Plan further ahead
19% lowerforecast error
See six months ahead.
Lower error than using last month’s sales to forecast six months ahead.
Explainable AI inventory planning
See Demand Before
You Commit to More Stock
Review seasonal demand, stock coverage, and incoming supply in one place. Use Demand ESP to inform your next reorder, then use Sparki in app or your own AI assistant over MCP to prepare a draft for your review.
Agentic onboarding · recommendations with reasoning · your team approves
Put inventory to work
12% lessstock held
Hold less. Keep orders moving.
Less inventory than last-month forecasting, at the same 98% fill rate.
Get value sooner
Under 1hour to forecast
Bring your files. Start planning.
From export files to a working forecast with Sparki or your own AI assistant.
Powered by Demand ESP
Every SKU. Its own forecast.
Demand ESP reads each item’s demand pattern and automatically selects a forecasting method. See the method chosen and the history behind your forecast.
Forecast and inventory results from public-data testing.
How we measured it
The 19% and 12% figures were measured in September 2026 on an earlier version of our forecasting engine.
A method for each demand pattern
Demand ESP selects statistical methods for steady, trending, seasonal, and intermittent demand, using the available sales history and fallbacks for limited data.
What the 12% measures
In an inventory simulation using 400 Iowa public-data items, average on-hand inventory cost was approximately $4.12M for Spark Inventory versus $4.67M for last-month forecasting at an interpolated 98% fill rate. Both used the same order-up-to policy, a 14-day lead time, lost sales, and a uniform $12 unit cost.
What the 19% measures
Retrospective testing on 799 Iowa items found six-month pooled weighted absolute percentage error (WAPE) of 22.0% for Spark Inventory versus 27.1% for last-month forecasting, a relative reduction of approximately 19%.
What onboarding time measures
Measured from customer export files to a working forecast. Setup time varies with data readiness and scope.
Connect the commerce and accounting tools you already use
.svg)
Decision-ready forecasting
Every reorder starts
with the evidence
See the sales, stock, lead-time, and policy signals behind every recommendation, then decide what your team wants Spark to do next.
Inventory risk engine
See the stockout before itbecomes a fire drill
Spark turns the signals already inside your business into an explainable risk window and a purchasing decision your team can approve.
SKU-level risk windows
Reasoning attached
Human-approved action
REPLENISHMENT / RISK TRACE
LIVE MODELSTEP 01
Unify demand and supply signals
Bring sales history, stock, incoming supply, and lead times into one planning view.
INPUT SIGNALS
Sales history
On-hand + incoming
Lead times
SIGNAL COVERAGE
Demand + supply
Explainable inventory control
Move from “low stock” to a decision your buyer can defend
Spark connects the demand evidence, supply position, timing, and purchasing workflow behind every replenishment recommendation.
Detect the change
Compare recent velocity with historical demand, current stock, incoming supply, and lead times to surface the risk window early.
Demand analysis, forecast, and stock-insight tools
Show the reasoning
See the demand and inventory evidence behind each recommendation.
Demand evidence and readiness gates
Approve the action
Preview quantities and supplier context, adjust when judgment is needed, and approve the purchase action with a decision record.
Governed preview, approval, and PO lifecycle
Decision trace
Every recommendation carries its evidence forward
01 / RISK
Projected coverage gap
Timing by SKU and location
02 / EVIDENCE
Velocity + lead time + supply
Inputs remain visible
03 / ACTION
Draft purchasing plan
Reviewed before execution
The replenishment operating layer
Connect the early warning to the purchase decision
Stockout prevention works when demand changes, current coverage, incoming supply, supplier timing, and the approved response live in one traceable workflow.
Detect the coverage gap
Combine on-hand inventory, recent velocity, forecast demand, incoming supply, and lead time to see when each SKU becomes exposed.
Risk has a date and a cause
Build the buying decision
Move beyond a low-stock alert to a proposed quantity, supplier context, timing, and the evidence that changed the recommendation.
Alert becomes a plan
Protect cash from the opposite error
Evaluate the stockout risk alongside overproduction, carrying cost, order cadence, and the inventory already on the way.
Availability and cash considered together
Keep people in control
Let Spark prepare the action while buyers review exceptions, adjust judgment calls, and approve what actually reaches the supplier.
Human-approved execution
A reorder point is a trigger. A replenishment plan is a decision.
Static thresholds can tell a buyer that stock is low. Spark connects the changing evidence needed to decide what to do next.
| Decision layer | Static reorder workflow | Spark replenishment workflow |
|---|---|---|
| Demand | Fixed average or manual forecast | Recent velocity and forecast context remain visible |
| Supply | Lead time entered once | On-hand, incoming, supplier, and timing considered together |
| Output | Low-stock alert or reorder point | Reviewable quantity and timing with evidence |
| Execution | Buyer rebuilds the PO manually | Draft purchasing action moves through approval |
Integrations
Connects to everything you already use
Bring storefronts, marketplaces, accounting, payments, fulfillment, and more into one operating view. Start with guided setup and keep every system current with real-time sync.
See All Integrations
Forecasting FAQ
Know what the recommendation is built on
Clear answers about the signal, the approval model, historical data, and how quickly the first plan becomes useful.
How does Spark predict a stockout?
Spark combines SKU-level sales history, changing velocity, current and incoming stock, supplier lead times, and inventory policy. It shows the risk window and the reasoning behind the recommended action.
Does the AI place purchase orders automatically?
Spark prepares the recommendation and draft purchase order. Your team can review the evidence, change quantities, and approve before anything is sent or committed.
Can Spark help with overstock as well as stockouts?
Yes. The same demand and supply model identifies inventory accumulating faster than expected, changing velocity, and cash tied up in stock that is unlikely to move on the current plan.
How much sales history should we bring?
Bring the reliable history you have. Twelve months or more is ideal for a stronger seasonal signal, and Spark can use up to the latest 36 months during onboarding.
How quickly can we see our first inventory plan?
Sparki or your own AI assistant over MCP can inspect, map, repair, and validate the data you already have. Once the forecast foundation is ready, Spark can surface the first planning recommendations without a traditional migration project.
See the next stockout before it happens
Bring your actual sales and inventory signal. Spark will show the risk, the recommendation, and the reasoning your team can review.
Free for one user, refreshed monthly · human-approved actions