Smart sizing doesn’t have to be one-size-fits-all. If you want to build a custom size finding
experience tailored specifically to your use cases, it’s possible with the Bold Metrics API. Your
developers can use this “application programming interface” to integrate Bold Metrics with
other pieces of your tech stack to develop your innovative solution.
Let’s check out how fashion retailers are already doing this (plus some never-seen-before ideas
to inspire you).
Mizzen+Main offers hundreds of dress shirt styles and colors. They wanted to combine the value of a fit finding quiz with a traditional “product finder” experience to serve personalized shirt results, while also filtering out unsuitable (and out-of-stock) sizes and fits.
They built their own stepped quiz flow, combining body specifications with style preferences and integrating Bold Metrics’ fit visualizer mid-quiz to enable shoppers to explore how both Trim and Classic styles will fit their bodies in their recommended and adjacent sizes.
Sun Day Red supports a similar fit finder quiz when you sign up for an account, which populates size recommendations on every product detail page for logged-in users who have filled in their profile.
Clicking the Recommended Size badge opens a fit profile to compare fit for one size up and down.
For Sun Day Red, this first-party data can also help personalize email and website merchandising and customize loyalty offers.
Suit Shop’s business is — you guessed it — suits. Most of their product pages include both jacket and pants together and require separate size selections.
Because custom suits have unique fit specifications, Suit Shop knows that asking for relative body shape and shoe size can help get even more accurate sizing recommendations, and they’ve included these steps in their finder quiz accordingly (in addition to standard fit questions like height, weight, age, and body measurements).
Featured steps within the Suit Shop fit finder flow
Shoppers who complete their fit finder quiz see both jacket and pants sizes pre-selected as they browse product pages.
The best part? This compact summary removes the default fit, size, and length variant chips that take up valuable screen real estate, bringing the Add to Cart button into clearer focus — a win for conversion rates.
Shoppers can easily expand their size tabs to review or change their applied selections. (You can see how much space is saved with this feature!)
Custom suit tailor Eph Apparel takes a different approach from the examples above, saving its fit quiz for a later step in the customer journey.
Once you’ve customized your suit through their customization solution and added your selection to cart, you’re then prompted to provide your fit details through a customized flow to build your measurement profile, which is passed on to the tailor for production.
Like Suit Shop, Eph includes its own unique questions that provide helpful context for suit tailors: chest type, shoulder shape, and posture.
In addition to standard measurements and fit preferences, Eph also includes optional inputs for neck size, arm length, and chest size. Given that their quiz is presented closer to conversion, it’s important that questions that may be difficult to answer — or that buyers prefer not to share — do not prevent a transaction.
By combining detailed body shape and measurement responses with Bold Metrics’ digital twin data set, Eph can provide more accurate tailoring without the need for an in-person measurement appointment.
When it comes to fit recommendations for minors, asking for sensitive information like body measurements or showing fit visualizers on child bodies is problematic. For this reason, French Toast needed a custom solution for its assortment of school uniforms for toddlers to teens.
Using the Bold Metrics API, French Toast customized its fit profile quiz to ask for only age, height, weight, and shoe size and to exclude any fit visualization graphics.
After completing the quiz, parents and guardians can navigate directly to Best Sellers or filter the catalog to the child’s school. The recommended sizes appear on product pages within the session and across sessions when saved to an account.
Integrating the Bold Metrics API with other applications can bring to life new and never-before-seen experiences. These are just a few ideas for what you can build with Bold Metrics.
Search and recommendation engines that support visual search can suggest close-match alternatives for out-of-stock sizes. Rather than request an email for a restock that may never happen, shoppers get instant gratification without the friction of returning to a product list or conducting a new search on their own.
For example, Retrofete displays small icons (similar to Google Lens’ visual search icon) with struck-out size chips that users can click to open a modal window with visually similar styles.
The recommendation engine filters styles without the specified size in stock and sorts matching options by how visually similar they are to the original item.
This experience reduces both customer disappointment and lost revenue from stockouts.
Now imagine if such a capability could be enhanced further through integration with the Bold Metrics API:
1. Customers with saved fit information can see their usual size, plus one size up or down if this fits their fit preferences (tighter, looser, etc.).
2. Visually similar styles that match the customer’s fit profile can be embedded directly into the product detail page in a recommendations pod above the fold.
Most shoppers won’t click on an out-of-stock variant when it’s struck out. Showing products immediately without any clicks required from the customer overcomes this challenge.
A number of customer review applications allow merchants to include additional customer data in their submission forms that can be displayed with user reviews and enable filtering by customer attributes like size purchased, usual size, and customer height, weight, age, and body type.
Depending on the review vendor, a customer’s saved profile information may always be displayed with their reviews, or the customer may have to provide this information every time they submit a review.
Integration with Bold Metrics’ API can auto-populate relevant fields, reducing friction and user effort in the submission process — especially when paired with purchase data (through integration with account history for logged-in users).
This can streamline the form to only require: “how’s the fit?” and “how comfy is it?”
Another opportunity to enhance the review experience is to personalize the sort order of customer reviews to the individual user based on reviews from “customers like you.”
For example, a product with different available fit options like Slim, Regular, Relaxed, Tall, and Extended Sleeve can show reviews with a given user’s saved fit preference at the top of the list, and apply a weighting factor for reviews from customers with similar body specifications.
Because ranking happens algorithmically — rather than requiring customers to manually apply filters — and most relevant reviews are pushed to the top, this can improve conversion, especially on mobile.
Many merchants include “Shop by Size” links with promotional emails to send shoppers directly to a filtered product list, especially during sales and clearance events.
Imagine if these size links could be scoped to an individual customer’s saved fit profile – showing a tighter range of sizes, rather than the full set – how much screen real estate could be saved, how easier the list would be to scan, and how more personal the email would feel.
This isn’t the only use case possible with the Bold Metrics API.
You could, for example, create cohorts of customers within your CRM system that only shop a specific size and offer them first-dibs discounts on products with overstocks in that size to help move inventory — without reducing the public-facing price on your website.
Rapid advancements in generative AI mean we’re getting closer to virtual try-on experiences that generate fit visualizations on our own body avatars. Walmart’s “Be Your Own Model” feature already lets shoppers upload a full-body selfie to virtually try on clothing, as do a handful of third-party consumer mobile applications.
However, there are several challenges to this approach to consider:
Lighting and image quality: Poor lighting and image quality can result in unrealistic or distorted generative images.
Usability and accessibility: Taking full-body selfies is difficult, time-consuming, and may feel intrusive.
Privacy concerns: Not all customers are comfortable uploading sensitive data like photos, and collecting this data adds complexity to retailers’ privacy and data governance requirements.
Unless a shopper is wearing a skin-tight outfit from top to bottom, the generative application will struggle to assess their actual body shape. Studies have shown conversion rates through body scanning mobile app solutions are as low as 0.25% due to these friction points.
We anticipate that generative AI vendors will soon make commercial applications available to support “be your own model” experiences on any merchant website. But these experiences will suffer from low adoption unless they solve for these friction and failure points.
Integrating Bold Metrics’ API with virtual avatar generation is one potential solution.
With just a facial selfie (which many consumers are already comfortable with) and a four-question quiz, online shoppers can see full catalogs “on themselves” with a more realistic picture of true-to-life fit.