Personalized Marketing Success Turns First-Time AI Users Into True Believers

Founded in 2006, Miustar is a long-standing women’s apparel site specializing in affordable Korean style clothes, shoes, bags and accessories. As they’ve kept up with K-fashion trends over the years, they’ve also stayed ahead in ecommerce, always adopting the best marketing practices for converting more customers online.


More competition than ever, conversion rate could be better


Backend automation, site-wide personalized recommendations



conversion increase


greater return on ad spend


“When AI solutions began getting popular we had questions, like just how accurate is automated data analysis anyway? We didn’t know if recommendations really improve conversions or not. So long story short, when we got the demo and gave it a try, our conversions went up before the demo period was done!”

Miustar-Intro01.jpg case study


Miustar has managed to stay relevant for over 15 years in the hyper-competitive Asian fashion industry because they constantly change with the times.

For example, when the biggest websites began using Instagram user-generated content on their landing page, Miustar did too and improved customer engagement immediately.

Their willingness to adopt AI-driven personalization paid off especially well because the solution is optimized to provide unique shopping experiences for individual fashion enthusiasts — a strategy that has also worked for the world’s biggest brands.
The Problem

Fashion ecommerce customers expect personalization more and more

With so many apparel websites out there nowadays, the Miustar Growth Management team knew that selling the hottest items and spending more on advertising are not long-term solutions.

They needed to work more efficiently with what they already had which meant converting more of their existing traffic and getting more return on their ad spend. But both of those goals depend on something more fundamental: consumer insights.

Processing product/customer data isn’t easy

In the past, Miustar had to spend a lot of time updating product information manually and analyzing consumer shopping journeys one by one in order to get consumer insights.
For years they suspected this wasn’t the ideal way to handle their data.

The new direction

Leading brands these days are focused on the customer. They've realized that understanding consumer preferences and providing an enjoyable shopping journey are the best ways to make more conversions.

This was the direction that Miustar needed to go and was there to guide them.
The Solution

Backend automation, site-wide personalized recommendations

The personalized marketing solution is unique because it uses a visual-AI-driven recommendation engine optimized for fashion ecommerce.

For Miustar, it improved their backend efficiency and provided accurate recommendations to shoppers, thus streamlining the workflow and increasing conversions.

Most importantly though, the extra-detailed tags get used to build up shopper preference profiles.

Automated product tagging and analysis

Automating product tags relies on computer vision trained by beauty and fashion industry experts. The training makes sure that the AI writes the tags to industry standards.

Category tags can be set to #style #pattern #material #sleeves #color, and up to 30 more. The tags can also be written into product descriptions, and used as clickable links to other related categories, improving product discovery.

Auto-generated product tags via image-based AI.

The additional product tagging detail in shopper preference profiles helps Miustar make accurate recommendations, a subtle part of the CX that goes a long way to build better customer relationships.

Plus, the automation saves time spent on maintaining web pages with product updates.

AI-powered personalized marketing concludes consumer journey

The awesome accuracy and high level of detail of the tags provides deeper insight into consumer preferences so the AI can predict what shoppers intend to buy at just the right time.

But the AI understands more than just consumer preferences. It also analyzes on-site behavior in real time, paying attention to clicking, scrolling, browsing and buying on every page.

With this data in hand the AI runs a recommender box featuring items “You might really like” on the Miustar product details page, just when the shopper is close to purchasing.

Seeing products with attributes that are similar to ones seen before converts shoppers at a higher rate, and even drives up average order value.

The “You may also like” recommendation on the Miustar product page.

Personalized recommendations ease product discovery, letting consumers find lots of their favorite things grouped together. Saving time while searching this way removes the friction-point of searching for products one by one. It makes them click through to more products, stay longer and buy more!


conversion increase


greater return on ad spend

After adding personalization to their online marketing strategies, Miustar improved CVR, AOV and revenue, as well as saving time and lowering costs to maintain the product information.

The increased cross-sells on the Miustar’s product page have achieved a 32% conversion rate increase and 1.47x greater effectiveness for ROAS.

With AI-powered personalized marketing, Miustar grew business, saved time, lowered costs and became able to predict more future purchases based on accurate customer preferences (thus reducing stockpiling).

Above all, personalized recommendations have levelled-up Miustar’s understanding of each customer’s behavior and preferences at the critical last phase of the customer journey. The deeper consumer insights create authentic connections with consumers, greatly increasing the likelihood of many happy returns!

Enabling merchants to understand the language of fashion.

Get Started