Out-of-Stock Statistics
A sourced roundup of stockout statistics: how often shoppers hit out-of-stock products, what they do next, how much revenue it costs, and how to measure back-in-stock recovery on your own store.
Out-of-Stock Statistics
Data on how often shoppers find products unavailable, how they react, and how much revenue stockouts cost retailers.
Out-of-stock (OOS) statistics measure three things: the stockout rate (the share of products or SKUs unavailable when a shopper wants them), the shopper response (switch brand, switch size, wait, buy from another retailer or abandon), and the financial loss (sales that never come back).
Most of the best-known figures come from large retail studies, such as the global meta-analysis by Gruen, Corsten and Bharadwaj, and from IHL Group's inventory distortion reports. For an online store they work as a floor, not a ceiling. On the web, a competitor is one tab away, so the 'buy elsewhere' response costs you more than it does in a physical aisle.
Key statistics
• Out-of-stocks were projected to cost retailers $1.2 trillion in lost sales worldwide in 2023 (IHL Group, 2023). • Total inventory distortion (out-of-stocks plus overstocks) was projected at $1.77 trillion in 2023 (IHL Group, 2023). • Overstocks added a further $562 billion in losses in 2023 (IHL Group, 2023). • Inventory distortion losses fell by $172 billion between 2022 and 2023 (IHL Group, 2023). • The average worldwide retail out-of-stock rate is 8.3% (Gruen, Corsten & Bharadwaj, 2002, 40 studies). • 31% of shoppers who hit a stockout buy the item from another retailer (Gruen, Corsten & Bharadwaj, 2002). • 9% of shoppers who hit a stockout buy nothing at all (Gruen, Corsten & Bharadwaj, 2002). • A typical retailer loses about 4% of sales to out-of-stocks (Gruen, Corsten & Bharadwaj, 2002). • Shoppers saw more than 2 billion online out-of-stock messages in October 2021 (Adobe Digital Economy Index, 2021). • 75% of US consumers tried a new shopping behaviour during the pandemic, with product availability among the top reasons for switching brands (McKinsey, 2020).
These numbers frame the scale, but the figure that matters for your P&L is narrower: how many product page sessions land on an unavailable size, shade or variant, and how many of those sessions end without an order. The sections below give you the public benchmarks first, then show how to translate them into your own revenue at risk.
How often shoppers hit out-of-stock products
Out-of-stock frequency and causes: headline benchmarks
| Metric | Figure | Scope | Year |
|---|---|---|---|
| Average out-of-stock rate, worldwide | 8.3% | Meta-analysis of 40 retail studies | 2002 |
| Average out-of-stock rate, United States | 7.9% | US retail studies | 2002 |
| Average out-of-stock rate, Europe | 8.6% | European retail studies | 2002 |
| Average out-of-stock rate, other regions | 8.2% | Studies outside US and Europe | 2002 |
| Stockouts caused by store-level ordering, forecasting and replenishment | 70–75% | Root-cause share across studies | 2002 |
| Online out-of-stock messages seen by shoppers | 2 billion+ | US e-commerce, October 2021 | 2021 |
| Change in online out-of-stock messages vs pre-pandemic | +172% | US e-commerce, Oct 2021 vs Jan 2020 | 2021 |
The 8.3% figure is two decades old and comes mostly from grocery and consumer goods. It is still the most-cited baseline because nobody has replicated the study at the same scale. Adobe's 2021 data shows how badly online availability can slip during supply shocks: out-of-stock messages almost tripled compared with early 2020.
For apparel and footwear stores, a product-level rate understates the problem. A jacket can be 'in stock' while the M and L sizes are gone, and those sizes carry most of the demand. Measure availability at variant level, weighted by traffic. A 5% SKU stockout rate can easily mean 15% or more of PDP sessions see a sold-out size.
What shoppers do when a product is out of stock
Shopper response to a stockout (share of shoppers)
Only the 19% who substitute within the same brand are a clean save for a single-brand store. If you sell only your own products, the 26% who switch to a different brand are lost to you. Add them to the 31% who buy elsewhere and the 9% who give up, and roughly two-thirds of stockout encounters are at risk.
The 15% who delay are the audience for back-in-stock alerts. You only keep them if you capture an email or SMS opt-in on the product page and restock before they forget. McKinsey's 2020 research showed how loose loyalty becomes when availability fails: three in four US consumers tried new shopping behaviours, and availability was among the top reasons for trying a new brand.
These reactions were measured in physical stores, where switching retailer means walking out. Online, switching is a Google search or a marketplace tab. So expect your 'buy elsewhere' share to sit at or above the 31% baseline, especially for branded electronics and beauty SKUs that marketplaces also stock.
Lost sales: what stockouts cost retailers
The financial cost of out-of-stocks
| Metric | Figure | Scope | Year |
|---|---|---|---|
| Lost sales from out-of-stocks | $1.2 trillion | Global retail | 2023 |
| Losses from overstocks (markdowns, spoilage) | $562 billion | Global retail | 2023 |
| Total inventory distortion | $1.77 trillion | Global retail | 2023 |
| Year-on-year change in inventory distortion | −$172 billion | Global retail, 2022 to 2023 | 2023 |
| Total inventory distortion (derived) | ~$1.94 trillion | Global retail | 2022 |
| Total inventory distortion | $1.1 trillion | Global retail | 2015 |
| Share of sales a typical retailer loses to stockouts | ~4% | Retail average | 2002 |
Out-of-stocks account for about two-thirds of IHL's inventory distortion total. That makes them a bigger leak than overstock markdowns, even though markdowns tend to get more attention in planning meetings. The 2023 improvement came mostly from supply chains reopening, with better forecasting systems also contributing.
Put the 4% benchmark on a store doing €6M a year and the stockout leak is about €240,000 in revenue. That loss never shows up as a line item. It is hidden inside a conversion rate that looks 'a bit soft' on your best-selling products.
Back-in-stock alert conversion: treat vendor claims with caution
There is no independent, published cross-industry benchmark for back-in-stock alert conversion. Most figures you'll find are from vendors that sell alert apps. Measure it yourself instead, as three numbers: opt-in rate (alert sign-ups ÷ sold-out PDP sessions), notification click rate, and purchase rate within 7 days of restock. Alert flows usually beat broadcast email because the intent is explicit. The main thing that decides whether alerts recover revenue is restock lead time, since delayed buyers drift away fast.
How to use these numbers on your store
Start with your own exposure rate: the share of PDP sessions that land on a sold-out product or variant. Multiply those sessions by your normal PDP-to-order rate and AOV to get revenue at risk. Then apply the response mix above. About 19% may be saved by same-brand substitution and some of the 15% who delay can be recovered. Treat the rest as lost.
Next, fix the experience on the product page. Show sold-out variants clearly instead of hiding them, offer an in-stock alternative in the same style or shade, and put the alert sign-up where the add-to-cart button was. Each of these is a testable change. Your PDP optimization roadmap should rank them by the revenue at risk you just calculated.
Finally, track the trend over time. With historical analytics data you can see whether stockout exposure spikes around launches, promotions or Black Friday. Those are the moments when inventory planning and merchandising have the most leverage, and when an empty size grid costs you the most paid traffic.
Out-of-stock statistics: frequently asked questions
The most-cited figure is 8.3% worldwide, from Gruen, Corsten and Bharadwaj's 2002 meta-analysis of 40 studies. The US averaged 7.9% and Europe 8.6%. Online stores selling sized or multi-variant products often see higher variant-level exposure.
IHL Group projected $1.2 trillion in global lost sales from out-of-stocks in 2023. At store level, the classic benchmark is that a typical retailer loses about 4% of sales to stockouts.
According to Gruen, Corsten and Bharadwaj (2002), 31% buy at another retailer, 26% switch to a different brand, 19% switch to another item from the same brand, 15% delay the purchase and 9% buy nothing. For a single-brand store, only same-brand substitution and delayed purchases are recoverable.
The best-known response data was collected in physical stores, where switching retailer takes effort. Online, a competitor is one search away, so the 31% 'buy elsewhere' figure should be treated as a minimum. This is especially true for products that marketplaces also stock.
The 2002 global study attributed 70–75% of stockouts to store-level practices: ordering, forecasting and replenishment. For online stores the equivalent is usually weak demand forecasting by variant, and slow reorders on best-selling sizes or shades.
No independent cross-industry benchmark exists; most published figures come from alert-app vendors. Track opt-in rate on sold-out PDPs, click rate and 7-day purchase rate after restock, then improve against your own baseline. Restock lead time has the biggest influence on the result.
Hiding them removes the chance to capture alert sign-ups and can break links from ads and SEO. Most stores do better by keeping the page live, clearly marking sold-out variants, recommending an in-stock alternative and offering a back-in-stock signup. Test both approaches on your highest-traffic products.
Send an event or parameter when a PDP loads with the selected or default variant unavailable. Then compare the conversion rate of those sessions with in-stock sessions on the same products. The gap, multiplied by affected sessions and AOV, is your revenue at risk.
Ads that drive traffic to sold-out products waste spend and lower ROAS, because the click is paid for but the sale can't happen. Pause or redirect ads for unavailable items, or exclude out-of-stock products from your Shopping feeds automatically.
The sold-out state is one of the product page states that most affects conversion. Treating it as part of PDP optimization means designing it deliberately, with clear variant availability, alternatives and alert capture, then A/B testing it like any other page element.
Find what's costing your store conversions — then test the fix
Metricuno shows where shoppers drop off, lets you change the page without a developer, and runs the A/B test that proves it. Free for 30 days. No credit card.