← All posts

Replenishment

Weeks of supply formula, and how to convert it from days of stock

Weeks of supply is days of stock in a coarser unit. The arithmetic is one line. What changes with the unit is how people use the number, and that is the useful part.

Marcus VolstedCo-founder, Restocio Published 10 October 202611 min read

Every source on this page was opened and read, and each one is dated in the Sources section at the foot.

Weeks of supply is units on hand divided by the units you sell in an average week. With 480 units on the shelf and sales of 42 a week, you have 11.4 weeks of supply. It is the same number as days of stock in a coarser unit, which is why a merchandise planner and a Shopify report can describe one shelf and appear to disagree. This page is the conversion, and the reading of it. How many weeks you should hold is a separate question with its own page.

What is weeks of supply?

How long your current stock lasts at the current pace, counted in weeks. Retail buyers and merchandise planners have used it as their working unit for a long time, mostly because their calendar runs in weeks. Deliveries land weekly and a season is a count of weeks rather than days. When a retail partner or an investor asks what your weeks of supply is on a line, that is the dialect they are speaking, and they usually mean it as a target as much as a measurement. "We run six weeks" is a policy. "We have 41 days" is a reading.

A Shopify store meets the same number in days. Shopify's own inventory report gives days, and so does Restocio. Most spreadsheets do too, because sales arrive as daily rows. Nothing about the arithmetic changes between them. Seven days is one week, and the only real difference is what each unit tempts you to do with the figure, which is where this page spends most of its words.

How do you calculate weeks of supply?

weeks of supply = units on hand ÷ average units sold per week

The weekly rate is the part with choices in it. Take a product with 480 units on the shelf that sold 336 units over the last eight weeks. That is 42 a week, and 480 divided by 42 is 11.4 weeks of supply. The same fact stated daily is 336 over 56 days, 6 a day, and 480 divided by 6 is 80 days, which is 11.4 weeks again.

Count the divisor in days the product was actually in stock, then multiply by seven, rather than counting weeks on a calendar. Suppose that product was sold out for 14 of the 56 days. Its honest pace is 336 over 42 in-stock days, 8 a day, 56 a week, and the shelf holds 8.6 weeks rather than 11.4. Nearly three weeks of difference on one product, and in the direction that flatters a product which keeps running out. The mechanism, and why it lands on your best products, is written up under what breaks every forecasting formula, so it is not repeated here.

Use the window you would use for a daily rate. Eight weeks is 56 days, close to the 60 that suits most stable products. Four weeks reacts faster and is jumpier, and for anything selling under a unit a day it holds so few sales that the weekly figure is mostly noise. A product moving four a week shows 16 sales in a four-week window, and one good weekend moves its weeks of supply by about a fifth.

Days of stock, weeks of supply and months of inventory are one number

Every version of this figure is stock divided by pace. What changes is the unit the pace is stated in, and different tools pick different units without saying so.

Divide days by seven for weeks and by 30.4 for months. The months column is where the public benchmarks live, since the US Census Bureau publishes inventories divided by monthly sales for the whole retail sector, and it is also the column most likely to be misread, which is the next section.

Two tools you may already have state the figure in days. Shopify's inventory reports, read on 27 September 2026, include one called Inventory remaining per product, defined as the quantity still in inventory divided by the average sold per day, with that average taken over the last 28 days, and a sibling called Inventory sold daily by product, which divides the quantity sold by the number of days in the period. Restocio shows the same figure in days, as Stock lasts on the Replenishment page and Days left on the Stock status page, on a 60-day window by default and counted over days the product was recorded in stock. Neither tool shows weeks. Divide by seven and you have your weeks of supply from either one.

What is a good weeks of supply?

There is one public benchmark worth knowing, and it is worth knowing mostly so you can see why it cannot be borrowed. The US Census Bureau's Manufacturing and Trade Inventories and Sales release for July 2026, published on 16 September 2026, puts the seasonally adjusted inventories/sales ratio for all US retail trade at 1.27, meaning 1.27 months of inventory against one month of sales. The same release breaks it down by kind of business. Converted to weeks:

Read it for the spread rather than the values. Grocery runs about a third of what clothing runs, and both are ordinary. Then notice what the ratio is made of. The Census values inventories at cost and sales at what customers paid, net of refunds, so a clothing store selling at twice what it pays holds, in units, roughly double the 9.1 weeks the ratio suggests. It is a dollar ratio, for whole sectors of businesses with employees, and it says nothing about any one product. A store that runs six weeks as a policy still has products sitting at two and at thirty.

The honest answer to what a good weeks of supply is comes from your lead time, not from a table. A number that keeps a grocer comfortable would empty an importer's shelf while the boat is still at sea. How to set the level, delivery time plus a cushion plus the length of one order cycle, is the subject of the days of stock page, and the result converts to weeks the same way everything else here does.

Read the number against your lead time, in the same unit

A weeks of supply figure is only useful next to a second figure in weeks, the time it takes to get more. Retail buyers do this comparison instinctively, which is most of why the unit persists. Ten weeks on the shelf against a two-week supplier is comfortable. Ten weeks against a ten-week boat is the last day you can order at sea freight prices, and if the factory only starts cutting after your deposit clears, it is already late.

Convert the lead time first, then compare. A 70-day sea passage is ten weeks. Restocio's cushion, the stock left when the boat arrives, is 60 days by default and adds another 8.6, so on those defaults an importer's order fires at about 18.6 weeks of supply, 130 days, before any production days at the supplier are counted, with the shelf still looking full. Which figure is the trigger, and how safety stock enters it, is worked through on the reorder point page. The countdown that turns it into a date is in the Shopify replenishment guide.

Trailing or forward, which weekly sales do you divide by?

Every report gives you a trailing figure, stock divided by what sold over the last few weeks. A merchandise planner going into a season often means something else by the same words, stock divided by what the coming weeks are expected to sell, and on a seasonal line the two can be very different. Eleven weeks of supply on a winter jacket in August, measured against August sales, is not the same shelf as eleven weeks measured against what October will do.

The forward version is only as good as the forecast under it, and for most products there is no forecast better than the trailing pace, so trailing is the right default. It goes wrong on products with a real calendar, and there the fix is to divide by the same weeks last year rather than the last few weeks this year. That is what Restocio does for products marked seasonal, where the rate in the run-up to the season is last year's same-season pace, falling back to the trailing pace when there is no last season to read. How much history each kind of estimate needs is in the forecasting guide.

Where the weekly average quietly goes wrong

Weeks are a coarse unit, and coarseness hides things. Count sales by week and a product that was empty from Wednesday to Sunday still shows a full selling week in the divisor. Count in days, drop the days with zero recorded stock, then multiply by seven, and the empty days stop voting. It is the same correction the daily rate needs, only easier to forget when the spreadsheet is laid out in weekly columns.

Short windows are the other trap. Four weeks is popular because it feels current, and it is fine for a product selling dozens a day. For the long tail of a catalogue it is far too few sales to average. Eight weeks is a better default for most products, and for genuinely slow lines the window should stretch until it holds enough sales to mean something, though never further back than the product has existed.

A campaign week belongs outside the window, or inside it on purpose. A weekly average that includes a 30 percent off weekend describes a different product at a different price, and the weeks of supply it produces will look thin for as long as that weekend stays in the window. Both distortions are covered properly in the forecasting formulas post, so take them as given.

What Restocio shows, and what it does not

Days, not weeks. The Replenishment page shows each product's stock in units and as Stock lasts, in days, which is units on hand across the SKU's active variants divided by its average daily sales. That average is measured over the last 60 days by default, or over the product's own age when it is younger than that, and it counts only the days the product was recorded in stock rather than days on the calendar. The deadline shown with it is that figure minus the delivery time, the cushion and the supplier's production days, shown as a countdown.

There is no weeks of supply column and no toggle to switch the unit, so if your buyer or your board thinks in weeks, you divide by seven yourself. We chose days because the deadline is a date and a date is counted in days, and because an importer's numbers are large enough that weeks would round away the last few days before a boat has to be booked. That is a real limitation for anyone reporting upward in weeks, and it is stated here so it does not surprise you on the first day.

Common questions

What is the weeks of supply formula?

Units on hand divided by average units sold per week. With 480 units on the shelf and sales of 42 a week, weeks of supply is 11.4. Measure the weekly rate over days the product was actually in stock and multiply by seven, rather than counting calendar weeks, or a product that keeps selling out will look better covered than it is.

How do you convert days of stock to weeks of supply?

Divide by seven. Sixty days of stock is 8.6 weeks of supply, 90 days is 12.9 weeks and 130 days is 18.6 weeks. For months, divide days by 30.4 or weeks by 4.33. The three figures are one measurement in different units, so a report showing days and a buyer talking in weeks are not disagreeing.

What is a good weeks of supply?

The only public benchmark is sector-wide. The US Census Bureau's July 2026 release puts US retail inventories at 1.27 months of sales, about 5.5 weeks, ranging from 0.77 months at food and beverage stores to 2.63 at department stores. Those are dollar ratios for whole sectors, with stock valued at cost and sales at retail, so they understate unit cover and say nothing about a single product. Your own good number is your lead time in weeks plus a cushion, which is far higher for an importer than for a store with a local supplier.

Does Shopify show weeks of supply?

Not in weeks. Shopify's inventory reports include Inventory remaining per product, which divides the quantity still in stock by the average sold per day over the last 28 days, and Inventory sold daily by product, which divides by the days in the period. Both are in days, and neither says it leaves out the days a product was sold out. Divide the days figure by seven for weeks of supply.

Is weeks of supply the same as weeks of cover or weeks on hand?

In everyday use, yes. Weeks of cover, weeks on hand, weeks of stock and weeks of supply all mean stock divided by weekly sales. The distinction worth keeping is trailing against forward: a report divides by what sold in recent weeks, while a merchandise planner ahead of a season may divide by what the coming weeks are forecast to sell, and the two can differ a lot on seasonal products.

Is weeks of supply the same as inventory turnover?

No, though they are related. Turnover is how many times a year the whole stock is sold and replaced, calculated from cost of goods sold against average inventory value. Weeks of supply is one product's stock against its weekly sales, right now. You can convert a turnover ratio into an average cover, 52 divided by turnover gives weeks, but that average hides the products sitting at two weeks and at thirty.

Sources

Every source below was opened and read on the date shown against it. Rules, rates and schedules change, so check the current page before you act on a number that matters. If anything here is out of date or wrong, tell us and we will correct it.

• US Census Bureau, Manufacturing and Trade Inventories and Sales, July 2026 (release CB26-154, 16 September 2026), Table 3 (read 27 September 2026)
• US Census Bureau, Monthly Retail Trade definitions (inventories valued at cost, the inventories/sales ratio) (read 27 September 2026)
• Shopify Help Center, Inventory reports (read 27 September 2026)

Marcus Volsted

Co-founder, Restocio

Marcus co-founded Restocio and works on it daily with a Swedish importer who plans their purchasing in it every working day. Restocio is built in Sweden by two founders, one Swedish and one Danish, and Marcus is the Danish one. Why we are building it.

Related reading

• How many days of stock should you hold
• Inventory forecasting formulas, and which one to use when
• Inventory forecasting, without the mystique
• Shopify replenishment, end to end
• Reorder point formula, with a worked example
• Inventory turnover and sell-through, and which one tells you something