Spark Inventory Blog

A spreadsheet can be a perfectly reasonable place to start planning inventory. The problem with spreadsheet inventory forecasting is not that spreadsheets are inherently wrong. It is that the workbook usually stops matching the operating reality once you have enough SKUs, channels, suppliers, and incoming inventory to manage.

At that point, the weekly planning file becomes a fragile manual process. Someone exports sales, updates receipts, adjusts a forecast, checks stock, and tries to calculate what needs to be ordered. The output may look precise, but it can still miss the question that matters: what should we buy now, from which supplier, to arrive before we run out without tying up unnecessary working capital?

For a product business selling across direct, marketplace, and wholesale channels, forecasting must lead to a buying decision. Here is where spreadsheets begin to fail, what a usable planning process requires, and how to decide whether your current file is still doing its job.

What spreadsheet inventory forecasting does well

A well-built spreadsheet is useful for narrow, stable planning problems. If you have a limited SKU count, one warehouse, predictable supplier lead times, and a person who owns the file, it can provide a helpful weekly view of on-hand stock, recent sales, and expected replenishment needs.

It is also flexible. An experienced operator can add a one-time promotion adjustment, correct an unusual sales spike, or model a supplier delay without waiting for system changes. That flexibility is valuable, particularly when the business is still establishing its inventory policies.

The trade-off is that flexibility often depends on individual knowledge. When formulas, overrides, and assumptions live in one person's workbook, the process is difficult to audit and even harder to repeat. A cell reference can break quietly. An import can be pasted into the wrong tab. A new SKU can be excluded from a lookup. The risk is not merely a bad report. It is an order that arrives late, or an order that is too large to justify.

The point where the workbook becomes a risk

Most teams do not abandon spreadsheets because they dislike them. They outgrow them when the number of changing inputs makes manual reconciliation unreliable.

Demand is the first issue. A trailing 30-day average may be adequate for a steady SKU, but it can distort a product with seasonality, a recent launch, a stockout period, or channel-specific demand. When one SKU sells through Shopify, Amazon, wholesale, and retail, a planner needs a forecast that distinguishes actual demand from sales constrained by unavailable stock.

Supply is the second issue. On-hand inventory alone is not a useful reorder signal when purchase orders are already in transit, partially received, delayed, or allocated to another warehouse. A buyer needs to see projected inventory over time: current available stock, incoming inventory by expected receipt date, forecast demand, and the date when stock risk begins.

Then there is timing. A reorder point only works when it reflects demand during the supplier lead time, not an old average entered when the supplier relationship began. If the supplier needs 45 days to produce and deliver, and the SKU sells faster than expected, the order decision has to happen before the stockout is visible in a basic stock report.

A spreadsheet can technically handle each of these factors. In practice, it handles them only if someone continuously validates data, maintains formulas, and updates assumptions. That is why a planning file often becomes more labor-intensive as inventory complexity increases, even when the team is trying to save time.

A buying plan needs four connected decisions

An inventory forecast is useful only when it supports a buying plan. For every SKU that needs attention, the planner should be able to answer four connected questions: how much demand is expected, when supply will arrive, how long current stock will last, and what action is required.

Forecast demand at the SKU level

The forecast should be per SKU and based on a clear demand history. That does not mean blindly extending recent sales. It means accounting for the conditions that shaped those sales. If a product was out of stock for 10 days, its recorded sales may understate true demand. If it had a temporary promotion, its recent volume may overstate the baseline.

The right forecast horizon depends on the buying cycle. A domestic supplier with a short lead time may require a near-term view. A supplier with production, freight, and receiving time may require several months of projected demand. The key is consistency: use a forecast that is appropriate to the replenishment decision, then allow the operator to review and adjust an exception when business context requires it.

Calculate days of supply using projected inventory

Days of supply should reflect more than the quantity physically in the warehouse. It should consider available inventory, committed demand where relevant, incoming inventory, and forecasted consumption.

For example, a SKU with 600 units on hand and forecast demand of 20 units per day appears to have 30 days of supply. But if 300 units are already allocated to wholesale orders, the usable position is closer to 15 days. If 500 units are due to arrive in 12 days, that incoming receipt changes the reorder recommendation. If the receipt is delayed, it changes again.

This is why a static days-of-supply column often creates false confidence. The planner needs a dated inventory projection, not just a ratio calculated from today's stock.

Apply supplier lead time and inventory policy

A reorder recommendation should combine the projected stockout date with the supplier lead time and a defined inventory policy. The policy may include safety stock, a target coverage period, minimum order quantities, case packs, and supplier-specific order schedules.

There is no universal safety stock number. A high-margin SKU with volatile demand may justify more buffer than a slow-moving accessory with a long holding cost. Likewise, a supplier with inconsistent arrival dates may require a different buffer than a reliable domestic supplier. The point is to make the policy explicit rather than burying it in an unexplained spreadsheet formula.

Convert the recommendation into an executable purchase decision

A buyer should not have to reconstruct an order from a risk report. The planning process should group reorder recommendations by supplier, account for minimums and order constraints, and present a clear proposed quantity and required order date.

That is the operating gap between reporting and inventory execution. A report tells you a SKU is low. A buying plan tells you what to buy, when to buy it, and why that quantity supports the policy.

Where manual spreadsheet processes usually fail

The common failure is not poor math. It is a gap between the data used in the model and the data used to run the business.

A planner may update sales weekly but receive inventory daily. Purchasing may track open orders in email or a separate system. Warehouse transfers may be missing from the planning file. One channel may report sales differently from another. By the time the workbook is updated, the team is working from a snapshot that no longer represents the current inventory position.

Manual exception handling compounds the problem. A buyer may know that a supplier is late, that a product is being discontinued, or that an inbound container is split across shipments. Those adjustments are sensible, but they need to be visible to the next person reviewing the plan. Otherwise, the workbook becomes a collection of decisions without an operating record.

The cost appears in both directions. Stockouts lose sales and disrupt customer commitments. Excess inventory consumes warehouse space and working capital, then pressures the team to discount product that was purchased on an outdated assumption.

Moving beyond spreadsheet inventory forecasting

The practical goal is not to remove operator judgment. It is to remove repetitive reconciliation so judgment can focus on exceptions and trade-offs.

A stronger process starts by bringing sales history, on-hand stock, incoming inventory, supplier lead times, and purchasing data into one planning view. Data should be validated before it drives recommendations. If an incoming purchase order has no expected receipt date, or a SKU has a mismatched identifier across systems, that should be visible as a data issue rather than silently absorbed by the forecast.

From there, the system should maintain a per-SKU forecast, identify stock risk and excess-inventory risk, and turn the result into a buying plan. The buyer can then review the recommendations by supplier, adjust quantities for commercial context, and approve the final decision.

Spark Inventory is designed around that workflow. It can use supported connections and CSV or XLSX imports to combine the inventory signals operators already manage. It prepares reorder recommendations and draft purchase orders, while the operator reviews, adjusts, and approves before anything is sent or committed.

That distinction matters. Inventory planning should be disciplined, but it should not ignore informed human judgment. A system can flag that a SKU will fall below safety stock before the next receipt arrives. The buyer may still decide to buy less because a replacement product is launching, or buy more because a supplier has a temporary order cutoff. The decision is better when the underlying forecast and inventory position are clear.

A simple test for your current process

Your spreadsheet may still be adequate if it can reliably answer, for every replenished SKU, what your projected stock position will be on each expected receipt date and which purchase orders must be placed this week. It should also show the assumptions behind those answers.

If getting there requires several exports, manual copy-paste work, formula checks, and a meeting to determine what the report means, the spreadsheet is no longer just a tool. It is a dependency.

The next step is not necessarily a heavyweight ERP implementation. Start by establishing a trusted monthly per-SKU forecast and a buying plan that turns demand and supply data into clear reorder decisions. Once the team can see stock risk early and act from the same inventory position, planning becomes less about maintaining a file and more about making the right call before the window closes.