Sell-Through Rate and Assortment KPIs: What to Track and How

Sell-through rate is the share of the stock you received that sold over a fixed window: units sold ÷ units received × 100. It is a warehouse metric, calculated per product from purchase and sales records. In ad data, the obvious stand-in is the share of advertised products with a sale. It mostly counts your orders. To judge an advertised catalogue, track the share of revenue from new products, the share of budget on products with no sales, and how fast your tail of rarely ordered products renews.

What is sell-through rate?

Sell-through rate tells you how much of what you bought has sold. Wikipedia (Sell-through) defines it as the percentage of a product that a retailer sells after its supplier ships it, usually counted over one month. Shopify’s guide (Sell-Through Rate (STR): How to Calculate & Improve It (2026)) puts it the same way: inventory sold against the units received from manufacturers over the same window.

Sell-through rate is one of the numbers behind assortment planning, the decision about what to carry, what to add and what to keep. It covers the buying side of that decision: whether a product, a size or a collection sold the stock you ordered for it. Shopify adds that it helps you compare how styles and sizes of the same product perform against each other.

How do you calculate sell-through rate?

Sell-through rate = units sold ÷ units received × 100

Lightspeed’s guide (Sell Through Rate: Definition, Formula, and Importance) gives this formula. Shopify writes it as total sales divided by stock on hand. Both guides count units, not money, in their examples. Shopify applies the rate to one product or one collection at a time; Lightspeed lets you calculate it by product type, category or brand.

The two guides differ on the bottom of the fraction. Lightspeed divides by units received. Shopify’s definition also uses units received. Its monthly calculation, though, divides by all the stock available for sale that month, which can include stock left over from before. Either works, as long as you pick one and stick to it.

1. Choose the window. Lightspeed says most retailers calculate it every 30 days. Shopify adds that you can also use a week, a quarter or a year. Judge seasonal products over their selling season.
2. Take the base for each product or size: units received in the window, or all units available for sale, whichever rule you picked.
3. Take units sold in the same window, across every channel that sells from that stock.
4. Divide and multiply by 100.
5. Compare like with like: the same product type, the same length of window, the same season. Shopify’s guide makes the same point.

Example store, not client data.

A tableware store received 400 units of a new tea and coffee collection in September and sold 260 of them by the end of the month. The collection’s sell-through rate is 260 ÷ 400 × 100 = 65%. Product by product, it looks less even:

The 65% average hides a teapot that sold fewer than one in four of the units received. Read sell-through per product and per size first, and only then for the whole collection.

What is a good sell-through rate?

There is no single good number, and the published ones we checked come without a sample behind them.

Neither guide says which stores or how many products these figures come from, and Shopify calls its category figures illustrative. Read them as the view of two retail software vendors: rules of thumb, not a market benchmark.

The bigger issue is the window. Shopify’s own category examples show it. Fragrance sells through about 23% after 8 weeks and 63% after a year. Home improvement sells through about 55% after 8 weeks and 90% within a year. The longer you wait, the higher the rate for the same product. So compare a product with itself, or with products of the same type, over the same window.

Our data comes from ad accounts and Merchant Center, where stock shows up only as an availability flag, not as units received. That’s why we publish no sell-through benchmark of our own.

How is sell-through rate different from inventory turnover?

Sell-through looks at one product or collection over a short window. Inventory turnover looks at the whole stock over a longer one. In Shopify’s guide, turnover shows how many times you sold and replaced all your products, usually over a full year. For a store that sells through Google Ads, a third number belongs in the comparison: the stand-in for sell-through built from ad data.

The turnover formula comes from Shopify’s guide Inventory Turnover Ratio: Formula and How to Improve. Use sell-through for buying decisions on single products and turnover for the health of the whole stock. Both need data from your store platform or warehouse system. The third row looks like a shortcut for a store that sells through Google Ads, but it misleads.

The ad-data version of sell-through mostly counts your orders

Your ad account knows which products got a conversion, so the obvious stand-in for sell-through is “products with at least one sale ÷ products advertised”. We tested that number on GetProfit data from June 2025 to June 2026. It failed two checks.

It recounts orders. Across 1,514 store-months, the number of products with at least one sale rose and fell with the number of orders. The rank correlation was +0.912, where 1 would mean that whenever one rises, so does the other. Month to month, changes in the share tracked changes in orders (+0.748) more closely than changes in revenue (+0.622). More orders mean more products with a sale, whatever happens to the assortment.

It changes with the window. In 114 stores, the median share of advertised products with a sale was 5.0% over one month and 26.5% over 13 months. Same catalogues, a 5.3× gap. So compare your store’s monthly figure only with other monthly figures.

Example store, not client data.

The tableware store shows 1,200 products in its ads. In September, its 300 orders included 240 different products, so its ad-data sell-through was 240 ÷ 1,200 = 20%. In October, orders rose to 375 with the same products, prices and ads, and included 288 products: 24%. The metric rose by 4 points while the assortment stayed the same. Over 12 months, 636 of the 1,200 products sold at least once, so the yearly figure was 53%. All three numbers describe one catalogue.

The rule we took from this applies to any share you track. If the top of the fraction is a result and the bottom is nearly the same result, the share will move with almost anything. Before you trust a share, ask two questions. Does it follow your order count? Does it change when you change the window length?

Which assortment KPIs show whether your catalogue works?

The figures in this section come from GetProfit data for June 2025 – June 2026. They are observations from stores we work with, not an experiment. They describe how the median store behaved, not a target you must hit.

Share of revenue from new products

In our analysis of 100 stores over June 2025 – June 2026, we compared the first three months of the window with the last three. In stores that doubled revenue, 47.3% of revenue in the last three months came from products that weren’t there in the first three. In stores that declined, the figure was 26.1%. The median across all 100 stores was 35.7%.

The link is modest (rank correlation +0.208, n = 100), but it held whether stores cut, held or raised spend. It also passed a check for the trap that caught the ad-data sell-through: a share that only restates the result. In stores that doubled, revenue from old products grew too, to 126% of the starting level. New products added revenue on top, so the higher share isn’t an artefact of the arithmetic.

What counted was how much revenue the new products went on to earn. The number of new products a store tested was not linked to growth on its own.

Share of budget on products with no sales

We looked at GetProfit data on 114 stores with at least 8 months of history (June 2025 – June 2026). The median store spent 18.8% of its budget on products with no sales in the window. For the middle half of stores, the figure ranged from 9.5% to 36.7%.

This share depends on the window: the longer it is, the more products get a sale. Some of this spend is the cost of testing, because a product needs clicks before you can judge it. That’s why the share stays above zero.

Products that spend noticeable money and never pay it back fall into what we call spend with nothing back. Track your own figure month by month over a fixed window, and compare it only with your own earlier months. Our article on Performance Max spend on unsold products covers how much of this share is normal and why Google keeps spending there.

Tail renewal

We call products with one or two orders in a month the tail. In GetProfit data on 116 stores (June 2025 – June 2026), the tail made up 64.9% of selling products and 57.0% of revenue in the median store-month. It changes fast. A median 75.0% of this month’s tail products didn’t sell the month before. For the head, products with six or more orders, it was 12.5%.

So read the tail as a flow. A product that sold once last month and not this month can still be fine: most of the tail changes every month. Judge your head products month by month and your tail over a longer window.

These KPIs sit next to store-wide numbers like revenue, ROAS and conversion rate. Our guide to ecommerce KPIs covers which of those explain your Google Ads results.

How to set up assortment KPIs in an advertised store

1. Keep stock metrics and ad metrics apart. Calculate sell-through and inventory turnover in your store platform or warehouse system, per product and per size. Take spend and sales per product from your ad data. Don’t build one from the other.
2. Set the windows before you look at the numbers. Use a month for products you sell all year and the selling season for seasonal ones. Keep the length the same every time, and never compare a monthly figure with a yearly one.
3. Track the three ad-data KPIs every month: share of revenue from new products, share of budget on products with no sales, and tail renewal.
4. Run the two-question check on any new share before you report it: does it follow your order count, and does it change with the window?
5. Put sell-through into your feed. Give slow-selling products a custom label value such as “low sell-through”. You can then group or exclude them in campaigns. Google’s Merchant API reference (ProductAttributes) describes custom labels 0–4 as fields for custom grouping of items in a Shopping campaign. The Google Ads API guide Listing Groups for Retail explains that Performance Max listing groups include or exclude products in each asset group. It also notes that Custom Attribute in Google Ads equals the custom label in Merchant Center.
6. Check what you don’t carry. Google’s Best sellers data (BestSellersProductClusterView, Merchant API) ranks product clusters by country and category, based on estimated units sold. Google marks each cluster as in stock in your product data, out of stock, or not in your inventory at all. The “not in inventory” rows in your categories are an assortment KPI no stock report gives you.
7. Review it all once a month, in a fixed order. The monthly assortment review checklist lists what to check.

The catalogue from feed to sales, on a fixed window

The portal’s Assortment section walks the whole catalogue through the stages from the feed to sales. The stages are: in the catalogue, eligible for ads, showing in ads, spending budget, bringing conversions, and paying off at target efficiency. The portal counts impressions, spend and sales over 90 days. The section also lines up the biggest product groups by share of budget, each next to its payback against a benchmark.

The account score works the same way. Products are one of its four areas. The portal computes the score on a fixed 90-day window, whatever period you pick on screen. The fixed window makes this month’s figure comparable with last month’s. The portal reads Google Ads and Merchant Center. Sell-through and inventory turnover come from your store platform or warehouse system, so they stay your own calculation.

See which part of your catalogue works. The portal walks the whole catalogue through the stages — from the feed to sales, shows where the budget goes by product group, and answers whether you really have a season. The portal changes nothing without your consent.

Check your catalogue →

Sources

• Sell-through (Wikipedia, last edited 28 June 2025) — definition of sell-through; calculated over a period, usually one month. Checked 2 October 2026.
• Sell-Through Rate (STR): How to Calculate & Improve It (2026) (Shopify, 3 October 2025) — definition against units received; formula as total sales ÷ stock on hand; monthly calculation on all stock available for sale; 70–80%, above 80%, 40–60% and “at or above 80%” targets; illustrative category figures for fragrance and home improvement; like-for-like periods; sell-through vs inventory turnover. Checked 2 October 2026.
• Sell Through Rate: Definition, Formula, and Importance (Lightspeed, 7 May 2026) — formula as units sold ÷ units received × 100; calculation every 30 days; 60–80% healthy, above 80% and below 40% readings, 75% in 30 days. Checked 2 October 2026.
• Inventory Turnover Ratio: Formula and How to Improve (Shopify, 9 June 2023) — inventory turnover formula and average inventory. Checked 2 October 2026.
• ProductAttributes (Google Merchant API reference, last updated 23 September 2026) — custom labels 0–4 for custom grouping of items in a Shopping campaign. Checked 2 October 2026.
• Listing Groups for Retail (Google Ads API, last updated 30 September 2026) — listing groups include or exclude products per asset group; Custom Attribute equals the Merchant Center custom label. Checked 2 October 2026.
• BestSellersProductClusterView (Google Merchant API reference, last updated 1 July 2026) — best sellers ranking by estimated units sold; in stock, out of stock and not in inventory status. Checked 2 October 2026.
• GetProfit data: 114 stores, June 2025 – June 2026 — share of advertised products with a sale over one month and over 13 months; share of budget on products with no sales. 1,514 store-months — products with a sale against orders.
• GetProfit data: 100 stores, June 2025 – June 2026 — share of revenue from new products by growth group; link held across spend bands; old-product revenue in stores that doubled.
• GetProfit data: 116 stores, June 2025 – June 2026 — tail share of selling products and revenue; tail and head renewal.