March 26, 2020
No big seller is left without an online retail store. Some retailers are even closing their physical stores and focusing on e-commerce. For example, in August 2019, Fortune announced that Zara and H&M are “closing their stores to get ahead.” With this move, the fashion giants are answering to the rise of e-commerce and re-locating their resources towards online sales.
And they’re right to do this. Previous predictions said that 2020 will bring us a total number of 2.05 billion online shoppers. However, with the new COVID-19 situation, these numbers are expected to grow even more.
Now that the virus is spreading, 27.5% of US internet users said that they are avoiding public places, and 58% of them will avoid them in the future if the coronavirus outbreak goes worse. Although shopping malls are expected to be the most avoided places, more than half of the users are avoiding shops in general.
This year is going to change e-commerce. We expect most of the shoppers to go online. Besides the good old “enjoying the comfort of their own homes” reason, they will now do this for safety reasons.
That’s why retailers need to create a convenient, user-friendly process of online shopping that will satisfy the needs of shoppers. They’ll have to get ready for a large transformation—from physical stores to e-commerce shops.
However, this isn’t so simple. Especially for shops that have a large number of products in their offering. A large number of products, or visitors, or both, means a lot of on-site events that are hard to manage. For this to be effective, shops need good catalog management, so let’s see how to achieve this.
What is catalog management?
A product catalog is a place where all the online retail store products are precisely ordered. Catalog management is a process that organizes the products in a catalog in a specific way in order to make them consistent and relevant. The optimization and modification of product data are also included here.
The catalog should provide information such as names, categories, price, brand, color, suppliers, and other relevant information. These categories need to be accurately ordered in order to make it easy for customers to find the right product.
Catalog management isn’t always very simple. The more structured the products in the catalog are, the more it helps companies to market their products across channels and increases product exploration, which is an outlet for conversions.
Bad catalog management means unstructured and inaccurate data attached to the products. This can result in several issues:
Chaotic supply chains
Retailers need to have accurate data in order to organize their supply chains. They must know the availability of the products in their physical stores and in their warehouses, and make sure it aligns with what’s offered on their e-commerce store. There are retailers who must be ready to ship millions of products every year, and in order to achieve this, their online and offline data must be reconciled.
Inaccurate data in the product catalog means a discrepancy between demand and supply, leading to a waste of time, money, and full warehouses of unsold goods. Effective catalog management is crucial to achieving optimal costs and efficient delivery. Moreover, it prepares retailers to meet shoppers’ demands in real-time.
A complicated digitization process
New products are coming out every day. All these products that come in the warehouse and then in the physical stores also need to be digitized and presented in the online catalog. The process of digitization comes with creating a lot of data about the product, which can take up some time.
An unorganized catalog makes this process even longer and more complicated. Online retailers must create a smooth digitization process in order to make sure their products become available to the customers immediately after their release.
Incorrect search results
Bad catalog data means bad search results. And this is a great disadvantage since searching is often the first thing website visitors do when they come to a website. If the information attached to the products is incorrect, the search engine will display the wrong results. In order to give customers a delightful shopping experience, they must be able to find the products they’re looking for.
Displaying the wrong results can make the visitor leave immediately. What is more, customers often aren’t very precise in their searches, so the search engine has to be very smart in order to find what they really mean.
Incorrect search results can also lead to more returns, since what people found and ordered might be different from the tagged data associated with it. And this aspect of the online shopping experience has to be well managed because according to Shopify, the search tool can generate up to 13.8% of an e-commerce store revenue.
The solution to these issues is good catalog management. This means a simple and clean catalog that contains all the information, accurately structured.
How to get a well-managed catalog?
Firstly, all products have to be tagged. A catalog can have 70,000 products and they all need to be tagged correctly, with all specific details included. Traditional methods of tagging are really long-lasting and exhausting. Tagging all products in a catalog can take up several workweeks of one employee. This process requires a lot of focus and accuracy, which means that mistakes can occur very easily.
Second, all tags need to be in context. Tags aren’t just about the characteristics of the product. They also have to be in context, so the employee performing the tagging process has to be educated on these specific products.
The growing number of products and their specifications makes manual tagging almost impossible. Luckily, the advancements of AI have brought a solution to this major challenge for online retailers—automatic tagging.
Here’s automatic product tagging in online retail, explained:
What is automatic product tagging?
Every product of an e-commerce store is made up of several tags that are set to describe its characteristics, features, and the category it’s part of. Every product has different tags, so every product is unique.
These tags include everything about the product—color, size, type, brand, use, sale, etc. For example, a dress could have these tags—red, midi, evening, summer, Zara, silk, long-sleeve, summer sale, etc.
Visitors and shoppers are supposed to get information about the product from these tags. Tags play a great part in influencing shoppers in their buying decision. They allow them to filter products based on the categories they want to explore.
Having descriptive tags improves the online store’s filters, allowing visitors to find the products they are looking for in no-time. Powerful descriptive tags bring deeper insights into customer intentions, even without shopping history. They provide smart analytics which helps make smart decisions.
Tagging products is also a way to improve the organization of the product catalog. Not to mention that products become much more relevant when tagged with the correct keywords. However, the process of manual product tagging can be very inaccurate. It’s even more complicated for stores with large catalogs, where product tagging can last forever.
Automatic product tagging is here to eliminate manual product tagging. Automatic tagging organizes and tags photos in the product catalog based on their characteristics, leveraging advanced AI algorithms. Thanks to Deep Learning, these algorithms speed up the tagging process making it automated and eliminating the need for human intervention. This is basically a process that generates metadata for catalog assets. It works in a way that it scans the image and detects features that are connected to particular keywords.
Automatic tagging is a trained AI system that can recognize clothing in images as we humans do. For us, it takes only a glimpse of an image to recognize a clothing item to decide what it is — a dress, a blouse or jeans. For computers this is not an easy task, and that is why the current process of catalogue tagging is done by humans.
With the advancements of Computer Vision and Deep Learning, scientists created Neural Networks that mimic the human brain, and can be trained to recognize what is in an image just as we humans do. These Neural Networks can take an image, process it, and give us semantic information in a form of text.
In order for AI systems to recognize if a fashion item is in an image, they need to learn from a lot of images on how clothing and apparel look like, and what attributes can be used to describe them. And this is exactly what automatic tagging for fashion is. A trained AI system that has seen thousands of fashion images, carefully tagged. Now, it can do the same—tag images with fashion categories and attributes.
During the automatic tagging process, the Deep Learning algorithm processes the pixel content of visuals, like images or videos, extracts their characteristics, and discovers relevant objects. An automatic product tagging model can increase catalog processing time by up to 90%.
With automatic tagging, every picture gets several attribute labels attached. These attribute labels are much more than generic tags—they add deep, specific insights about the products.
After a while, aside from only collecting catalog data, tags gather information about their use, who in particular is using them, and their connections to other tags.
Products don’t have only one tag. In fact, they can have many different tags. For example, an image of a blue shirt with flowers can have several tags attached by the machine learning technology—”blue shirt”, “flower shirt”, “slim-fit”, “formal”, “long-sleeve”, “buttoned-shirt”, etc. Even though the shopper might only remember to look for a blue shirt, the AI technology takes all features into consideration. This allows people who are looking for a blue shirt or a slim-fit shirt to find the same shirt in the product catalog.
Automatic tagging reduces the time it takes to tag products, improves the accuracy of the tags and the website’s search results, and plays a significant role in the reduction of operational costs. Brands can use automatic tagging to reduce human efforts and therefore minimize the chances of making a mistake in the tagging process. Moreover, it can reduce the time-to-market of new products, automating the entire digitization process.
Advanced image algorithms allow the entire tagging process to be automated and performed in only a day, replacing days and weeks of an average worker’s effort. That’s why a lot of e-commerce stores have adopted this technology, making it an essential part of their online retail strategy.
The evolution of automatic tagging
The first “sights” of automatic tagging date from the end of 2010, when Facebook introduced its facial recognition feature. This service was supposed to eliminate manual tagging when people uploaded photos and recognize the people on them immediately. So, people didn’t have to type in their friends’ names anymore—Facebook started rolling out suggestions instead.
Many similar solutions started to come out after that. However, the full potential of this technology was yet to be discovered. Companies realized that these images were full of data and wanted to take advantage of it. The key was in finding out how to process this data and leverage it to grow the business.
In 2015, Google was probably further than anyone else when it came to processing image data. If a user had uploaded a photo, Google was able to recognize what was on them and thus offer them a category—people, places, and things.
In 2017, Facebook rolled out an update that said that their automatic tagging features could now find photos where users were tagged in and help them find out when others were trying to use their images in their profile pictures.
The next upgrade of the technology came from Google that same year, when they started offering various automatic tagging and automatic sorting options. Now you could sort images by date, place, and features, like “red”.
These advancements paved the way for the use of automatic tagging in e-commerce. In 2017, Pinterest was one of the first to introduce automatic object detection, after the launch of their visual search engine in 2015. This new feature made the platform very popular, especially because people were visiting it particularly due to its visual nature, which was now upgraded.
Then, the first sellers started using the technology. Automatic tagging for e-commerce started slowly, but surely, to take over the entire e-commerce space in the past several years. Even social media platforms like Instagram have started using it—pictures now have tags with shopping information on the items that users can buy.
Now, we expect retailers to adopt this technology and improve the customer experience. As deep learning technology evolves, we expect to see many more features that will transform digital selling forever.
Why is automatic tagging important?
Automatic tagging has become one of the most important AI-driven products in the retail market, as a technology that can detect different items in images, videos, or any other kind of visual content.
The automatic tagging engine is visually-intelligent and it can improve the store performance at multiple levels. By extracting various attributes, the technology helps stores create a persona for every customer, giving them a deeper dive into who their shoppers are and what they really want.
With automatic tagging, you should be able to enrich your data through a multi-level product tagging structure, with hundreds of tags at each level. With high-level categories and subcategories, you can add various features that enable you to have more control of your storage.
Moreover, you can also add mutually-exclusive tags that refer to a particular attribute of a certain product. This will allow you to sort your catalog easier and find your products much faster. For example, you can have “accessory” as a category, and “necklace” as a subcategory. These subcategories, on the other hand, can have more attribute tags that enrich product data, like “large”, “golden”, “asymmetric”, etc.
Automatic tagging also results in clean catalogs. This means that all the products get accurate and rich tags that are suitable for their characteristics, enabling better catalog management processes. Thanks to automatic tagging, catalogs get descriptive tags that improve the search engine of the e-commerce store, showing more accurate search results for the shoppers.
These accurate descriptive tags make every information connected to a product more quality, improving the catalog’s metadata. Metadata is the description of other data. These are the keywords that describe an image, so it’s very important that they are relevant and accurate. However, metadata is not a process like photo tagging.
Improving the catalog’s metadata is very important for the overall quality of the website, meaning it will rank higher when it comes to search results, which contributes to higher sales volumes.
The Benefits of Automatic Tagging
Online retailers get tons of benefits by using automatic tagging. Here are some of them:
More time for important tasks
As automatic tagging for e-commerce replaces the manual tagging process, this means that store owners, brand owners, and their employees have more time left for other more important tasks. They can focus on better marketing campaigns, on-demand customer support, and improved store management, without having to hire new people to do all these tasks.
Improved catalog management
Accurate product tags enable streamlined and well-organized backend product processes. Product catalog pictures with precise tags allow stores to track sales, discover the most searched products, find out and eliminate products that aren’t popular, and always have stocks under control.
Automatic tagging can organize products on different bases, like according to brand, design, style, color, and other criteria. More tags can be added from time to time so that the product catalog will always remain updated with fashion trends. These tags can seriously impact buying decisions and therefore bring deeper insights into attribute-level sales analytics.
Incorporating an automatic tagging engine in the e-commerce store can improve the decisions managers make, as it allows them to have a better overview of the products from their online retail store.
Reduce shopping cart abandonment rate
Shoppers abandon their shopping carts for many reasons. Products being out of stock is one of the most common reasons why this happens, leading to an average cart abandonment of 83.6% for some industries.
Automatic tagging can be a solution to this problem, helping e-commerce stores reduce losses. Tags show similar and related products, or products from different brands, giving the users the possibility to choose another product they like and therefore continue their shopping on the website.
Better search engine ranking
Automatic tagging doesn’t only improve catalog management—it can also be a serious SEO booster. Tagging reorganizes the internal hierarchical linking system of the website, placing connections on strategically better positions around the web.
This is especially useful because Google’s AI-based algorithm now understands search context and considers it when ranking results.
Moreover, tags also offer keyword-friendly links to other areas of the website that get clicked by visitors, improving the website’s SEO.
In order to enable automatic tagging to boost the website’s SEO, tags have to be accurate. This way, they can feature in a relevant search result. Automated tags automatically annotate images that have particular categories and reference marks.
Take content creation to the next level
Every online retail store should have a content section. This part of the website can be used to recommend different uses of the same product or to keep visitors up-to-date with news from the industry.
Besides helping customers to find products, automatic tagging can also help managers and content creations create more relevant and attractive content. Accurate product tags help content creators find what they’re looking for fast and in real-time.
Increase customer spend
Every online retail store has the goal to get customers to stay as much as possible on their websites and end their visits with a purchase. As automatic tagging connects products with product tags, it allows store owners to recommend customers more products during their checkout process.
So, customers will find even more related products they like when checking out, adding them to their shopping carts and going away with more products than planned.
For example, if the shopper is looking for a summer dress, they will be looking at dresses with the “summer dress” tag. However, if they are getting ready for the summer, they could probably also use some sandals or t-shirts the system can offer with the tag “summer”.
They’ve probably only come to your store for the dress, but the automatic tagging system is here to remind them to buy other summer items as well. Accurate product tags can enable suggesting other relevant summer items, eventually increasing sales figures.
Accurate predictive analytics
Deep Learning algorithms allow you to detect and understand the performance of your products and their attributes.
With automatic product tagging, you get very valuable insights about your customers’ behavior, preferences, and demographic and geographic characteristics. A larger number of descriptive tags allows you to get more details about what consumers want and create better personalized offers. This allows you to detect and analyze trends, getting your e-commerce store ready for future shoppers.
Moreover, clean and correct tags can reduce returns because buyers will get more information about the product, so they won’t be surprised when the fashion item is delivered to their doorsteps.
Thanks to the data you get from automatic tagging, you can make smart business decisions, reducing business risks with design, prices, and purchase processes suitable for your type of audience. You won’t suffer from overproduction, waste, and price subtraction. This way, you’ll be financially stable and have a more sustainable business model.
Product tagging is a MUST for a well-performing online retail store. Product catalogs that are accurately tagged improve the entire value chain of the store. Automated workflows cut costs and increase profits. This leads to more time to focus on the actual customers, giving them a better customer experience.
Automatic tagging shouldn’t be taken for granted—it should be, instead, understood as the core of the e-commerce store. A well-organized catalog with correct tags improves the entire retail chain, including the automation of human work, cost reduction, and improved product visibility. Most importantly, it improves the personalization of the e-commerce store, providing users exactly what they’re looking for.
Pixyle.ai’s state-of-the-art visual AI will scan your entire product image catalog and name it automatically—detailed and in the blink of an eye. Our AI solutions include visual search, similar recommendations, automatic tagging, and a fully-customizable business intelligence and data visualization platform.
Pixyle.ai’s automatic product tagging engine can help you get a clean catalog, improve your SEO ranking, and increase your sales volumes by scanning your entire product catalog in just a few seconds.