Running an ecommerce store creates a deceptively simple inventory question:

How much product should you have on hand?

Too little inventory and popular products can sell out before you have time to replenish them. Too much, and cash that could be used for marketing, product development, or other parts of the business ends up sitting on a shelf.

Inventory planning is the discipline of finding a better balance.

In this guide, we'll cover the fundamentals of inventory planning—from forecasting demand and calculating reorder points to understanding the metrics that can tell you whether your inventory is healthy.

1. Introduction: What is inventory planning?

Inventory planning is the process of determining what products you need, how much you need, and when you need them.

For an ecommerce merchant, that usually means combining several pieces of information:

• What you've sold in the past

• What you currently have in stock

• How quickly each product is selling

• How long suppliers take to replenish inventory

• How much demand you expect in the future

• How much buffer inventory you want to maintain

• How much cash you can reasonably commit to purchasing

The objective isn't simply to keep as much inventory as possible. It's to maintain enough stock to meet expected customer demand without unnecessarily tying up capital in products that may take months to sell.

Inventory forecasting is one part of that process: it uses historical sales, seasonality, trends, and other demand signals to estimate future stock needs. Replenishment then turns that forecast into decisions about when and how much to reorder.

Why inventory planning matters for cash flow

Inventory is an asset, but it's also cash that has been converted into products.

Imagine you have $30,000 available for your next purchasing cycle.

You could buy:

$30,000 of inventory because you have the cash available.

Or you could determine:

Which products are most likely to sell, which are at risk of running out, and how much inventory you actually need—and invest $18,000 accordingly.

The remaining $12,000 stays available for other business needs.

That is one of the fundamental benefits of inventory planning.

Overstock ties up working capital and creates carrying costs. Products can also become outdated, damaged, discounted, or eventually written off.

At the other extreme, understocking can lead to stockouts, missed sales, frustrated customers, and lost momentum on popular products.

Good inventory planning therefore isn't about minimizing inventory.

It's about having the right inventory at the right time.

Understanding the different types of inventory

Depending on your business model, inventory can appear in several forms.

For most Shopify merchants selling finished products, the most relevant is finished-goods inventory: products that are ready to sell to customers.

Manufacturers may also manage:

Raw materials — components or materials used to create products.

Work in progress (WIP) — products that have entered production but aren't finished.

Finished goods — completed products available for sale.

You may also encounter inventory that is operationally important rather than directly sold, such as packaging and maintenance supplies.

Within sellable inventory, it is useful to think in another dimension as well:

Cycle stock is the inventory you expect to sell under normal conditions.

Safety stock is extra inventory kept as a buffer against uncertainty.

Excess inventory is inventory beyond what you're reasonably expecting to need.

The distinction between safety stock and excess stock is particularly important. Safety stock serves a deliberate purpose: protecting the business against unexpected demand or supplier delays. Excess inventory generally represents capital that isn't being used efficiently.

2. Key concepts and methods

Inventory planning begins with a forecast.

You don't need a sophisticated machine-learning model to start. For many growing merchants, a useful first forecast can come from understanding their own historical sales.

Start with historical demand

There's clearly some growth.

Simply using the average of all four months may underestimate what happens next because recent demand is higher than older demand.

On the other hand, assuming the growth will continue indefinitely could lead you to over-order.

That's why forecasting is not simply calculating an average. You want to look for patterns in the data.

Start by asking:

What is the normal sales rate?

Calculate average daily or weekly sales.

Is demand increasing or decreasing?

Compare recent periods with previous periods.

Is the product seasonal?

A swimsuit that sells 20 units per day in July may sell very differently in January.

Were there unusual events?

A promotion, influencer mention, stockout, product launch, or unusually large order can distort historical sales.

Historical data gives you a baseline, but a useful forecast should also consider seasonality, planned promotions, launches, and other known changes in demand.

Common forecasting approaches

There are many forecasting models, but a small ecommerce business doesn't need to begin with the most complicated one.

Simple moving average

Take sales from a defined number of recent periods and calculate the average.

For example:

Last 30 days: 300 units sold
Average daily sales: 300 ÷ 30 = 10 units/day

It's easy to understand and works reasonably well for products with stable demand.

Its weakness is that it reacts slowly when demand is changing rapidly.

Weighted moving average

Give more importance to recent sales.

For example, instead of treating the previous three months equally, you might weight them:

Most recent month: 50%
Previous month: 30%
Third month: 20%

This allows the forecast to respond more quickly to changing demand.

Seasonal forecasting

Compare the product with the same or comparable periods in previous years.

This can be especially useful for businesses selling products influenced by holidays, seasons, back-to-school periods, or other recurring events.

The important principle is:

The forecasting method should fit the product's demand pattern.

You don't necessarily need the same model for every SKU.

Lead time: the missing piece

Knowing how quickly something sells isn't enough.

You also need to know how long it takes to replace it.

If you sell 10 units per day and your supplier needs 14 days to deliver a new order, you could sell approximately:

10 × 14 = 140 units

while waiting for the shipment.

This is called lead-time demand. Shopify similarly defines it as expected demand during the period required for replenishment.

That brings us to two of the most useful concepts in inventory planning.

Safety stock

Forecasts are estimates. Real life rarely follows them perfectly.

A product expected to sell 10 units per day might suddenly sell 16. A supplier expecting to deliver in 10 days might take 15.

Safety stock creates a buffer against those uncertainties.

One common formula is:

Safety Stock = (Maximum Daily Sales × Maximum Lead Time) − (Average Daily Sales × Average Lead Time)

Suppose:

Average daily sales = 10 units
Maximum daily sales = 14 units
Average supplier lead time = 10 days
Maximum lead time = 14 days

Then:

(14 × 14) − (10 × 10)
= 196 − 100
= 96 units of safety stock

This isn't the only safety-stock method. More sophisticated approaches account statistically for demand variability and desired service levels. But the basic principle is the same: maintain a deliberate buffer rather than adding inventory based purely on instinct.

Reorder point

Once you know expected demand during lead time and the safety stock you want to maintain, you can calculate your reorder point (ROP).

A reorder point is the inventory level that tells you:

It's time to place another order.

The standard formula is:

Reorder Point = (Average Daily Sales × Lead Time) + Safety Stock

Suppose:

Average daily sales = 10 units
Supplier lead time = 14 days
Safety stock = 40 units

Your reorder point is:

(10 × 14) + 40 = 180 units

When available inventory approaches 180 units, it's time to consider placing the next order.

The 180 units aren't necessarily what you should order. They're the level at which you should act. Determining order quantity is a separate purchasing decision.

That distinction—when to order versus how much to order—is one of the most useful ideas to understand when moving beyond basic inventory tracking.

3. Essential inventory metrics

A merchant can have thousands of data points available and still struggle to make a purchasing decision.

The goal isn't to track everything. Start with metrics that answer practical questions.

Average daily sales / sales velocity

What it tells you: How quickly a product is selling.

A basic calculation is:

Average Daily Sales = Units Sold ÷ Number of Days

If you sold 450 units over 30 days:

450 ÷ 30 = 15 units/day

Be careful with periods in which the product was out of stock. Zero sales on those days may reflect a lack of inventory rather than a lack of customer demand.

Days of inventory / days of supply

What it tells you: Approximately how long your current inventory will last.

A simple calculation is:

Days of Supply = Available Inventory ÷ Average Daily Demand

If you have 300 units and sell approximately 15 per day:

300 ÷ 15 = 20 days of supply

This is often more intuitive than simply knowing you have “300 units.”

Three hundred units might be plenty for a product selling five units per day and dangerously low for one selling 50.

Lead time

What it tells you: How long replenishment takes.

Measure the time from placing an order until the inventory is actually available to sell.

Don't assume the supplier's quoted lead time is always the real lead time. Tracking actual delivery performance can reveal whether a supplier routinely takes longer than expected.

Reorder point

What it tells you: When you should start replenishment.

As covered earlier:

ROP = Lead-Time Demand + Safety Stock

Reorder points should be reviewed as demand and supplier performance change. A threshold calculated six months ago may no longer make sense for a rapidly growing product.

Safety stock

What it tells you: How much buffer you've intentionally maintained against uncertainty.

More safety stock reduces stockout risk, but it also increases the amount of cash committed to inventory.

There is no universally “correct” safety-stock number. It is partly a business decision about the tradeoff between availability and carrying cost.

Sell-through rate

What it tells you: How much of received inventory actually sells during a period.

A product with strong sell-through may justify faster replenishment, while persistently weak sell-through can be an early signal of overstock.

Shopify identifies sell-through alongside inventory turnover and days on hand as useful measures of how effectively inventory is moving.

Inventory turnover

What it tells you: How frequently inventory is sold and replaced.

A common financial formula is:

Inventory Turnover = Cost of Goods Sold ÷ Average Inventory Value

Higher turnover generally means inventory is moving efficiently, although what's “good” depends heavily on product category and business model.

Stockout risk

What it tells you: Which products may run out before they can be replenished.

Rather than thinking of stockout risk as a single universal formula, consider several signals together:

• Current available inventory

• Forecast demand

• Days of supply

• Supplier lead time

• Incoming inventory

• Safety stock

For example, a product with 12 days of supply and a 21-day supplier lead time deserves attention even though it isn't technically out of stock yet.

Overstock risk

What it tells you: Where you may be holding significantly more inventory than expected demand requires.

Look for combinations such as:

• High days of supply

• Falling sales velocity

• Weak sell-through

• Large quantities on hand

• No expected seasonal increase

Overstock isn't simply “a lot of units.” It is too much inventory relative to expected demand.

That's an important distinction.

4. Best practices and tools

Once the fundamentals are in place, good inventory planning becomes a repeatable process rather than a one-time calculation.

Keep your inventory data accurate

Forecasting software cannot compensate for fundamentally incorrect inventory data.

Regularly reconcile actual stock with recorded quantities and investigate discrepancies caused by returns, damaged goods, transfers, fulfillment errors, or other adjustments. Physical inventory counts remain useful for verifying that system records reflect reality.

Plan at the product or SKU level

A store might look healthy overall while individual products are severely overstocked or approaching stockouts.

Inventory decisions should therefore generally be made at the SKU level—and, when relevant, by location.

Don't treat every product equally

Your best-selling product and a slow-moving accessory probably shouldn't receive identical safety-stock policies.

Products can be prioritized according to factors such as:

• Sales volume

• Revenue contribution

• Margin

• Lead time

• Demand stability

• Business importance

ABC analysis is one established approach for segmenting inventory so that the products with the greatest business impact receive closer attention.

Refresh forecasts regularly

Demand changes.

A forecast created six months ago shouldn't silently drive today's purchasing decisions.

For many stores, reviewing forecasts monthly is a reasonable starting point, with more frequent reviews for fast-moving products or volatile periods. Shopify likewise recommends refreshing forecasts and considering promotions and external demand signals rather than relying indefinitely on historical averages.

Account for promotions and seasonality

Historical sales aren't always representative of what happens next.

Before accepting a forecast, ask:

Are we running a promotion?
Is a holiday approaching?
Is this product seasonal?
Did a previous stockout suppress historical sales?
Has the supplier's lead time changed?

Forecasting should support judgment, not replace it.

Separate recommendations from decisions

Inventory-planning software may recommend purchasing 200 units.

That doesn't automatically mean 200 units is the right business decision.

You might know that:

• A replacement product launches next month.

• The supplier has a minimum order quantity.

• A promotion is scheduled.

• Cash is needed elsewhere.

• The product is being discontinued.

Good planning tools should help merchants understand why an action is recommended while leaving room for business context.

Graduate from spreadsheets when they stop helping

Spreadsheets are perfectly reasonable when a catalog is small.

A simple sheet containing:

SKU → On hand → Daily sales → Lead time → Safety stock → Reorder point

can be enough to establish a useful planning discipline.

The challenge appears as catalogs, locations, suppliers, and purchasing frequency grow. Calculations need to be refreshed repeatedly, historical data becomes cumbersome, and identifying which products deserve attention can become time-consuming.

Inventory-management applications can automate data collection, low-stock alerts, forecasting, and related planning processes.

The right progression is often:

Manual tracking → Repeatable spreadsheet → Automated inventory planning

rather than adopting the most sophisticated system on day one.

Bringing it all together

Inventory planning ultimately comes down to a handful of questions:

What are customers likely to buy?
Use historical demand and forecasting.

How much inventory do I have?
Maintain accurate inventory records.

How long will it last?
Monitor sales velocity and days of supply.

When should I reorder?
Combine lead-time demand with safety stock to establish reorder points.

How much should I purchase?
Consider expected demand, current and incoming inventory, target coverage, supplier constraints, and available cash.

You don't need perfect forecasts to improve inventory decisions.

Start with accurate sales and inventory data, establish a few consistent metrics, create sensible reorder rules, and review them regularly. As your store grows, those basic practices provide the foundation for more sophisticated forecasting and purchasing.

The goal isn't to predict the future perfectly.

It's to make better inventory decisions with the information you have today.