Photo: Linkedin
Size Matters: How the Retail Industry Loses $1 Trillion a Year on Inaccurate Sizing
Size matters, especially for fashion brands. In fact, it matters so much that retailers are losing a whopping trillion dollars a year due to messed-up size availability. Let’s talk about how we can fix this colossal problem, and no, there is no “one size fits all” solution.
The Sad Truth About Size Availability
Here’s a scenario we’re all too familiar with: You spot your dream dress on Instagram, you tap to shop, and bam—your size is out of stock. Heartbreaking, isn’t it? Well, it’s not just a tragedy for you, the loyal customer, but also for the retailer who has lost out on a valuable sale. This happens millions of times a day across the world.
In fact, retail analyst IHL Group estimates that retailers globally lose $1 trillion in revenue annually due to out-of-stock situations. It’s astounding that such a vital aspect of the buying and fashion merchandising cycle, like sizing, is often overlooked by retailers of all sizes.
The Nitty-Gritty of Size Curves and Ratios
Before we dive into the solution, let’s break down some of the key terms and concepts related to sizing in the fashion industry:
- Size run: All the sizes that a particular product is produced in (e.g., XS to XL for clothing, or 5-13 for footwear).
- Size ratio: A relative ratio that can be applied to your total number of buy units (e.g., a 1:2:1 ratio for Small, Medium, and Large sizes).
- Size curve: A curve based on the distribution of buy units across the available sizes for a product, determined by the merchandising teams.
Now that we’re all on the same page, let’s talk about the elephant in the room: the current problems with sizing in the industry.
Where We Go Wrong with Sizing
There are several flawed methods commonly used in determining size curves and ratios, such as analyzing sales units versus intake units or full-price sales versus intake. These approaches often fail to take into account factors like size sellouts, age of stock, and markdown-driven sales, leading to inaccurate size curves.
So, how can we do better? The answer is to calculate the rate of sale when all sizes are in stock, ensuring you’re analyzing the true sales potential of each size. But with massive data sets, that’s easier said than done.
The Importance of Sizing and Its Impact on Your Business
Getting sizing right is crucial for several reasons:
- Reduces the need for heavy discounting, preserving profit margins
- Gives customers a reason to choose you over competitors
- Reduces waste and improves cash flow
- Enhances customer loyalty and satisfaction
By offering an optimal size mix that caters to the needs of various customers, retailers can improve customer satisfaction, create a positive shopping experience, and foster customer loyalty (Bolton, 2020). When customers can consistently find their size at a particular store, they are more likely to become repeat buyers and recommend the store to friends and family.
Minimizes markdowns and inventory holding costs: A well-planned size curve strategy can help retailers avoid overstocking or understocking of certain sizes (Style Arcade, 2021). This can lead to reduced markdowns, as retailers will be less likely to have an excess of slow-moving sizes and can focus on selling inventory at full price. Additionally, efficient size allocation can lower inventory holding costs by reducing the need for excessive storage space.
To implement an effective size curve strategy, retailers should consider the following steps:
- Analyze historical sales data: By examining past sales data, retailers can gain insights into the size preferences of their target customers and make informed decisions about the sizes they should stock (Heikkinen, 2020).
- Monitor customer feedback: Retailers should actively listen to customer feedback regarding size availability and preferences. This information can help retailers refine their size curve strategy to better meet customer needs (Lomas, 2018).
- Use technology to optimize size curves: Retailers can leverage advanced software solutions to analyze data, forecast demand, and optimize size allocation (Style Arcade, 2021).
- Continuously evaluate and adjust size curves: A successful size curve strategy requires ongoing evaluation and adjustments based on sales trends, customer feedback, and market changes. Retailers should be prepared to adapt their size curve strategy as needed to ensure continued success.
In conclusion, adopting an optimal size curve strategy is essential for retailers looking to maximize sales, reduce inventory costs, and enhance customer loyalty. By implementing the right size mix for their target customers, retailers can improve their bottom line and create a better shopping experience for their clientele.
Chase Hutchison is the head of business development at Style Arcade, an AI-powered merchandising platform. View the original article post on Linkedin.
Discussion Questions
DISCUSSION QUESTIONS: In your opinion, what is the main problem with sizing in the fashion industry? What are the steps that retailers can take to implement an effective size curve strategy?
Poll
Loading …BrainTrust
All good points on the urgency of getting sizing right, but the macro focus needs to zoom in on the online fashion shopping experience as well. Bracketing is the bain of an online retailer’s existence. We need to get better at this, and to me a solution is user-generatied content (UGC) and AI. With UGC you can see someone who looks like you, you can understand the fit, the drape and the feel of the product and not from some avatar, but from a real person. We need to get better at sizing in real time, and this is just one way to stop the madness.
Taking this one step further, I’d like to ‘follow’ users who are ‘like me’ so I can see what they like, and what they don’t like.
Sourcing isn’t mentioned yet it’s a fundamental variable, particularly in the wake of the pandemic. Retailers and brands have had to uproot long-standing sourcing partnerships and move at the drop of a hat (or disruption out of left field), further compromising size consistency. Until sourcing stabilizes, size consistency and accuracy will be a moving target.
As a former apparel merchant and product developer, the archaic planning process is at the root of retail’s sizing issue.
Factory partners also have a lot to do with it. Constraints with the way we create product, plan assortments and pack goods is only part of the issue. As brands and retailers improve technology and tools, factory partners need to do the same. Flexibility is crucial in reacting to consumer insights and feedback in size and fit.
Let’s add another problem to this list — size names mean nothing. I’m an extra small at one brand and a medium at another. This self-inflicted confusion also retards sales and fuels the boom in returns.
Totally agree, Cathy. We have managed to standardize so many things, why not sizes? There would be so many benefits to consumers and retailers alike. For starters, doing so would instantly decrease the need for size bracketing online. Imagine the financial benefits if we just reduced bracketing by 50%!
100% Cathy. Sizing is terrible. Especially for women’s apparel. The inconsistencies, even within the same brand, absolutely drive up returns. Brands that provide both consistent sizing, and a way for consumers to easily understand the size and fit of their products, will be the winners in ecommerce.
So true. Even in mens apparel, the size on the label often only means a 50% shot of actually fitting me correctly because different manufacturers use different patterns. In person returns have to be planned out, costing me time and money, while also costing the retailer money, labor. There has to be a way of reducing this smartly.
I totally agree that analyzing sales when all sizes are available and in stock is a great first step to analyzing true size demand. I also agree that size curves and ratios can not be managed with a “set it and forget it” mindset. Customer and demand patterns change constantly, so why shouldn’t our methods of analyzing their needs?
The lack of consistent global standards makes buying apparel tricky, especially online.
AI will help to analyze and normalize the abundance of non-standardized apparel data. AI will improve personalized fashion recommendations that fit.
AI can go a long way in filling in the blanks for missed size opportunities using demand planning and forecasting, so capturing opportunity with enhanced size curves is possible with the right technology. The other side of the sizing equation is that different silhouettes fit different body types differently. Size is not only a number but a highly variable combination of dimensions based on many body types and product silhouettes. Clear and consistent measurements executed by brands and available for view online, will aid in the correct size selection initially.
I have a problem with this anlysys.
Let’s see. “You spot your dream dress on Instagram, you tap to shop, and bam—your size is out of stock. Heartbreaking, isn’t it? Well, it’s not just a tragedy for you, the loyal customer, but also for the retailer who has lost out on a valuable sale. This happens millions of times a day across the world.”
It might be a tragedy for the single retailer, but not the marketplace. That shopper will not pout and never buy again. They will buy another choice.
I have criticized retail inventory systems for decades. Inventory management in apparel gives online retailers a huge advantage. If the retailer has hundreds, the retailer has to have hundreds of XS, S, L, XL, that will never be sold. Meanwhile, those mid-sizes go out of stock.
Consolidation with a limited purchase of XS, S, L, XL to serve those needs and additional investment in middle sizes means more inventory turns.
Inventory turns are one of the biggest factors factors in retail profitability.of inventory
All fair enough but a bigger part of the problem is that fit by size is not consistent or standard even within the same brand across factories let alone across vendors. A size 32 jean on a size 32 person is no guarantee of fit which creates the basis for a lot of online purchase returns and messes up the statistics for all the curves and ratios discussed. Quality control on production also creates issues. It’s not just the math it’s the silhouette and the pattern grading
“So, how can we do better? The answer is to calculate the rate of sale when all sizes are in stock, ensuring you’re analyzing the true sales potential of each size.”
In order to do that you need to be able to track what you had both in stock AND available to a customer at any given moment in time. What if a location was in-stock but the item wasn’t on the shelf (for physical customers), or a location was in-stock but their stock wasn’t available on-line due to maxing out capacity, a power outage, staffing issues, etc. (for digital customers).
The challenge is that most systems only track sales and current inventory positions. Not the inventory positions at the moment an order was placed. Nor the number of times a particular size was queried but shown as out of stock.
To track that requires event based data that can be used to recreate the actual inventory availability displayed when an order was placed. Not many solutions track all that data for digital customers. For physical stores that’s pretty much impossible without RFID – and even then, will depend on how often you scan and the range of the scanners.
Mr. Hutchinson makes some excellent points in the article, and I’d argue that lost sales is likely higher than the stated $1 Trillion. It’s a massive issue, but it’s not just about size. There are plenty of other product attributes that should be considered, such as feature, function, color, price, network availability. There’s interesting technology that considers substitution matrices, product demand share and fitnessed-based purchasing probability that optimizes inventory by product by store to mitigate against lost sales. AI and sophisticated algorithms play a central role in getting the right answers for retailers.
Was this written by ChatGPT? Out of stock and sizing are two different things.
Two very different things!
Even if a company gets the size data right on a chainwide basis, it must focus more on localization. This requires more logistical effort from the manufacturer, at the distribution center and in the retailer’s allocation systems. “One size does not fit all” if you consider regional differences within the U.S., and stores do their customers a disservice by ignoring the issue.
Interesting article but it is not about sizing. Cathy Hotka nailed the issue on actual sizing. Until retailers can create a universal size chart, and stick to it, shopping will continue to be a nightmare for consumers. And for retailers as well.
There’s no way I am believing that retail is losing $1 trillion a year on sizing issues. At $50 per garment, that suggests we’d be buying an additional 20 billion garments a year? An additional 20 billion garments of demand is out there? Nope. No way. And we think we have a waste problem now.
Of course sizing is an issue. It’s frustrating when you want to buy something and it’s out of stock. Shop earlier. Stop waiting for sales. Buy something else. And it’s frustrating when garments don’t fit the way we’d like. But that’s how brands become favorites. Style and fit.
Fit (garment specs) and size scales are very different issues. Solving one does not in any way solve the other. There was a time when buying 1/4/5/2 prepacks worked fairly well (menswear), both from a manufacturing and warehousing/distribution point of view. Life is more complicated than that now. But retailers have the data to solve that problem. They just have to be very surgical in how they look at the date to be sure items are fully in stock and they are measuring true demand.
If we look at overstocks / unsold inventory of sizes that didn’t match shopper measurements, plus what didn’t sell because forecasting didn’t guess high enough for actual shopper measurements, globally, then sure, a trillion is as good a guess as anything.
Print-on-demand data I get on clothing SKUs from the e-commerce site for the museum where I volunteer (we use the Printify network) isn’t robust enough to give confidence in a bulk buy of t-shirts or hoodies – just not enough data to plot into a normal graph. And those print vendors have their own sizing out-of-stock issues as well.
In the next decade, weave-to-order will be the way to go – collapsing supply chains of course and making retail and design truly local again.
I don’t think customers stop shopping when they run into an out-of-stock issue. They keep shopping, and buy choice #2 when choice #1 is unavailable. Maybe they wait until choice #1 is back in stock, but in either case, the demand had been satisfied. Maybe they bought #2 from the same retailer that they would have purchased #1 from. In either case, the demand was satisfied. The frustrating thing is that there may be no way of knowing that the sale was actually choice #2. So the demand for #1 is understated and the demand for #2 is overstated. But that customers demand was satisfied. It was not a “lost” sale. And if the customer ends up buying from a different retailer, then absolutely retailer #1 lost the sale. But retailer #2 captured it. The demand was satisfied.
That’s not to say that individual retailers don’t have both sales and margin upside if they are smarter about how they buy and distribute size scales. Selling more larges while not having to markdown extra smalls will of course be healthier for the business. But that still doesn’t solve the dilemma of where the customer bought choice #2 when #1 was out of stock. To me that begins to suggest it will be possible to offer narrower ranges of choices once the in-stock issue is under control.
is it possible that the retailer №2 from which the Clients will buy and satisfy their demand = will be from a different category than clothes? For example, instead of blouses / pants = buy a bag…or an umbrella…or a bottle of wine
Therefore, replacing Option 1 with Option 2 = is not only dangerous at the moment of loss of sales. Rather, the risk of loss of income in the category.
Returned items eat retailers’ profits and then some. It’s not just the impact of shipping costs, the loss of most or all of the original selling price and the cost of restocking, it’s also the impact on the shopper. When an item is purchased the customer is excited about it, and looks forward to delivery. When that item doesn’t fit, the disappointment is big enough to make a shopper consider other competitors next time.
We have seen an increased emphasis on merchandising, assortment, inventory optimization, and localization. However, as the IHL study indicates, one of the biggest challenges the retail industry faces is stockouts and the lack of sizes when the customer is looking to checkout.
With all the advanced predictive analytics and insights available, it’s shocking that the blending of the arts and sizes has not advanced far enough with size curve strategies. True fit technologies and virtual capabilities have recently advanced for made-to-order and customized items. However, the fundamental challenges the industry is experiencing are at the front end, during the product development and sizing optimization processes.
By leveraging a data and analytics-first approach, retailers and their manufacturing partners can advance with the appropriate size curves and meet consumer demands.
The “main” problem is hard to pin down. What I have observed over the years is:
-“Size creep” especially in woman’s clothing. A “used to be” size 4 is now a size 2.
-Sizes are not uniform. One Brands size 8 is another’s size 6.
-Sorry, but all of us, (as a group!) are getting bigger.
-This is completely speculative, but the market is not very “accepting” of larger sizes. Which triggers the size creep.
-Size/style/color continues to be difficult to “get right”
I think the overall problem is almost unsolvable for brands and retailers. Sadly.
These are all great points but there is one more that may be even more critical – sizing charts and the size labels themselves. Most consumers will agree, the size label at one brand means almost nothing when compared to another brand. Sometimes this is even true within the same brand. That lack of consistency in sizing causes consumers to either order multiple sizes when shopping online, or causes them to spend much more time than they likely desire trying on different sizes to make a purchase decision. Both of these factors can also result in lost sales, not to mention create a horrible returns situation for retailers. This lack of size consistency in fit needs to be addressed before meaningful change can happen in the apparel industry.
Sizing inconsistencies by vendor is a ripe opportunity. As well, fashion trends change every several years, loose fit to tight, back to loose, with variations in between.
Creation of agreed upon industry standards, application of data analytics and standardized consumer tools will help shoppers, but also buyers, merchandisers.
Mens shirt companies like Charles Tyrwhitt, and Lorenzo Uomo, even Bonobos are brand examples having made small headway here, improving shopper confidence in sizing and style variation. Why couldn’t this be pursued further, applied in all clothing, for men and women alike?
I’ll say this, here the bigger problem for retailers is the inability to make sales from wherever the customers are present. Many retailers lack visibility into their own inventory which results in loss of sale, despite having a chance that the shopper’s ordered size for apparel might be available. If retailers modernize their approach by equipping Order Management Systems (OMS) into their operations, they would know the availability of each & every item under their belt at different locations from stores to warehouses. The absence of an OMS results in retailers having no clue over their inventory and being unable to make it available on different channels like physical stores, eCommerce, or even social media.
Secondly, many retailers lose sales on inaccurate sizes due to their lack of capability in taking pre-orders. Accepting pre-orders not only allows retailers to sell their future inventory at full price but also helps in assessing the demand for a particular item, in this case, a particular apparel size, which proves crucial in preparing a better size curve and planning a better strategy. Retailers who successfully manage to take pre-orders find it useful in tracking the current demand for their products and plan their inventory accordingly to avoid missing out on a sales opportunity.
a trillion dollar problem is unlikely to contain only one variable = size. Especially the practice of recent decades with the use of elastic fabrics (even for jeans, jackets). The general popularization of clothes from elastic fabrics – t-shirts, hoodies – instead of clothes of fixed sizes. The specified problem on a trillion = did not solve.
After all, it is possible to analyze even more deeply = and how much money frozen in non-odorous sizes?
Reducing discounts if only the right size = reducing the impulses for visits = how much loss of income?
And the author wondered why not just add equipment/capacity at the factory for the missing sizes?
If you think in the proposed course = you can calculate how much money is lost food retail, if their goods are not at hand when a person has a desire to eat. But isn’t that absurd?