YesPlz Product Recommendation System vs. Others

The Difference Between Product Recommendation Systems

by Jess Erdman, Content Marketing LeadNovember 2021

The best things in life take time. And, when it comes to product recommendation systems, taking an extra few minutes can make all the difference in showing your customers relevant, accurate product recommendations.

Each product recommendation system is different, based on the underlying technology that supports each. Unsurprisingly, product recommendation systems are becoming more and more accurate based on new technologies.

In this article, we’ll go over:

A history of product recommendation systems
Visual search for product recommendations
What happens behind the scenes when you install the YesPlz Product Recommendation System on Shopify

The YesPlz Product Recommendation System is built to understand your store’s products and make fashion-forward recommendations. You can learn more about how our Product Recommendation System works in this blog.

Let’s dive into product recommendation systems.

A history of product recommendation systems

In the past, product recommendation systems were based on matching the texts of a product title to a recommended product. So, if your customer were to search for the word “boots,” they would see recommended products that are boots. Easy, right?

But, once we begin to introduce more complex search terms into the mix, it becomes proportionally more complicated. For example, if your customer were to search for the term “white boots,” they might receive a mix of recommended products--some which are white, and some which are boots.

Text-based search is limited to the quality of the keywords. The customer searching for a product needs to know which search terms to type in. And, on the other side, the eCommerce company needs to have a system that can recognize both keywords and search intent.

The complicated fashion taxonomy behind text-based search can limit the success of a product recommendation system. If your customer is searching for a white boot and receives recommended products that include a running sneaker, it’s unlikely that they will click through to an entirely different product.

Product recommendations should be within the same category and/or complementary to the search--for example, a winter jacket search might return winter accessories or other winter jackets in a similar style.

The sheer complexity of different product attributes (for example, a white boot description could include heel height, type, and material). Any combination of these text-based keywords could create a different set of product recommendations, depending on which keywords are selected.

Keyword-based product recommendations are unpredictable, especially as product searches become more complex.

And, from the point of view of the retailer, it’s overwhelming to create product descriptions. It’s hard to create a balance between product descriptions that are easy for a keyword-based search engine to understand--and to create product descriptions that are engaging and fun for customers to read. You shouldn’t have to make a trade-off that can affect customer engagement.

What’s the solution for a problem with serious consequences, such as customers not completing sales or bouncing from websites?.

The advantages of visual search for product recommendation systems

Keyword-based search is burdensome for customers and retailers. Enter visual search.

No more vague contextual clues for a computer to likely misinterpret. No more strange product recommendations that include inaccurate sub-categories of the original search.

Visual search can completely change the course of product recommendation systems. Customers are more engaged and more loyal. Through visual search, customers can also experience online shopping that’s closer to in-store. All in all, visual search has countless advantages, especially when building a product recommendation system.

When setting up a product recommendation system, whether text-based or visual, the technology needs time to scan products to create the recommendations. Text-based search can take a few minutes to learn 100 products. Visual recognition can take up to 10 minutes to learn 100 products.

But, back to the original question:

Would you rather have product recommendations that are accurate (and takes 10 minutes to set up) or inaccurate (with a few minutes of set-up time)?

The answer is obvious--the best things in life (and product recommendations) take a few minutes more to set-up.

What happens behind the scenes when you install the YesPlz Product Recommendation System on Shopify

When you first install the YesPlz Product Recommendation System on Shopify, you’ll be greeted with a screen to get started and connect your store. You also have the option to learn more about the YesPlz Product Recommendation System.

After you connect your store, our artificial intelligence will study your store’s products in order to make the best possible recommendations to your customers. By taking the time to truly learn your store’s products, our product recommendation system will be able to give quality recommendations that lead to higher click-through rates.

YesPlz secured $1.51 million in total seed funding and we are excited to announce the launch of the world’s first visual search filter that helps style online consumers through personalized AI technology. By selecting just a few design parts, the AI-powered filter makes it possible to instantly provide suggestions from thousands of clothing and footwear options.

\r\n

YesPlz is led by CEO and Co-founder Jiwon Hong and CTO, Sukjae Cho. Hong has spent her career building search and recommendation engines, and Cho brings experience as a former member of the Microsoft XBox team and ML engineer. The founding members have teamed up to leverage AI technology to create an ingenious and personal experience that brings the best of offline shopping, online.

\r\n

“This launch has been a labor of love - of fashion and of user friendly technology,” shared Jiwon Hong, CEO and co-founder of YesPlz. “It’s a dream to finally launch the world's first AI based visual search filter that allows for an intuitive, fun and curated online shopping experience. No more endless scrolling down to see irrelevant clothing.” With this investment, we look forward to growing our team and scaling our production right away.”

\r\n

Keeping up with constant updates to technology and fashion trends can be both expensive and difficult to execute. To address this challenge of the e-commerce industry, YesPlz has created a universal ‘fashion language’ relatable to any consumer, through their Style Filter. The unique search capabilities of the filter allow consumers to visually communicate their style preferences, guiding the YesPlz proprietary algorithm to deliver search results curated for that consumer. The filter empowers consumers to find a product they are looking for without having to be a fashion expert.

\r\n

“With the challenges of living in a Covid-19 world, online shopping has only increased,” shared Gyyoung Kim, EVP of KOLON Industries, Inc. FnC Organization, the leading global fashion company in Korea. “As an online retailer, creating the best virtual shopping experience you can is even more critical than it ever was. What YesPlz has created both empowers the consumer today and makes it personal with their innovative technology.”

\r\n

Boston Consulting Group (BCG), reports that COVID-19 lockdowns have led to an uptick in first-time e-commerce shoppers. 14% of consumers in the United States and 17% of consumers in China bought fashion items online for the first time because of the pandemic.

\r\n

While the Style Filter debuted with its first international client in Korea - Kolonmall.com - in September of this year, the company is planning additional client launches for this fall and the end of the year, in time for the busy holiday shopping season.

","published":1,"pub_date":1603756500,"publishedon":1603756500,"alias":"yesplz-raises-1.5mm-in-seed-funding","template":12,"resource_author":"33","resource_cta-text":null,"resource_cta-link":null,"resourcecta":null,"resource-ebook":null,"tags":"48","chips":null,"preview_picture":"/assets/images/yesplz-funding-blog.jpg","detail_picture":"/assets/images/funding-annmnt-blog-notext.jpg","subtitle":"by YesPlz"},{"id":59,"pagetitle":"YesPlz Product Recommendation System vs. Others","introtext":"Each product recommendation system is different, based on the underlying technology that supports each. In this article, we'll go over the history of product recommendation systems, how visual search affects product recommendations, and the YesPlz system.","content":"

\r\n

In this article, we’ll go over:

\r\n

A history of product recommendation systems
Visual search for product recommendations
What happens behind the scenes when you install the YesPlz Product Recommendation System on Shopify

\r\n

A history of product recommendation systems

\r\n

Keyword-based product recommendations are unpredictable, especially as product searches become more complex.

\r\n

What’s the solution for a problem with serious consequences, such as customers not completing sales or bouncing from websites?.

\r\n

The advantages of visual search for product recommendation systems

\r\n

Keyword-based search is burdensome for customers and retailers. Enter visual search.

\r\n

No more vague contextual clues for a computer to likely misinterpret. No more strange product recommendations that include inaccurate sub-categories of the original search.

\r\n

But, back to the original question:

\r\n

Would you rather have product recommendations that are accurate (and takes 10 minutes to set up) or inaccurate (with a few minutes of set-up time)?

\r\n

The answer is obvious--the best things in life (and product recommendations) take a few minutes more to set-up.

\r\n

What happens behind the scenes when you install the YesPlz Product Recommendation System on Shopify

\r\n

$\"\"$

\r\n

Your product data will be synced with YesPlz AI, and finally, you’ll decide where to display the recommendations and be able to customize the types of recommendations:

\r\n

$\"\"$

\r\n

The final recommendations displayed to customers will look like this:

\r\n

$\"\"$

\r\n

Relevant, fashion-forward product recommendations. A search for a brown, men’s dress shoe will give product recommendations that are within the same fashion category and style.

\r\n

As you can see, good things take time--and that includes product recommendation systems. While the YesPlz Product Recommendation System takes the time to study your products, the system can truly make the best possible recommendations for your customers.

\r\n

You can try the Product Recommendation System here--and don’t forget to let us know your thoughts!

","published":1,"pub_date":1635771600,"publishedon":1635771600,"alias":"yesplz-product-recommendation-system-vs.-others","template":12,"resource_author":"35","resource_cta-text":null,"resource_cta-link":null,"resourcecta":"[]","resource-ebook":"79","tags":"49","chips":null,"preview_picture":"/assets/images/yesplz-blog-cover-images-3.jpg","detail_picture":"/assets/images/yesplz-blog-cover-images-4.jpg","subtitle":"The Difference Between Product Recommendation Systems"},{"id":117,"pagetitle":"12 Product Filter Examples for Fashion Website","introtext":"Service planning is hard when you’re trying to integrate technology with service offerings. In this guide for fashion eCommerce, we go over step-by-step, how to plan fashion product filter powered by AI. You can also discover our product filter templates for fashion websites to help you along the way.","content":"

Service planning is hard when you’re trying to integrate technology with service offerings. What do you offer? How do you know whether you’re planning too little or overplanning? And how do you even begin?

\r\n

Fashion eCommerce stores know this dilemma all too well: you want to integrate new service offerings, and you know that there’s a lot of technology out there to help: but where do you begin?

\r\n

By taking the time to plan your service offerings, you’re setting up your eCommerce for success.

\r\n

Vertical Product Filter

\r\n

$\"modern$

\r\n

What technology should you integrate?

\r\n

Fashion artificial intelligence (AI) is artificial intelligence that understands the ins-and-outs of fashion. Fashion AI can understand fashion taxonomy, and even develop fashion taste that fits your customers. It can include search, recommendations, personalization–and can perform the same job as a helpful in-store sales associate.

\r\n

The number of ways to implement fashion AI are dizzying–and many fashion eCommerce stores are concerned about the implementation process. But, once you start to break down the steps to create an action plan, it’s not as overwhelming.

\r\n

Why use fashion AI in your eCommerce?

\r\n

Nearly every major retailer is using fashion AI in some form, from predictive analytics to search. Utilizing fashion AI will help your brand keep up with the competitive industry landscape, and stay ahead of competitors.

\r\n

The benefits of fashion AI include:

\r\n

Increased rates of customer satisfaction, resulting in higher rates of return

\r\n

Easy-to-navigate search tools for customers, leading to more completed purchases

\r\n

More accurate product recommendations, with higher basket sizes

\r\n

Automatic product tagging of new SKUs, saving retailers precious time and energy

\r\n

You can learn more about all of the benefits of fashion AI in our Complete Guide to Fashion AI.

\r\n

The Steps to Creating a Service Plan

\r\n

By following the below checklist, you can make sure that fashion AI planning and integration is seamless, and aligned with your eCommerce’s goals.

\r\n

If you’d like more in-depth advice on creating a service plan, contact us for a free 20-minute consultation where we can go into more details.

\r\n

Step 1: Define service scenario.

\r\n

Forget all of the different types of technology, and the question of whether technology can help or not. First, define the service scenario. But, how? We recommend identifying either a pain point or an experience that has the potential to delight your customers. Perhaps you find that customers are struggling to complete their purchases after repeated search attempts. Or, customers simply aren’t engaging with product recommendations. You’ve identified your pain point.

\r\n

Alternatively, some eCommerce stores want to go above and beyond their competitors–by creating experiences that not only solve problems, but delight and bring joy to customers. In this case, you can brainstorm different scenarios.

\r\n

Some examples could include:
-a style quiz
-a personal wardrobe
-daily style curation over email
The opportunities are sky-high for those eCommerce brands willing to look into the future.

\r\n

Step 2: Make the logic in order to implement the service.

\r\n

Unfortunately, there isn’t a magic wand you can wave to translate the service into some form that your website can understand. But, there is good news: it’s not as difficult as it seems.

\r\n

Simply set a rule for your service. Let’s say you’re planning to integrate recommendations into a product page. You can set a rule to show recommendations from the most popular brands first. Or most-clicked products. Or visually similar products. You can set the rule yourself.

\r\n

Tip: The more attribute data you have, the more services you can create. For example, you can use fashion AI to pull only “floral print” tops or “spring vibe” skirts that match the tops, and display the collection under the spring banner “Our spring collection.” Fashion AI makes this easy.

\r\n

Step 3: Test with users.

\r\n

Without a test run with users, you won’t know how the service is helping or delighting your customers. By testing and measuring, you can know the actions that users take. Are your customers clicking more products? Placing more items in a shopping cart? End up purchasing more items overall?

\r\n

Step 4: Refine usability and polish design.

\r\n

After testing, you’ll likely receive feedback that will cause you to go back and refine the service design. Don’t worry–it’s completely normal (and troubling if you don’t have any feedback to implement). This will get you closer to the service implementation that your customers want.

\r\n

Step 5: Continuous reiteration

\r\n

Part of the design process includes continuous reiteration–because nothing is ever perfect. There’s always room for improvement, and by implementing a mindset of continuous reiteration, you can continue to stay on top of the latest trends and keep customers satisfied.

\r\n

Do you have an idea to improve your fashion eCommerce, but lack the engineering resources to implement it? At YesPlz AI, our fashion AI can power your ideas. We have multiple templates for you to choose from that are powered by fashion AI.

\r\n

GET READY FOR THE 2021 HOLIDAY SEASON WITH OUR MUST-READ EBOOK.

Tips and tricks for personalizing your eCommerce
Why personalization is the key to eCommerce success this holiday season
Common mistakes made by eCommerce companies

\r\n

YesPlz eCommerce Product Filter Templates, Powered by Fashion AI

\r\n

Template libraries with fashion AI

\r\n

Save time, but don’t sacrifice on quality with pre-built, customizable templates from our library. We have templates for fashion search, product recommendations, and mobile filters. In addition, each template category design is customizable, so you can install the template that works best for you–including the design that suits your website.

\r\n

Why fashion AI matters

\r\n

Our templates are more than aesthetically pleasing–they’re also powered by real, trained fashion AI. We’ve spent years training our fashion AI to understand the ins-and-outs of fashion, from product attributes to product recommendations, resulting in templates that are powered by technology.

\r\n

Why use a template?

\r\n

When templates are well-built and well-designed, they offer a seamless way to integrate a new service offering into your online store. At YesPlz, our templates are comprehensive–and customizable, so you’re not stuck integrating a template that doesn’t fit your business needs.

\r\n

Our Services, Powered by Fashion AI:

\r\n

Fashion search: Our Style Filter makes search navigation easy by using a virtual mannequin to narrow down search results and filter by product attributes/occasion

\r\n

Product recommendations: Whether your customers want similar products or to complete their look, our product recommendations can find what they’re looking for

\r\n

Personalization engine: An engine that learns shopper preferences, tastes, and behaviors, and creates curated collections for them

\r\n

Image tagging: Our technology can quickly and accurately tag product attributes from images, setting up your eCommerce to implement powerful personalization tools

\r\n

A Taste of Our Template Library

\r\n

Fashion focused product filter templates:

\r\n

We compiled fashion search templates after years of interviews with users about their preferences and search behavior. All of our templates are powered by fashion AI, eliminating the need for manual tagging. Here's a sample of our templates (you can download the full template eBook below).

\r\n

Side Filter

\r\n

$\"\"$

\r\n

Hybrid Filter

\r\n

$\"\"$

\r\n

Different Look

\r\n

$\"\"$

\r\n

Curious to see mobile and more design template ideas and use cases for your website?

\r\n

Download our free template ideas:

\r\n

GET READY FOR THE 2021 HOLIDAY SEASON WITH OUR MUST-READ EBOOK.

Tips and tricks for personalizing your eCommerce
Why personalization is the key to eCommerce success this holiday season
Common mistakes made by eCommerce companies

\r\n

Tip: You can always plug in our fashion AI to make the template come to life without the burden of manual tagging.

","published":1,"pub_date":1651579380,"publishedon":1651579380,"alias":"how-to-plan-a-new-service-using-a.i","template":12,"resource_author":"33","resource_cta-text":null,"resource_cta-link":null,"resourcecta":"{\"fieldValue\":[{\"resource-cta_text\":\"Schedule a free 20-minute consultation\",\"resource-cta_link\":\"www.yesplz.ai/contact\"}],\"fieldSettings\":{\"autoincrement\":1}}","resource-ebook":"120","tags":"48||49","chips":null,"preview_picture":"/assets/images/12-1-22_1.jpg","detail_picture":"/assets/images/12-1-22_2.jpg","subtitle":"A YesPlz Guide"},{"id":128,"pagetitle":"Battle of eCommerce Search Filters: UO vs. In the Style","introtext":"Ecommerce search filters affects not only how customers find products, but the overall experience. In the first of our product filtering battle series, we put Urban Outfitters and In the Style to the test based on YesPlz filtering criteria.","content":"

$\"\"$

\r\n

Ecommerce search filters impact not only how your customers find products, but the overall user experience. Despite the thousands of search solutions available, only 16% of eCommerce websites provide a “positive” filtering experience.

\r\n

How can the majority of fashion eCommerce companies improve the fashion search experience? We created the answer in YesPlz’s Product Filter Evaluation Checklist–a guide for fashion eCommerce when evaluating their own product filtering.

\r\n

Today, we’re going to apply the product filter checklist to two filters: Urban Outfitters (UO) and In the Style, and based on YesPlz criteria, determine which filter is stronger.

\r\n

GET READY FOR THE 2021 HOLIDAY SEASON WITH OUR MUST-READ EBOOK.

Tips and tricks for personalizing your eCommerce
Why personalization is the key to eCommerce success this holiday season
Common mistakes made by eCommerce companies

\r\n

YesPlz Best Practices for eCommerce Search Filters

\r\n

Based on our own research and user research from The Good, we’ve determined the following criteria for product filter & search:

\r\n

1. Keep filter language jargon-free

\r\n

The subtle differences in fashion language can cause users to become confused, overwhelmed, and simply miss out on your products because of a lack of understanding.

\r\n

Ideally, you should look to your customers for cues on language to describe products–if your customers aren’t using the term “cap sleeves” to describe a sleeve type, then you shouldn’t either.

\r\n

Using consistent, easy-to-understand language in eCommerce search filters helps guide users to the products they’re looking for while continuing to discover your product catalog.

\r\n

2. Include thematic filters to keep users engaged

\r\n

Our research shows that users have an “open search mindset”--they want to search within broader categories, but still have a checklist in mind for fit and silhouette.

\r\n

For example, “statement dresses for work that hide my arms” is how users actually think about search–but users are limited by burdensome, unhelpful product filter & search. By offering thematic filters, users can quickly navigate to the occasion they want, and continue to explore.

\r\n

Unfortunately, manually creating thematic filters can cause inaccurate search results and waste hours of time, due to human error and subjectivity. Even the most sophisticated fashion experts may define a “work dress” differently from each other.

\r\n

Fortunately, fashion AI can learn to correctly tag based on theme, removing human error.

\r\n

3. Include key product specifications in eCommerce search filters

\r\n

Even the most open-minded shopper wants to filter by specific product specifications like price and size.

\r\n

Product type and material are the second most asked-about specifications. And finally, an ideal product filter & search solution will also include product filtering by style, such as neckline, sleeve length, and silhouette.

\r\n

Product specifications should be organized in a way that makes sense to the browsing experience (and not include long lists of text filters).

\r\n

4. Make navigation easy and seamless

\r\n

Product specification-rich filters will be meaningless unless users can easily make changes like adding or removing them. An ideal eCommerce search filter navigation experience will include:

\r\n

Truncated design
Easy to add and remove filters
No zero-search results (hide the filters that don’t return search results)
Strategically-ordered filter categories

\r\n

5. Advanced search interface such as unique filters for mobile

\r\n

Users are more sophisticated now than ever, and the standard for product filter and search has risen in the past 2 years. Users expect to have a superior mobile search experience, optimized specifically for small screens. And, shoppers are unwilling to read heavy text filters on mobile–meaning that visual cues are becoming more common to guide users.

\r\n

Faceted product filtering is also becoming more common, allowing users to filter based on multiple selections (for example, color and sleeve type). This requires more advanced data management on the part of the retailer, as certain product specifications will be excluded depending on the search results. However, faceted product filtering is a more seamless way for users to search for clothing.

\r\n

$\"\"$

\r\n

Battle of eCommerce Search Filters: UO vs In the Style

\r\n

Based on the above criteria, we’re comparing the product filter & search at Urban Outfitters and In the Style, two retailers with large product catalogs.

\r\n

We’ll rate each retailer for each of the 5 YesPlz eCommerce search filter checklist items, with a maximum score of 5 out of 5.

\r\n

Battle #1- No Fashion Jargon

\r\n

Winner: It's a tie!

\r\n

Urban Outfitters: 👠👠👠 (3/5)

\r\n

In the Style: 👠👠👠 (3/5)

\r\n

Urban Outfitters’ language is (mainly) easy for shoppers to understand, with universal terms like “halter” and “square neck” to describe necklines.

\r\n

In the Style, on the other hand, didn’t include the naming of style product attributes. Even when product specifications like size are named, they’re difficult to understand. The naming conventions for size at In the Style are inconsistent, bouncing from numbers to years, making it difficult to understand what the product specifications are referring to.

\r\n

However, In the Style explains each type of filter selection (like “Bardot Dresses”) with a text explanation, which we think adds value for the user.

\r\n

$\"\"$

\r\n

Room for Improvement:

\r\n

There is language that could be improved in Urban Outfitter product filtering, such as “mock neck.” Other language on the website included terms that could be difficult to decipher, like “babydoll” tops. Alternatively, we like that Urban Outfitters is actively thinking about fashion jargon–by placing “mock neck” in the same category as “turtleneck,” the user can infer that a “mockneck” is similar to a “turtleneck.”

\r\n

In the Style chose to explain eCommerce search filter terms using sentences, which becomes very text-heavy for users that are browsing and looking for a visual experience.

\r\n

Battle #2: Include thematic filters

\r\n

Winner: In the Style

\r\n

Urban Outfitters: 👠👠 (2/5)

\r\n

In the Style: 👠👠👠 (3.5/5)

\r\n

$\"\"$

\r\n

$\"\"$

\r\n

Urban Outfitters chose to curate their thematic filters to include “festival outfits” and “sun shop”--but we’d love to see more thematic filters that fit the mood and vibe. Users can also find thematic filters under each product category (for example, choosing a “casual” or “going out dress” after selecting “dresses”). Oveall, there simply weren’t enough thematic filters offered–and the filter categories were subjective with inaccurate search results.

\r\n

In the Style offers thematic filters as well, ranging from “graduation dresses” to “occasion dresses.” Although the categories were sometimes difficult to decipher, there was a richer offering than Urban Outfitters–for example, we saw “Wedding Guest Dresses” included alongside “Daytime Dresses”; “Summer Dresses”; and “Graduation Dresses”--all very unique and relevant occasion/themes.

\r\n

Room for Improvement:

\r\n

Simply offering thematic eCommerce search filters isn’t enough.

\r\n

One of the biggest problems with thematic filters is inaccurate search results, often stemming from problems with fashion tagging. Shoppers not only expect to see thematic filters, but accurate search results under their selections. This becomes more difficult to achieve in thematic filters, because of the subjectivity of humans when tagging manually.

\r\n

Urban Outfitters chose to curate their thematic filters to include “festival outfits” and “sun shop”--but we’d love to see more thematic filters that fit the mood and vibe. Users can also find thematic filters under each product category (for example, choosing a “casual” or “going out dress” after selecting “dresses”). Overall, there simply weren’t enough thematic filters offered–and the filter categories were subjective with inaccurate search results.

\r\n

After selecting “party” dresses at both retailers, we were met with mixed results, ranging from patterned orange to floral dresses. Urban Outfitters search results also showed pants, clearly in the wrong product category.

\r\n

Perhaps it is our own bias, but we would expect to see a mix of sequins and black dresses under “party and going out dresses.”

\r\n

$\"\"$

\r\n

Eliminating subjectivity in thematic filters is key–and fashion AI plays an important role in creating accurate, consistent occasion filters.

\r\n

Battle #3: Show key product specifications in filters.

\r\n

Winner: Urban Outfitters

\r\n

Urban Outfitters: 👠👠👠 (3/5)

\r\n

In the Style: 👠 (1/5)

\r\n

Key product specifications like price, size, color, pattern, and material should be included in filters, as well as style attributes like neckline, sleeve type, and fit. Urban Outfitters does a good, if not inconsistent job, of providing this information–users can easily filter by neckline or sleeve type.

\r\n

In the Style, however, does not include filters for style attributes, making the search experience frustrating for users with specific style preferences.

\r\n

Room for Improvement:

\r\n

While we love that Urban Outfitters is starting to include material, not all of the products are being captured by the eCommerce search filters (likely a problem due to fashion tagging). For example, there are 540 dresses listed, but only 37 materials tagged. This means that shoppers are missing out on products.

\r\n

$\"\"$

\r\n

Battle #4: Easy-to-navigate product filters

\r\n

Winner: Urban Outfitters

\r\n

Urban Outfitters: 👠👠👠👠(4/5)

\r\n

In the Style: 👠👠 (2.5 /5)

\r\n

We found Urban Outfitters filters to check all of the boxes: truncated filters, easy to add and remove filters, and filters like price and size that were strategically located at the top.

\r\n

On the other hand, In the Style filters had categories with only one selection, and seemed to be a long list of attributes without thinking about usability. In the Style filters were straightforward and showed the number of matching products for each search parameter, which is helpful for users. However, we wish that the filters included color chips and labels to make them more readable and easier to navigate.

\r\n

Poorly planned information hierarchy leads to dead-end searches, and frustrated shoppers, making the overall experience overwhelming.

\r\n

We also recommend never showing users a zero-search results page. Users are likely to bounce after seeing the message below:

\r\n

$\"\"$

\r\n

Room for Improvement:

\r\n

We also recommend skipping broad curation categories like “new arrivals” as they’re not impactful, especially for retailers with thousands of products. We’d also like to see less text, and more visual cues from both retailers.

\r\n

Battle #5: Advanced Product Filtering Options

\r\n

Winner: Urban Outfitters

\r\n

Urban Outfitters: 👠👠👠 (3/5)

\r\n

In the Style: 👠👠 (2/5)

\r\n

Faceted Filters: When searching for a “blue, long-sleeve dress,” a user isn’t looking for either a blue dress or a long-sleeve dress; she wants to see the combination of the two. Through faceted filters, users can quickly navigate through product specifications to create their ideal combinations. However, retailers should beware of leading users to dead-ends as they continue to filter.

\r\n

Urban Outfitters has strong faceted filters, allowing users to seamlessly shop their preferences. In the Style, on the other hand, does not offer faceted filters, which does not mirror the way that users naturally search. In the Style includes basic filtering options with limited advanced options in top navigation–and while their filters “get the job done,” they could be more advanced.

\r\n

Mobile UX: We liked that Urban Outffiter’s mobile filters allowed us to select and unselect items, and were generally easy to navigate. However, we would prefer that mobile filters don’t take up the full screen, because that requires users to toggle back and forth between selections, making it more likely that they’ll bounce.

\r\n

In Urban Outfitters, users have a horizontal filter option to select “product category,” but all of the other filter categories require users to click to a vertical filter.

\r\n

In the Style has all filter options in a vertical filter that takes up the entire page, and is difficult to navigate because it isn’t truncated. It seemed like the experience was not especially designed for mobile.

\r\n

$\"\"$

\r\n

Room for Improvement:

\r\n

On smaller screens, space is key–and taking up too much space with text can lead users to miss important information. By incorporating visual cues, which naturally align with the mobile experience, retailers can play on the strengths of mobile (capturing user attention in a highly visual platform) and keep users engaged.

\r\n

The Final Scorecard:

\r\n

How to Take eCommerce Search Filters to the Next Level

\r\n

Bare minimum eCommerce search filter experiences aren’t enough to keep shoppers satisfied. With limited thematic options and laundry-list style product attributes, traditional product filters are filled with obstacles for every type of shopper.

\r\n

Fashion AI technology gives us the ability to re-imagine how shoppers engage with product filters and search. Rather than provide barebones, dated filters, fashion AI makes the search experience more intuitive–meaning more shoppers can purchase the products they want.

\r\n

At YesPlz, we’re re-inventing product filters by:

\r\n

1. Expanding occasion filters to include mood and vibe

\r\n

Our filters aren’t only limited to the most basic thematic filters, we can go beyond basic filters to create mood and vibe tags like “statement” or “romantic.”

\r\n

2. Including visual-first interface in the product filtering and search experience

\r\n

Traditional product filters rely heavily on text, expecting that shoppers will engage with long-winded descriptions. Shopping is, inherently, a visual experience. With YesPlz’s Style Filter, shoppers can visually (and intuitively) search in a way that feels natural.

\r\n

3. Offering automated, AI-powered fashion tagging to reduce inaccuracy

\r\n

When shoppers see inaccurate search results, it’s unlikely that they’ll continue to trust the search experience, and may even bounce. And, inaccurate search results are inevitable when manually tagging hundreds of product attributes per product. By using AI image tagging, YesPlz can accurately tag products in seconds–and use that information to create consistent filters.

\r\n

$\"\"$

\r\n

The eCommerce search filter experience can give your eCommerce the unique opportunity to stand out from competitors–and most importantly, stand out to shoppers. By using AI-powered filter and search solutions like YesPlz AI, shoppers will remember you as a positive product filtering experience.

\r\n

","published":1,"pub_date":1656515100,"publishedon":1656515100,"alias":"battle-of-product-filter-and-search-urban-outfitters-vs.-in-the-style","template":12,"resource_author":"35","resource_cta-text":null,"resource_cta-link":null,"resourcecta":"{\"fieldValue\":[{\"resource-cta_text\":\"Schedule a free 20-minute consultation \",\"resource-cta_link\":\"www.yesplz.ai/contact\"},{\"resource-cta_text\":\"Contact us to learn more about thematic filters powered by AI\",\"resource-cta_link\":\"https://yesplz.ai/contact.html\"}],\"fieldSettings\":{\"autoincrement\":1}}","resource-ebook":"111","tags":"48","chips":null,"preview_picture":"/assets/images/10-2022-updatedbattleofsearch.png","detail_picture":"/assets/images/06-28-22_15.png","subtitle":"Series #1"},{"id":154,"pagetitle":"Traditional vs. Enhanced eCommerce Search: Unlocking New Text Features","introtext":"eCommerce search is an essential part of the shopping experience, particularly when it comes to text search. Users type in their search queries and expect to receive relevant search results. But, traditional eCommerce site search experience is falling short in meeting user and retailer needs. In this article, we demonstrate how you can meet the full potential of your text search, using visual and AI technology from YesPlz. ","content":"

$\"Instant$

\r\n

What is traditional eCommerce search?

\r\n

eCommerce search is an essential part of the shopping experience, particularly when it comes to text search. Users type in their search queries and expect to receive relevant search results.

\r\n

In a traditional eCommerce site search experience (without any AI technology), a users’ keyword search is matched against a manually configured glossary of keywords.

\r\n

Unfortunately, due to ever-changing fashion vocabulary and the complexity of search queries, traditional eCommerce search still falls short in reaching its full potential, leading to poor matching results and frustrated shoppers.

\r\n

The Current State of Search Technology in 2023

\r\n

In 2023, the fashion eCommerce industry has evolved to use algorithms like NLP (Natural Language Processing) to further interpret shoppers’ search language, attempting to define intent and meaning behind search queries more effectively.

\r\n

But, the hypothetical value is often not the same as the reality, because of:

\r\n

1-high prices

\r\n

2-large data sets needed to train the algorithm

\r\n

3- the sheer complexity of search queries

\r\n

The two types of algorithm-powered text search models are:

\r\n

1. Contextual:

\r\n

In theory: Matches user search intent using context clues relevant to the individual user. Search results are specific to a user.

\r\n

Reality: Difficult to provide relevant search results, and expensive to run

\r\n

2. Semantic:

\r\n

In theory: Uses NLP to identify meaning and logic behind search. For example, a “work dress” search is synonymous with “wear-to-work” or “office dress”

\r\n

Reality: Depends on quality of product tags, which may be mismatched.
Too slow to capture up-to-date, modern fashion keywords.
Requires manually updating glossaries.

\r\n

In 2023, the majority of search technology still requires manual configuration, is expensive, and is still prone to errors.

\r\n

And, it’s not that these search models are bad–but there’s still untapped potential in making text search even better, more optimized, and more relevant.

\r\n

But, how?

\r\n

By using fashion tagging at its core, we’ve created text search that combines different types of AI, unlocking a new, revolutionary approach to traditional text search.

\r\n

How YesPlz Is Solving Common Text Problems in eCommerce Search

\r\n

The most common problems in “traditional” eCommerce search are:

\r\n

1) irrelevant search results;

\r\n

2) poor understanding of search intent;

\r\n

and 3) low quality (or missing) product tags.

\r\n

At YesPlz, we’ve built a comprehensive text search system to solve these problems.

\r\n

Our formula to resolve the problems includes:

\r\n

Laser-accurate AI image tagging, to better create matching, accurate search results, without the need for product descriptions.

\r\n

Machine learning that analyzes past user inputs and interactions to predict best search intentions.

\r\n

Fresh, updated keywords to keep search keywords relevant and modern.

\r\n

Thanks to machine learning, we can customize search results for each of our clients’ specific needs.

\r\n

Here’s the YesPlz antidote to inaccurate, one-dimensional eCommerce text search:

\r\n

\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n

	\r\n Outdated Solution \r\n	\r\n The YesPlz Solution \r\n
\r\n Fresh, popular keywords: Continually updated keywords with most popular trends and suggestions to keep search fresh, relevant and accurate \r\n	\r\n Lacks up-to-date, popular keywords. Product discovery isn’t quick or effective. \r\n	\r\n Automatically update with latest keywords based on past user input \r\n
\r\n Autocomplete suggestions \r\n	\r\n No keywords to nudge or guide shoppers. Mismatched search results. \r\n	\r\n Suggested autocomplete keywords, based on popularity or similarity. \r\n
\r\n Instant search results with product preview \r\n	\r\n Search results only show after typing in complete search keywords. High error rates for typos, and unmatched keywords. \r\n	\r\n Updated search results displayed with each keystroke. Actively guides shopper discovery for accurate search results. \r\n
\r\n Additional Filter Paired with Text Search: Extra Side Filter for shoppers to further filter \r\n	\r\n Hundreds of low-accurate matching search results, leading to higher bounce rate. \r\n	\r\n Fully integrated filter alongside search results. \r\n Shoppers filter down further, increasing possibility for add-to-cart. \r\n
\r\n Enhanced image tagging \r\n	\r\n Search results limited to matching text descriptions. \r\n Huge missed potential for matching results. \r\n	\r\n Key product attributes extracted from image, not product descriptions. \r\n Increases percentage of higher matching results. \r\n
\r\n Brand recognition \r\n	\r\n No brand recognition from catalog. \r\n Missed opportunity to help shopper discover favorite products from brands. \r\n	\r\n Brand recognition from search, leading to search results that match brands. \r\n

\r\n

YesPlz Enhanced eCommerce Site Search Solutions vs. Top Competitors

\r\n

YesPlz’s eCommerce site search solutions are designed to stand out to both retailers and shoppers, from its underlying technology to its customizable features.

\r\n

But, what exactly is the difference between YesPlz eCommerce search and top competitors?

\r\n

Here’s a quick overview of all the features your eCommerce could be unlocking:

\r\n

$\"A$

\r\n

The impact of added features for text search:

\r\n

Accurate product tags when using enhanced image tagging, building the foundation for personalization

\r\n

Higher conversion rates– Customers who land on an autocomplete page suggestion are six times more likely to convert than those who don’t.

\r\n

Increased average cart size and volume with additional integrated product filter

\r\n

Lower bounce rate with intuitive UX design

\r\n

Fresh updated keywords keep eCommerce relevant and on top of the latest trends

\r\n

More relevant, precise search results when powered by both visual and text AI

\r\n

The power of the above features, when combined, means outstanding text search that always provides relevant, updated results.

\r\n

Before/After: Creating Enhanced eCommerce Search Solutions for W Concept

\r\n

W Concept’s story:

\r\n

A trendsetting online retailer for luxury fashion, W Concept has captivated the attention of shoppers worldwide with a vast selection of clothing, shoes, handbags, and beauty products.

\r\n

But, W Concept had a problem. Its current text search technology didn’t have instant search, lacked autocomplete to make search intuitive and accurate, and had low matching search results.

\r\n

Here’s the before/after snapshot of how YesPlz transformed W Concept’s text search:

\r\n

1. Before: No automated product tagging, leading to inaccurate search results

\r\n

After: Product tagging for entire catalog, completely automated by fashion-trained AI, that isn’t reliant on vendor information, increasing the accuracy of matching search results

\r\n

$\"YesPlz$

\r\n

W Concept’s robust product catalog also required robust product tags to set the foundation for enhanced text search.

\r\n

No product tags? No problem.

\r\n

With thousands of vendors, W Concept was missing product information, and had some inaccurate information from third-party vendors. YesPlz easily generated these product tags using visual AI.

\r\n

2. Before: Plain text match searching

\r\n

After: Rich search suggestions based on both keyword and similar search

\r\n

By training the data to understand what other styles resemble a search term, we made W Concept’s search more accurate and robust.

\r\n

For example, if a shopper searched for “tank top,” behind the scenes, search results are based on data that understands the number of similar items to a tank top (including products that don’t have the term “tank top” in their description). A sleeveless shirt, ruched tank top, or square neck tank top still fall under the search query–but wouldn’t be picked up by traditional search unless the terminology was an exact match.

\r\n

In addition, autocomplete offers the best of both visual and text based search, with updated keywords on the most popular, modern trends.

\r\n

$\"YesPlz$

\r\n

3. Before: No product preview images, not sparking the visual nature of shopping

\r\n

After: Product previews combined with instant search

\r\n

All it takes is a single keyword to show matching results with a product preview, resulting in a much more intuitive and better quality discovery experience.

\r\n

4. Before: Shoppers needed to click on a search term to filter further

\r\n

After: Integrated side filter helps shoppers quickly filter down as they search

\r\n

W Concept shoppers needed to click on a search term, load a new page, and then were able to filter based on their preferred product attributes. This not only wasted time, but wasn’t providing the most optimized product filtering experience.

\r\n

With an added integrated side filter, shoppers can filter based on their text search (for example, a white tank top text search would lead to further filter options related to brand, silhouette, fit)--with white tank top as a preselected first filter. YesPlz filters are designed to be visual, and UX-friendly, unlike text-heavy filters.

\r\n

$\"YesPlz$

\r\n

By creating a custom, enhanced eCommerce search that solved all of the above problems, YesPlz transformed W Concept’s text search–resulting in always relevant, effective eCommerce site search that converts.

\r\n

Can your eCommerce text search do all of this? When powered by visual and text AI, YesPlz text search is revolutionizing the way shoppers search. If your text search provider isn't doing enough, contact us for a free demo.

\r\n

","published":1,"pub_date":1673870820,"publishedon":1673870820,"alias":"traditional-vs.-enhanced-ecommerce-search-unlocking-new-text-features","template":12,"resource_author":"35","resource_cta-text":null,"resource_cta-link":null,"resourcecta":"{\"fieldValue\":[{\"resource-cta_text\":\"Discover the full potential of eCommerce site search\",\"resource-cta_link\":\"www.yesplz.ai/contact\"},{\"resource-cta_text\":\"Schedule a demo\",\"resource-cta_link\":\"www.yesplz.ai/contact\"}],\"fieldSettings\":{\"autoincrement\":1}}","resource-ebook":"79","tags":"149","chips":null,"preview_picture":"/assets/images/01-13-23_cover1.png","detail_picture":"/assets/images/01-13-23_cover2.png","subtitle":"A Comparison of Text Search"},{"id":195,"pagetitle":"YesPlz AI Stylist vs. Traditional Chatbot for Retail: 2023 Updates ","introtext":"How does YesPlz’s ChatGPT AI Stylist stand against traditional chatbots for retail, regular ChatGPT, human stylists, and AI stylists? We’ll show you why it’s a step above the rest, and why it stands out in comparison.","content":"

The re-emergence of conversational AI is the next frontier in every aspect of our lives. Even Google is considering upending its traditional search results for a ChatGPT-style conversation, showing that the future of search is conversational.

\r\n

Although the fashion industry has experimented with chatbots before, the release of Open AI’s ChatGPT has led to a rush of retailers looking to supplement the eCommerce journey with technology like chatbots.

\r\n

Retailers are also wondering whether ChatGPT can be a standalone solution to provide better eCommerce product recommendations.

\r\n

However, both chatbots and “regular” ChatGPT have limitations in this area.

\r\n

YesPlz’s AI Stylist (powered by ChatGPT for retail) is the world’s first AI stylist that uses combined AI technologies to find curated clothing for shoppers.

\r\n

But, how does YesPlz’s ChatGPT AI Stylist stand against traditional chatbots for retail, regular ChatGPT, human stylists, and AI stylists? We’ll show you why it’s a step above the rest.

\r\n

Online Stylists Are Limited In Efficiency and Shopper Transparency

\r\n

One solution that fashion eCommerce retailers have tried in the past is to use a real, human stylist that connects with shoppers via chat to find clothes. While this was an innovative solution in the past, there are better options in 2023 for retailers looking to make better product recommendations.

\r\n

The limitations of online stylists include:

\r\n

Time-zone Differences: Online stylists are sometimes simply not available in a shopper’s time-zone because of time differences. This requires shoppers to adhere to the stylist's schedule

\r\n

Lack of knowledge of product catalog: Online stylists are humans, and may not remember the entire product catalog or specific product attribute details to make extremely accurate recommendations, leaving retailers with unsold products

\r\n

Inefficiency: Only 1 stylist can serve 1 client at a time, leading to unavailable stylists and shoppers that are waiting on hold

\r\n

Overall, using a human stylist is a missed opportunity for retailers to gather valuable, transparent shopper data.

\r\n

Shoppers may not feel comfortable sharing personal style detail preferences.

\r\n

For example, YesPlz research shows that shoppers want to hide or flaunt their body parts–and it can be daunting to tell a human about these specific preferences.

\r\n

The 2023 Shopper Demands More Than A Traditional Chatbot for Retail Can Offer

\r\n

Traditional chatbots are rule-based computer programs that use pre-programmed responses to communicate with users. They rely on specific keywords or phrases to trigger their responses.

$\"\"$

\r\n

Traditional chatbots for eCommerce have been around for years, and while they have provided some benefits, their limitations are becoming increasingly apparent.

\r\n

$\"Chatbot$

\r\n

Limitations of Traditional Chatbots for Retail Include:

\r\n

Not UX Friendly: With automated responses, it’s difficult for shoppers to explore product catalogs in a user-friendly way.

\r\n

$\"Chat$

\r\n

Not conversational: Chatbots for retail lack the ability to understand natural language, which is a serious limitation for shoppers who do not want to be limited by pre-established choices or bogged down by never-ending menus.

\r\n

No Product Recommendations: Traditionally, they don’t make recommendations. Interacting with a chatbot is more similar to automated filtering, rather than having a two-way conversation.

\r\n

Overall, traditional chatbots are becoming less effective as eCommerce evolves and customers demand more personalized, human-like experiences.

\r\n

$\"YesPlz$

\r\n

ChatGPT AI Stylist uses machine learning algorithms and data analysis to create personalized recommendations based on individual user preferences. Chatbots, on the other hand, tend to provide more general responses that aren’t tailored to specific users.

\r\n

When chatbot knowledge is limited to a website, or rules-based, shoppers have limited engagements with technology.

\r\n

However, with YesPlz’s AI Stylist, shoppers can ask anything, and receive responses to guide the journey. And, responses are generated as the conversation evolves, making the entire interaction more natural than chatbots.

$\"YesPlz$

\r\n

Regular ChatGPT Lacks the Creative Thinking to Recommend Superior Products

\r\n

Most recently, ChatGPT for retail has come front and center in the discussion around integrating technology into the shopping experience. ChatGPT is strong in its ability to generate immediate responses to questions, but as it stands today, isn’t immediately able to integrate with retailer websites.

\r\n

Regular ChatGPT, despite all the hype around it, isn’t enough for retailers to make sophisticated product recommendations. Here’s why:

\r\n

Not fashion trained: ChatGPT isn’t trained in fashion specifically, so recommendations are broad, which doesn’t help shoppers with their specific fashion needs

\r\n

No integration with retailer product catalogs: Instead, shoppers receive long-winded text descriptions that are difficult to read and too general to spark discovery.

\r\n

Lacks the creative “spark” needed to make superior recommendations: ChatGPT responses are underwhelming because it tries to stay neutral, making recommendations dull and uninspiring.

\r\n

$\"ChatGPT$

\r\n

Regular ChatGPT requires more input from shoppers to get recommendations, making the discovery journey more burdensome.

$\"A$

\r\n

YesPlz AI Stylist (Powered by ChatGPT for Retail) Is a Step Above The Rest

\r\n

Enter YesPlz AI Stylist (Powered by ChatGPT for Retail), which uses state-of-the-art machine learning technology to deliver personalized recommendations based on shoppers’ individual style preferences.

\r\n

Recommendations are a step above others because they’re not only based on style or color, but also theme, vibe, and mood. Shoppers are given a curated selection of products based on their prompt, including styling tips and style-with products.

\r\n

And, recommendations are integrated with the retailer product catalog so shoppers can see the full depth of products.

\r\n

Behind the scenes, powerful AI creates real-time personalization. YesPlz combined different kinds of AI including:

\r\n

AI Image Tagging: Our advanced computer vision technology quickly scans product catalogs and assigns product attributes to each item, making it easy to successfully curate recommendations.

\r\n

Generative AI: We utilize Open AI's technology to create responses to prompts

\r\n

NLP: By leveraging NLP, we can better understand the meaning of prompts, even incomplete questions like \"Summer dresses?\", without requiring shoppers to adjust their language to fit the machine.

\r\n

Fashion Transformer: Proprietary to YesPlz AI, we built a fashion transformer that translates OpenAI’s recommendations, as well as product catalogs. Then, it finds the best matching results. Thanks to the fashion transformer, we can process the matching process from recommendations for thousands of products in a second.

\r\n

The end result: personalized style recommendations that mimic the experience of having a personal stylist at your fingertips.

\r\n

How YesPlz Is Solving The Biggest Problems With Chatbots and Regular ChatGPT

\r\n

Here are the biggest problems with typical retail chatbots–and the optimized solution, using the YesPlz AI Stylist (Powered by ChatGPT).

\r\n

Not Fashion Trained: YesPlz’s AI specializes in fashion training, and can therefore understand NLP in fashion contexts.

\r\n

Example: Our AI understands that lavender is a purple color, and that rust is a brown color.

\r\n

$\"YesPlz$

\r\n

No Integration with Retailer Product Catalogs: Since YesPlz’s AI Stylist is based on AI image tagging, our AI already knows the product information (such as attributes, theme, mood) of a retailer’s entire catalog. Then, YesPlz AI talks to Open AI to bring matching products based on shopper prompts.

\r\n

$\"An$

\r\n

Lacks Creative Intelligence: Unlike OpenAI, YesPlz’s AI Stylist can save past conversations (when shoppers opt-in) and analyze the key product attributes from liked items to make even more relevant recommendations.

\r\n

$\"AI$

\r\n

Are you curious to see how YesPlz AI Stylist can integrate with your product catalog?

\r\n

","published":1,"pub_date":1682425560,"publishedon":1682436384,"alias":"yesplz-ai-stylist-vs.-traditional-chatbot-for-retail-2023-updates","template":12,"resource_author":"35","resource_cta-text":null,"resource_cta-link":null,"resourcecta":"{\"fieldValue\":[{\"resource-cta_title\":\"\",\"resource-cta_text\":\"Schedule a free 20-min demo\",\"resource-cta_link\":\"www.yesplz.ai/contact\"}],\"fieldSettings\":{\"autoincrement\":1}}","resource-ebook":"161","tags":"50","chips":"168","preview_picture":"/assets/images/04-24-23_cover1.png","detail_picture":"/assets/images/04-24-23_cover2.png","subtitle":"Comparing Online Styling Options"},{"id":227,"pagetitle":"Fashion Face-Off: Virtual Stylists vs. GPT AI Stylist","introtext":"Virtual stylists are rising in popularity as a way to get personalized fashion advice without leaving the house. Fashion favorites like Nordstrom, Revolve, and other typical virtual stylist services now allow you to access a personal stylist online and receive styling advice over a video chat.\r\nBut, what's the best option for retailers looking for scalable, data-driven personal styling solutions?","content":"

Virtual stylists are rising in popularity as a way to get personalized fashion advice without leaving the house.

\r\n

Fashion favorites like Nordstrom, Revolve, and other typical virtual stylist services now allow you to access a personal stylist online and receive styling advice over a video chat.

\r\n

Typically, shoppers fill out a quiz and are matched, one on one, with a stylist.

\r\n

We believe that personal stylists will always have a place in the fashion world, but for busy, high-volume eCommerce retailers looking to maximize efficiency, accuracy, and scalability, GPT Stylist offers a more comprehensive solution that better fits the specific needs of retailers

\r\n

But, are virtual stylists a good fit for retailers looking for scalable, speedy iterative recommendations?

\r\n

$\"Virtual$

\r\n

The Scalability Challenge for Virtual Styling Services

\r\n

$\"Revolves$

\r\n

Process:

\r\n

Nordstrom or typical virtual stylist services rely solely on limited stylist availability, constrained by human realities like bandwidth, costs, and number of hours in a day. In short, with the limited number of hours available for shoppers, there are only so many shoppers that can be served at once.

\r\n

One review noted that Nordstrom stylists weren’t available on the weekends–a peak time for styling services, leaving shoppers outside the service with limited options.

\r\n

YesPlz ChatGPT Stylist is always available for shoppers, 24 hours a day, whenever they need virtual styling services.

\r\n

Labor Costs:

\r\n

$\"Nordstrom$

\r\n

YesPlz AI-powered technology can handle serving personalized recommendations to an unlimited number of shoppers simultaneously, making it easier to scale without needing to add more stylists.

\r\n

AI can scan through the entire catalog in a short time, for every request from a shopper, this is a big advantage of AI.

\r\n

Human stylists typically take 3-4 hours to have one complete the lookbook. Normally, it would take a number of iterations between a stylist and a client to refine the styling.

\r\n

Growing Product Catalogs:

\r\n

Scalability also applies to growing product catalogs–as retailers scale their product offerings, it’s impossible for virtual stylists to keep up with every new product (and rightfully so). But, that means that as product catalogs grow, virtual stylists will struggle to keep up with them, limiting the conversion rate on new products

\r\n

However, with AI, we can automate recommended promotions based on new products (or seasonal, or any preferences the retailers have).

\r\n

The Power of Collective Data Insights In AI Styling

\r\n

$\"5$

\r\n

Nordstrom stylists are limited in their recommendations by their own personal experiences and knowledge–as are all humans. But, that can lead to inconsistent recommendations

\r\n

Our ever-evolving AI draws on an exponentially larger pool of data points and sources of inspiration compared to any one person–and we can leverage data insights from shoppers to serve curated, personalized recommendations to every shopper

\r\n

AI is a strong pattern detector–and GPT Stylist can put together all of the pieces and nuances that lead to the reasons why a shopper might like a specific item but not another.

\r\n

For example, perhaps there’s a specific vibe or combination of design attributes that the shopper likes

\r\n

YesPlz’s AI gets smarter every second by continuously optimizing its recommendations and style knowledge through machine learning. Users can like or dislike products they see, leading to even stronger recommendations in the future

\r\n

Streamlining the Product Discovery Experience

\r\n

Getting Started

\r\n

Services like Nordstrom require customers to schedule an appointment in advance and coordinate schedules with a stylist. This adds friction and wait times.

\r\n

Shoppers are then tasked with another step–sitting on a video phone call.

\r\n

Then, the virtual stylist sends the products to the shoppers

\r\n

Then, shoppers need to click the links to see the items. Then, finally, they can checkout.

\r\n

But, that’s too many steps–especially when we know that the more steps in the discovery process, the less likely a shopper is to complete it.

\r\n

On the other hand, YesPlz’s GPT Stylist is instantly ready to start the conversation and start styling. Shoppers are never confused about what to write because there are AI-generated prompts to get them started or they simply start typing, leaving shoppers with no delay in starting the initial product discovery journey.

\r\n

The Styling Journey

\r\n

With a virtual stylist, shoppers see a limited number of outfit curations based on the pre-survey completed.

\r\n

But, what if shoppers want to see more outfit ideas? The shopper needs to ask for more outfit ideas, but that may not be realistic in the time-constrained time of a phone call.

\r\n

And, shoppers see a limited view of the product because it’s a video call.

\r\n

What if shoppers want to see more options? It can be difficult to tell another person “next” all the time–but with YesPlz GPT Stylist, we empower shoppers to be as picky as they want.

\r\n

With endless outfit remixes and infinite product refreshes, shoppers can click “no” (or “yes”) as many times as they want to find the perfect outfit combination.

\r\n

Shoppers can be shy in making requests and asking for more outfit ideas. But, with GPT Stylist, they can ask for outfit ideas as many times as they want–or ask uncomfortable questions, like how to hide or flaunt a certain body part.

\r\n

The AI styling process is much more beneficial for retailers–we offer retailers the possibility to see shopper behavior, and to better understand which products are doing well, and which aren’t–this isn’t possible with a virtual stylist.

\r\n

Compared to humans, AI can reiterate repeatedly & faster, making it more cost-efficient for retailers

\r\n

Delivering an Immersive Journey

\r\n

Nordstrom has focused its virtual styling services more narrowly on manually assembling complete outfits.

\r\n

YesPlz GPT Stylist aims to provide an engaging, end-to-end discovery experience beyond just outfit assembly- from curated closets to thumbs up/thumbs down functionality, we’re keeping shoppers engaged during the journey while providing valuable information to retailers

\r\n

Shoppers can actively customize recommendations, explore alternative products, and interact throughout the process.

\r\n

Shoppers rarely only have 1 styling request–or they have a ton of follow-up questions. But, with virtual styling services, shoppers are limited to their initial request (or else the stylist would need to start over). In AI styling, shoppers can make as many different unrelated requests as they want.

\r\n

With retailer product integration, shoppers only need to click to purchase. In virtual styling, shoppers need to wait to receive a message, then take the extra steps to purchase, creating more barriers to conversion.

\r\n

AI Styling Offers More Scalable and Data-Driven Solution for Retailers

\r\n

In conclusion, while personal stylists will always hold a place in the fashion world, the YesPlz GPT Stylist powered by AI offers a more scalable, efficient, and data-driven solution for busy eCommerce retailers.

\r\n

It surpasses human stylists in handling an unlimited number of shoppers, keeping up with growing product catalogs, and providing consistent and accurate recommendations.

\r\n

The AI-driven process streamlines the product discovery experience, empowering shoppers to explore endless outfit options and interact freely.

\r\n

By delivering an immersive journey and valuable data insights, the YesPlz GPT Stylist maximizes efficiency, accuracy, and engagement, making it a smart choice for retailers looking to enhance the fashion advice experience for their customers.

\r\n

","published":1,"pub_date":1690993080,"publishedon":1690993080,"alias":"fashion-face-off-virtual-stylists-vs.-ai-gpt-stylist","template":12,"resource_author":"35","resource_cta-text":null,"resource_cta-link":null,"resourcecta":"{\"fieldValue\":[{\"resource-cta_title\":\"\",\"resource-cta_text\":\"Learn more about AI Styling solutions\",\"resource-cta_link\":\"https://yesplz.ai/contact.html\"}],\"fieldSettings\":{\"autoincrement\":1}}","resource-ebook":"161","tags":"48||50","chips":"168","preview_picture":"/assets/images/08-11-23-2.png","detail_picture":"/assets/images/08-2-23_cover1.png","subtitle":"What's the difference?"},{"id":245,"pagetitle":"Shopify AI Shopping Assistant vs. YesPlz AI Stylist","introtext":"Today, we’re going to compare two popular shopping tools: the YesPlz AI stylist and Shopify’s AI shopping assistant to see which can produce the best fall outfits. ","content":"

\r\n

Today, we’re going to compare two popular shopping tools: the YesPlz AI stylist and Shopify’s AI shopping assistant to see which can produce the best fall outfits.

\r\n

Shop AI Shopping Assistant by Shopify

\r\n

Background:

\r\n

Shop by Shopify recently launched its AI shopping assistant, incorporating the ChatGPT API, and marking its entrance into the AI-powered shopping landscape.

\r\n

Shop exists as a way for consumers to access Shopify brands, and now, it's doing so with the help of AI.

\r\n

With its AI shopping assistant, Shopify is also confirming the need for immersive, engaging shopping experiences to guide shoppers through the difficult task of discovering and filtering products based on complex filter combinations.

\r\n

How Shop Describes Its Service:

\r\n

Shop notes that its AI shopping assistant is designed to streamline the discovery process. By scanning a vast array of products, it offers personalized recommendations, either aligning with the shopper's preferences or suggesting new finds.

\r\n

Our Testing Experience:

\r\n

Shop’s design is undeniably sleek, enhancing the overall user experience.

\r\n

However, the process felt burdensome, requiring the user to repeatedly prompt to actually see search results and extract more detailed recommendations.

\r\n

And, because the tool is simply an AI shopping assistant versus stylist, users receive long blocks of text in response, without integrated product links, resulting in a long, rambling shopping experience.

\r\n

Overall, Shop’s AI shopping assistant is an interesting addition to the Shopify ecosystem, but lacks the in-depth style knowledge of an AI stylist, resulting in very basic, fragmented chats.

\r\n

The YesPlz AI Stylist--More Than An AI Shopping Assistant

\r\n

Background:

\r\n

YesPlz AI Stylist is a transformative AI-powered fashion experience, fusing together deep learning, NLP, and computer vision to offer curated and personalized head-to-toe outfits.

\r\n

The YesPlz Fashion Transformer, the system's backbone, is trained specifically in fashion, ensuring that the recommendations are always on-trend.

\r\n

Our Testing Experience:

\r\n

The YesPlz AI Stylist is designed to be integrated and interactive, offering various touchpoints for shoppers to engage with their styling experience.

\r\n

Users take a short personalized style quiz to frame the conversation, and can initiate any conversation about styles or products.

\r\n

Instantly, users receive a rich variety of head-to-toe styles. Each style recommendation is thoughtfully curated to create a holistic look based on user preferences.

\r\n

With a simple click, shoppers can remix entire outfits, seeing an entirely new look. Or, they can refresh one product in an outfit, to keep the look fresh.

\r\n

Shoppers can also like or dislike products, to better train the AI on their preferences, and in turn, access curated products made especially for them, based on their likes.

\r\n

The shopping experience feels seamless, merging the capabilities of both a personal assistant and stylist.

\r\n

Head-to-Head Test: Finding Trendy Outfits for Fall

\r\n

$\"\"$

\r\n

We posed the same prompt to both tools: \"Find trendy outfits for fall.\"

\r\n

Shopify AI Shopping Assistant:

\r\n

$\"Testing$

\r\n

Filtering Process: We had to repeatedly ask the Shopify AI shopping assistant to show us products, instead of giving us long-winded text responses. Each filter had to be applied individually, making the process too slow.

\r\n

Search Results: The results were a mix of keyword-based recommendations, which at times, resulted in inaccurate search results. The products felt random at times, and there was a noticeable lack of cohesion in the overall outfit suggestions.

\r\n

User Experience: Because only products were shown side-by-side (versus full outfits), it was difficult to imagine how outfits would look together, putting the burden on the user to create full outfits.

\r\n

Fashion Sophistication: When we asked for casual chic styling suggestions for a denim jacket, we were given one answer in response: to layer a hoodie under a denim jacket–and then were shown multiple similar pictures with the style. It didn’t feel like the fashion styling was on par with a human stylist, leaving much to be desired with the experience.

\r\n

YesPlz AI Stylist

\r\n

$\"Testing$

\r\n

Filtering Process: We could make complex requests to the AI Stylist, and receive instant search results. We didn’t need to individually break down each filter request.

\r\n

Search Results: The results felt stylish, accurate for fall, and on-trend with the latest fashion trends, showing looks like an oversized sweater with a crossbody bag, and a plaid blazer with ankle boots.

\r\n

User Experience: It was easy to imagine the outfits when looking at complete looks, and felt easy to interact and customize the styles further with remixes and product refreshes.

\r\n

Fashion Sophistication: The looks felt like on-trend, fashionable looks, versus random products thrown together. Because YesPlz AI trains its AI based on vibe and design attributes, it can pull together looks that are more sophisticated than keyword matching.

\r\n

Final Verdict: AI Stylist vs. AI Shopping Assistant

\r\n

We love that personalization is coming front and center to the conversation around shopping experiences, but not all tools are created equal.

\r\n

YesPlz AI Stylist, with its blend of advanced AI capabilities and fashion-centric training, clearly provides a more intuitive, responsive, and stylish shopping experience. On the other hand, the Shopify AI shopping assistant, though functional, felt limited in its fashion understanding and scope.

\r\n

","published":1,"pub_date":1695833400,"publishedon":1695833400,"alias":"shopify-ai-shopping-assistant-vs.-yesplz-ai-stylist","template":12,"resource_author":"35","resource_cta-text":null,"resource_cta-link":null,"resourcecta":"[]","resource-ebook":"161","tags":"48","chips":"164||168","preview_picture":"/assets/images/09-27-23_cover1.png","detail_picture":"/assets/images/09-27-23_cover2.jpg","subtitle":"Battle of Fall Styles"},{"id":254,"pagetitle":"Using AI in Fashion for Online Stores: A W.Concept Case Study","introtext":"With the rise of platforms like TikTok and YouTube, customers are becoming more discerning about what an algorithm does and expect it to work on their shopping websites, too. W.concept recognizes this trend and is at the forefront of integrating technology and fashion, particularly through the use of artificial intelligence (AI).","content":"

Unveiling W.concept's Renewal Project: Transforming the Modern Shopping Experience

\r\n

W.concept is a rapidly expanding fashion eCommerce platform that ships to over 44 countries and offers more than 100,000 products. Like many other fast-growing fashion retailers, they aim to stand out and improve their service for their increasingly global customer base.

\r\n

With the rise of platforms like TikTok and YouTube, customers are becoming more discerning about what an algorithm does and expect it to work on their shopping websites, too. W.concept recognizes this trend and is at the forefront of integrating technology and fashion, particularly through the use of artificial intelligence (AI).

\r\n

They understand the significance of AI in fashion and have cleverly incorporated it into their renweal website to enhance the shopping experience. YesPlz AI was lucky to be part of this renweal project and today, we'd like delve into how W.concept has integrated AI technology into their platform.

\r\n

$\"Who$

\r\n

W.concept Introduces the Virtual Mannequin Filter, Powered by AI, to Simplify the Shopping Process for Customers

\r\n

With thousands of new products added weekly (including over 16,000 women's tops alone), it's crucial to provide shoppers with a fast and intuitive filtering solution so they can easily narrow down their search. Recognizing this need and the power of AI technology, W.concept opted for the Virtual Mannequin Filter over traditional filtering methods, which often involve browsing through lengthy lists of options. With the Virtual Mannequin Filter, shoppers can now easily customize their search by simply tapping on visual filters.

\r\n

What's even better for retailers? The Virtual Mannequin Filter automatically tags products using fashion AI, from design silhouette to fashion vibes, eliminating the need for manual tagging. This feature saves retailers time and effort.

\r\n

And the results speak for themselves: the Virtual Mannequin Filter has increased the average cart size by 1.7X, making it highly effective for shoppers with specific intentions.

\r\n

$\"The$

\r\n

Everyone Enjoys Knowing 'What's Trending'

\r\n

Shoppers like to easily find out what's popular, which brands are trending, and what they might be missing out on. This simple feature is highly effective in helping shoppers discover new favorite brands. Especially for retailers with a wide range of brands, it's important to introduce shoppers to other cool brands they might enjoy.

\r\n

$\"Show$

\r\n

Automate Occasion Tagging to Ensure Shopper Find Their Favorite Styles Effortlessly

\r\n

Based on our user interviews and surveys, we have repeatedly learned how much our shoppers enjoy shopping for fashion items based on different occasions. They may have a wedding to attend, a date night where they want to feel special, or or professional situations where they need to look polished and put-together.

\r\n

Again, when you're dealing with thousands of new products on a weekly basis, it's almost impossible for your merchandise team to tag different occasions for every single product (while possible, it's probably not the most exciting task!).

\r\n

When AI can understand fashion much like a human does, it can automatically tag which occasion this clothing belongs to. W.concept utilizes this feature to help their shoppers quickly navigate to the occasion they're looking for.

\r\n

$\"Automate$

\r\n

$\"\"$

\r\n

Streamline the Integration of Fashion AI and Get All Types of Fashion Recommendations Your Website Needs

\r\n

Shoppers seek seamless inspiration and discovery while browsing for fashion items. Upon landing on a product detail page, they anticipate further recommendations aligned with their product choices—be it similar items, complementary pieces, or selections resonating with others who share similar tastes.

\r\n

W.concept understands these needs and has integrated multiple product recommendations across various areas, including the shopping cart, quick view, and product detail page, ensuring a smooth and enjoyable discovery experience.

\r\n

Furthermore, W.concept enhances this experience by curating personalized recommendations for each shopper based on their preferred styles or purchases. Shoppers can navigate to 'My Feed' to begin exploring tailored curation.

\r\n

$\"Personalized$

\r\n

$\"Optimized$

\r\n

$\"Optimized$

\r\n

$\"Collaborative$

\r\n

Why Work with Multiple Solutions When You Can Streamline with a Single Solution?

\r\n

Good search and recommendations can only happen when there is a high quality of data available, especially in the case of fashion, where design attributes, style implications, and user preferences matter. YesPlz AI provides 9 product discovery solutions specialized in fashion. By working with YesPlz AI, a retailer gains access to all 9 solutions and future updates without the need to allocate extra resources or spend additional money.

\r\n

$\"All-in-one$

\r\n

$\"Contact$

\r\n

You can explore how YesPlz fashion AIs are integrated into W.concept's '24 renewal website. Enjoy a modern shopping experience with thousands of beautiful indie fashion brands here: https://us.wconcept.com/

\r\n

Do you like what you see? Schedule a demo today and discover how fashion AI can benefit your fashion eCommerce business.

\r\n

","published":1,"pub_date":1707490920,"publishedon":1707490920,"alias":"using-ai-in-fashion-for-online-stores-a-w.concept-case-study","template":12,"resource_author":"33","resource_cta-text":null,"resource_cta-link":null,"resourcecta":"{\"fieldValue\":[{\"resource-cta_title\":\"Curious to See How Our Fashion AI Works for Your eCommerce?\",\"resource-cta_text\":\"Schedule a Demo\",\"resource-cta_link\":\"/contact.html\"}],\"fieldSettings\":{\"autoincrement\":1}}","resource-ebook":"161","tags":"48||149","chips":"165","preview_picture":"/assets/images/2-4-2024-1.jpg","detail_picture":"/assets/images/2-4-2024-2.jpg","subtitle":"Step-by-step case reviews examine each feature how this global fashion retailer is embracing AI"}]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" alt="" width="1172" height="673" />

Your product data will be synced with YesPlz AI, and finally, you’ll decide where to display the recommendations and be able to customize the types of recommendations:

The final recommendations displayed to customers will look like this:

Relevant, fashion-forward product recommendations. A search for a brown, men’s dress shoe will give product recommendations that are within the same fashion category and style.

You can try the Product Recommendation System here--and don’t forget to let us know your thoughts!

Curious to see how the all-in-one discovery solution works for you?

Schedule 20 min Demo Try Live Demo

Follow us on social media

Written by Jess Erdman

Content Marketing Lead

I'm passionate about creating cool content. The best part? I get to learn new things about fashion tech and ecommerce everyday. Have an idea or opinion about this article? Reach out at jess@yesplz.ai

Recommended for you

6 Critical Flaws AI Search Engines Can Solve: Revolutionizing Online Fashion Search

Retailers lose 2 trillion dollars annually, not to competitors, but to Bad SEARCH. Discover how AI search engines fix this costly flaw for fashion eCommerce.

by YesPlz.AI

eCommerce Search Filters Comparison: Traditional vs. AI-Powered Filter

by YesPlz.AI

Blog Home