Today is a milestone for AppScout Labs.
MatrixFit Recommender — our first app, and the first real-world test of the AppScout Labs thesis — is now approved and live on the Shopify App Store.
You can install it right now: apps.shopify.com/matrix-fit-recommender
This post tells the complete story: where the idea came from, what we built, and what we're watching as it starts finding merchants.
Where It Started: An 87% Confidence Signal
In November 2024, AppScout's AI surfaced a pattern from 2,500+ merchant conversations — forum threads, Reddit posts, Shopify Community discussions, support tickets. The confidence score was 87%.
The pattern: apparel merchants don't have a good middle-ground tool for sizing.
Enterprise fit-tech solutions are accurate but priced and implemented for enterprise budgets — out of reach for most small brands. At the other end, static size charts are free but don't actually reduce returns. They just present the same ambiguous information in a prettier format.
The data behind the insight:
- 62% of apparel merchants in the sample cited sizing as their single largest cause of returns
- The problem was described as urgent, not abstract — merchants were losing money every month
AppScout's analysis pointed at roughly 45,000 qualifying Shopify merchants in the addressable market.
That signal was the starting gun for MatrixFit.
The Validation: Merchants Before Code
High confidence scores are a reason to investigate, not a mandate to build. Before committing to development, we put the insight in front of real apparel merchants. We wanted to know three things: How bad is this, really? What have you tried? What would you actually pay?
The conversations confirmed the direction the data pointed: static size charts weren't reducing returns, enterprise tools were out of reach, and the gap in between was real.
That's the bar we use for a go decision, and we hit it. We built.
What We Built: How MatrixFit Actually Works
The core mechanic is straightforward, and deliberately so.
Merchants upload their size charts as CSV files — a simple matrix of measurements and size labels. A women's apparel brand might upload a chart with bust and waist as the two axes; a lingerie merchant might use bust and underbust; a jeans brand uses waist and inseam. MatrixFit supports any two-measurement combination, so the app adapts to whatever sizing logic the merchant actually uses rather than forcing them into a predetermined structure.
Those size charts get assigned to products or collections through the merchant's admin panel.
On the storefront side, MatrixFit installs as a Theme App Extension — no code edits required. Shoppers see a "Find My Size" widget on product pages. They enter two measurements, and MatrixFit cross-references them against the merchant's uploaded chart and returns a recommendation with a confidence score: something like "Recommended: M — 94%."
The confidence score matters. It's not decoration. When the measurements fall squarely inside a size band, confidence is high. When they fall near a boundary between two sizes, the score reflects that honestly. We'd rather show shoppers genuine uncertainty than a falsely confident single answer.
Additional features that shipped in version 1.0:
- Automatic unit conversion — shoppers can enter measurements in inches or centimeters; MatrixFit converts automatically
- Analytics dashboard — tracks how many recommendations were given, which products generate the most size uncertainty, and recommendation conversion rates (did shoppers who got a recommendation actually buy?)
- Support for any measurement pairing — bust/underbust, waist/inseam, chest/sleeve, shoulder/length, custom labels
Pricing is deliberately simple. The free plan includes one size matrix, the size finder widget, 7-day analytics, and email support — enough to evaluate the impact on a real product before paying anything. The Plus plan at $9.99/month unlocks unlimited matrices, 90-day analytics, fit accuracy feedback, and live chat support.
The Build: What It Actually Looked Like
We gave ourselves 6–8 weeks from go decision to beta-ready — the standard AppScout Labs MVP window.
The defining technical bet was the CSV matrix approach. If you have a size chart — and every apparel merchant has a size chart — you can upload it as-is, with no data migration and no training period. The recommendation logic is deterministic and explainable: merchants can look at their chart and understand exactly why a customer got a particular recommendation. For a first version targeting adoption, simplicity beats sophistication.
The Theme App Extension was the other significant decision. The alternative — a script tag injection model — would have given us more flexibility but required merchants or their developers to manually add code to their themes. Extension-based installation is handled entirely through the Shopify admin.
Beta ran with 5 pilot merchants across women's apparel, athletic wear, children's clothing, menswear, and footwear — brands ranging from $100k to $5M in annual revenue, with different category types and different sizing logic. Exactly the spread we needed to pressure-test whether one mechanic could flex across the target market.
One thing that surprised us in pilot feedback: the analytics dashboard. We built it as a supporting feature, but pilots showed particular interest in it — the visibility into which products generate the most sizing uncertainty turned out to be valuable on its own.
This Is AppScout Labs' First App
It's worth being explicit about what today means beyond the app itself.
AppScout Labs exists to test a specific claim: that AppScout's AI can identify real market opportunities, not just interesting data patterns. An insight at 87% confidence is only meaningful if it leads to something merchants actually want and use. The only way to know is to build it.
MatrixFit is the first test of that claim at full scale — from signal, through validation, through build, through App Store approval. We now have a live data point.
We don't have enough data yet to declare the insight validated. That requires watching real-world adoption: how many merchants install it, how many stay, and whether it measurably reduces returns at scale. We'll track that and report it publicly, including the numbers that don't look good.
What we can say today: the 87% confidence signal identified a real, unsolved problem. Merchant conversations confirmed it. Five pilot merchants beta tested it across five apparel categories. Shopify reviewed and approved the app. That's a chain of evidence — and now the market gets the final vote.
The next milestone is the first paying merchant outside our beta program. We'll post metrics publicly as they come in.
Install MatrixFit
If you're a Shopify apparel merchant dealing with size-related returns, MatrixFit is available now.
Install from the Shopify App Store: apps.shopify.com/matrix-fit-recommender
Free plan available; Plus is $9.99/month. No coding required — it installs as a Theme App Extension.
Questions or feedback — especially if something doesn't work as expected — go directly to: labs@appscout.io
Follow the Journey
We publish metrics, development updates, and case study content at labs.appscout.io. Subscribe if you want to follow what happens from here: first customers, first churn, first decisions about what to build next.
If you're building your own Shopify app and want to see what AppScout's insight discovery looks like from the inside, the platform is at appscout.io.
— The AppScout Labs Team
Built by AppScout Labs. Validating AI-discovered opportunities through real execution.