasting engines, crown the best fit per SKU
Every SKU's sales history runs through five forecasting engines — TimesFM (Google), LightGBM (Microsoft), Prophet (Meta), Exponential Smoothing, and Moving Average — backtested over a rolling six-month window on data none of them has seen. Each model is scored on WAPE, bias, safety stock, and service level, and the closest fit is crowned the champion for that SKU.
- Function
- 5-engine backtest: TimesFM, LightGBM, Prophet, Exponential Smoothing, Moving Average
- Optimizes for
- Forecast accuracy per SKU, scored on WAPE, bias, safety stock, service level
Blended Demand Build
Build the demand signal from the champion model
Once a champion algorithm is selected per SKU, buffers.ai builds a blended demand forecast from its output — the single demand number that feeds every downstream buffer, replenishment, and supply calculation.
- Function
- Demand signal construction from each SKU's champion forecasting model
- Optimizes for
- Forecast accuracy feeding the supply calculation
Supply Calculation
Turn the demand forecast into a purchase order
Converts the blended demand forecast into a purchase recommendation, accounting for lead time, safety stock, minimum order quantities, budget limits, and open commitments.
- Function
- Lead time, safety stock, minimum order quantity, budget, and commitments applied to the blended forecast
- Optimizes for
- Right-sized purchase orders within budget and commitment constraints
Promotion Uplift
Measure each promotion's real lift, then forecast the next one with it
Every past promotion is measured on each SKU's own sales: average daily demand during the event against a clean baseline of the four weeks before and after it, with other event days left out and repeat runs of the same event pooled. A future promotion points at a comparable past one and inherits its measured lift, so a brand-new promo is forecast from real history instead of a guess. Prophet reads the events directly as holidays; the other engines get the measured lift applied to the forecast on the promotion days. One-off events nobody planned are smoothed out of the training data so they don't skew the model.
Measured: last spring's 30%-off sale
Applied: this summer's sale, same 7 days
- Measured lift
- 82 / 50 − 1 = +64%
- Applied to the forecast
- 55 × 1.64 = 90.2/day
- Over the 7 days
- 385 → 631 units
- Function
- Per-SKU promotion lift measured from history and applied to future promotion days
- Optimizes for
- Forecast accuracy through promotions, without one-off spikes distorting the baseline
Cold-Start Forecast
Forecast a brand-new SKU before it has sold a single unit
The five forecasting engines need a SKU's own history, so a new item would otherwise get no forecast at all. Instead it is matched to products that share its attributes, starting with the most specific group that has at least eight of them and widening only when it has to. Its demand level is the median of that group, shaped by the group's own weekday, month-of-year, and launch-ramp patterns. As the new SKU starts selling, its own sales take a growing share of the forecast.
Find similar products (needs at least 8)
- Running sneakers, $90-1305 SKUs, too few
- Sneakers23 SKUs, used
- Footwear140 SKUs, not needed
Blend with its own sales so far
- Weight on its own sales
- 14 / (14 + 21) = 40%
- Forecast level
- 0.4 × 6.0 + 0.6 × 4.2 = 4.92/day
- Before its first sale
- 4.2/day, from similar products alone
- Function
- Analog-based forecast for new and thin-history SKUs, blended with their own sales as they accrue
- Optimizes for
- A usable launch forecast and PO quantity from day one
Monthly Buy Plan (Buy Plan)
Budget-driven, periodic purchase planning — built for franchisees and multi-banner retailers who need every category and month to land inside a set buying budget.
Monthly Buy Plan
Hit a periodic buying budget, every category, every month
Turns a monthly or seasonal buying budget into a 12-month purchase plan by category and month — in units and dollars — built for franchisees and multi-banner retailers who must stay within a periodic budget rather than buy freehand. The plan can be viewed by week, month, or quarter, with every total re-summed to match. It also works with zero sales history, since most of what gets bought each period is new. Holidays borrow their lift from a comparable past event, promotions are sized from their discount (see Promotion Impact), and every number breaks down into base rate, seasonality, holiday impact, and promotion impact.
- Function
- Category-by-month unit and dollar purchase planning against a set budget
- Optimizes for
- Franchise/multi-banner budget compliance, buy quantities without sales history
Promotion Impact
See what a discount does to future sales before you buy for it
Add a future promotion to a population (a category, brand, or SKU list) with its discount and dates, and the buy plan lifts that population's months to match. The lift comes from the best evidence available: a number you pin yourself, a named past event you choose to copy, past promotions on the same population at a similar discount, or, when there is no history, a price-elasticity curve on the discount. The curve counts only the markdown on top of what the population already sells at, and the lift is spread over the days the promotion actually covers in each month.
- Lift on promo days
- (1 / 0.70)1.5 − 1 = +70.7%
- Spread over the month
- +70.7% × 14/30 = +33.0%
- Already selling at 10% off?
- +45.8% per day, 1,214 units
- Function
- Discount-driven demand lift per population, prorated into each month of the plan
- Optimizes for
- Buying enough stock for a promotion without overbuying the months around it
New Store Demand
Size the lift a new store adds to the whole chain
Opening a store adds demand permanently, not for a few weeks like a promotion. Enter the store's opening stock buy and the plan sizes the lift by comparing it with the stock the chain already holds, then phases it in over the store's first 13 weeks and keeps it from then on. The same lift feeds the SKU forecasts, and it updates automatically as chain stock changes. You can also pin the lift yourself when stock data isn't available.
- Extra demand at full size
- 1.56 / (52 × 1.3) = +2.31%
- Phase-in
- Straight line over 13 weeks, then held
- On 10,000 units/week
- +231 units every week after
- Function
- Permanent, phased-in chain demand lift from a new store's opening stock value
- Optimizes for
- Buying for a store opening without guessing its effect on the rest of the plan
Set a Budget, See It Everywhere at Once
Change the total and every category, month, and banner rescales instantly — no rebuild, no waiting.
Plan From Day One
Works before a single unit of a new item has sold.
One Plan, Every Banner
The same category structure runs per banner, so head office can hold every location to its own share of the budget.
AI Inventory Replenishment (Replenishment)
Nine features that manage day-to-day store and channel inventory: daily dynamic buffers, restocking, assortment, new product launches, new store openings, time events and promotions, promotional push, omnichannel balancing, and KPI reporting.
Dynamic Buffer
Recalculate the optimal stock level every day
Calculates the optimal stock level for every product at every store on a daily basis, so targets follow the latest demand instead of a fixed level set months ago.
- Function
- Daily optimal stock-level calculation per product and store
- Optimizes for
- Availability with the least inventory
Replenishment
Restock the right products at the right time
Sends the right amount to each store within local and global constraints, so shelves stay stocked without tying up capital in excess inventory.
- Function
- Store-level restocking under local and global constraints
- Optimizes for
- Stockout reduction, inventory availability
Assortment
Adjust your assortment dynamically
Adjusts your product assortment as sales data comes in, surfacing best sellers and matching the mix to what customers are actually buying. Also used to set the initial assortment when a new store opens.
- Function
- Real-time, data-driven product-mix adjustment per store
- Optimizes for
- Best-seller identification, new-store product fit
New Product Introduction
Find the right amount to launch new products
Sizes and distributes inventory for new product launches using sales forecasts, so you avoid over- or under-stocking on day one. Available as a rule-based or AI-based configuration.
- Function
- Launch inventory sizing and distribution, rule-based or AI-based
- Optimizes for
- Forecast-driven launch risk reduction
New Store
Decide the opening quantities for a new store
Helps decide how much of each product to send when a new store opens, before the store has any sales history of its own.
- Function
- Opening-quantity planning for new stores
- Optimizes for
- Right-sized opening stock
Time Events & Promotions
Plan stock around holidays, seasons and promotions
Adjusts store stock levels for time-based events such as holidays and seasons, and for planned promotions, raising inventory ahead of the event and bringing it back down once it ends.
- Function
- Stock-level adjustment for calendar events and promotions
- Optimizes for
- Event demand coverage without leftover stock
Push
Let the merch team increase inventory when needed
Lets your merchandising team push a one-time stock increase ahead of a promotion or event, covering the demand spike without carrying the extra inventory afterward.
- Function
- One-time manual stock increase for promotions or events
- Optimizes for
- Peak-demand coverage without overstocking
Omnichannel/Online
Balance inventory across online sales, physical stores and wholesale
Splits inventory between online, physical stores and wholesale customers according to your policy, so no channel runs short while another sits overstocked. Wholesalers are supported as a channel alongside your own stores and online sales.
- Function
- Inventory allocation between online, physical stores and wholesalers
- Optimizes for
- Cross-channel availability, policy compliance
Advanced Dashboard
Global KPIs dashboard for fast decision making
One dashboard surfacing KPIs from manufacturing, engineering, procurement, and supply chain, so managers can decide from current data instead of a weekly report.
- Function
- Real-time KPI reporting across manufacturing, engineering, procurement, and supply chain
- Optimizes for
- Decision speed
Trusted by Industry Leaders
How buffers.ai provides automated supply chain replenishment and assortment tools for global retail and manufacturing enterprises including H&M, P&G, Toshiba, Bath & Body Works, Victoria's Secret, and Strauss.
From consulting floor to automation platform
buffers.ai was established in 2015, growing out of a consulting practice that spent years inside the inventory and demand-planning operations of retail and manufacturing companies — from luxury and fast fashion to footwear, grocery, consumer goods, and manufacturing operations like Toshiba's.
The goal from day one was a holistic supply chain solution, not another siloed reporting tool: one system carrying an operational action all the way from demand planning to supply, first allocation for new products, and day-to-day store replenishment. We built it as a call-to-action system — one that outputs the next order, not a BI system that hands you a report and leaves the decision to someone else.
That work kept surfacing the same problem: stockouts and excess inventory weren't a data problem, they were a process problem — replenishment decisions run on static rules and spreadsheets that couldn't keep pace with real, day-to-day demand. Rather than solve it one client engagement at a time, we built the automation platform that consulting work kept trying to become.
buffers.ai is that platform: years of hands-on inventory and forecasting expertise, encoded into software that makes the call every day instead of a consultant revisiting it every quarter.
Results, Backed by Case Studies
Metrics reported by buffers.ai customers, sourced from published case studies
Bath & Body Works
Delta Israel Brands Ltd. — deployed across European stores and online channels
- ~70%
- Replenishment process automated
- 2 mo
- Initial deployment time
- 2
- Markets live (Israel, Germany)
- ~90%
- Target automation
“Among the systems currently in use, buffers.ai has proven to be the strongest performer in supporting our operational needs.”Read the Bath & Body Works case study
RAI
Grocery retailer managing 6,000–7,000 SKUs
- 95%
- Replenishment process optimized
- 2–3 hrs
- Saved per store manager, daily
- ~5%
- Stockout rate reduction (8–10% → ~4.5%)
- 1–2 mo
- Expected full ROI