What is Synonym Management

A synonym is a word or phrase that means exactly or nearly the same as another word or phrase in a given language. Without proper synonyms in place, a search query may not match any products, resulting in either:

• A zero results page, leading to a missed sales opportunity, or
• a partial match, where the search engine returns although returns some relevant products, but not all.

You cannot expect shoppers to know your store’s product terminology. Instead, you need a system to understand visitors’ search intent and set up synonyms so they can find what they’re looking for, regardless of the specific product naming conventions.

Adding synonyms to keywords helps increase the relevant result count for different search queries. You can also set negative synonyms to eliminate irrelevant results. There are even more synonym options, such as bi-directional synonyms or product synonyms, you can check this article for the details.

What makes synonym management difficult and time consuming?

• Prioritization

There can be tens of thousands of different search terms that visitors are searching for on a webstore, so it is impossible to set up proper synonyms for all of them, you need to efficiently select the most common terms to handle.

• Finding the right synonym

Identifying the correct synonyms for search terms is challenging. It requires understanding customer intent and ensuring the selected synonyms accurately match the products they are looking for.

Strategies for effective synonym management

Search term prioritization and finding the right synonyms require understanding and analyzing your search data.

When you’re getting started, a common practice is that you should look at popular search queries that have a low click-through rate (CTR).

Another useful tactic is investigating common search terms that shoppers use but then click on different results.

Salesforce B2C Commerce offers built-in Search Analytics that help you understand how users interact with search results. By tracking popular search terms and products, you can make search results more relevant and increase the conversion rate for your online store.

If you would like to have more granular and complex analytics, you can do so with various third-party vendors from the Salesforce AppExchange.

Synonym management in Salesforce B2C Commerce

Salesforce B2C Commerce offers built-in synonym management. You can configure synonyms and negative synonyms for each search term. Salesforce offers brand synonyms too, which are automatically generated for all the products in your product catalog (by the product system object’s brand attribute values). You can turn this feature off.

Using AI Search to minimize manual synonym management

Based on what’s discussed above, you can get a feeling that manual synonym management is a very meticulous and time consuming, but also critical task for your online store.

One of the main use cases for Artificial Intelligence platforms, like Salesforce’s Einstein 1 is that they can minimize such manual optimization tasks, so they are especially helpful in this case too.

AIs helps automating synonym management by the language model they leverage. Their model inherently identifies similar terms, and this functionality can directly be utilized for generating and maintaining synonyms.

AI models utilize vector databases, where every product in a retailer’s product catalog is vectorized. The model then calculates the nearest neighbors of a query vector to return products with the closest relation to the query. This process is natively able to find similar keywords too based on closeness in the vector space.

For example, some SFCC third-party providers offer features that automatically recommend problematic keywords that need synonyms to be set, and some even suggest specific synonyms for these keywords.

You can find even more use cases of AI in the E-commerce industry in this blog post.

Wrap Up

Synonym management is crucial for optimizing search results in online stores, but it is a time-consuming and complex task. Traditionally, retailers manually tag terms and products with relevant synonyms to ensure customers find what they’re searching even when not knowing the exact product names.

Advanced AI technologies now allow for the automation of this process, saving valuable time. AI models, like Salesforce’s Einstein 1, utilize vector databases to identify and suggest relevant synonyms automatically, enhancing search relevancy and boosting conversion rates.

It’s not hard to imagine a future where online shopping goes something like this:

While ChatGPT can recommend products by name, which shoppers can then go search online; it can also provide links directly to online stores, helping the product discovery process:

Having ChatGPT recommend links to products listed directly in your E-commerce store can be an important channel for Ecommerce product discovery and increase your conversion rates and overall customer satisfaction.

We’ll dive into the concrete things you need to do to optimize your online store for product discovery for AI search, but don’t forget to start with the basics – building your brand reputation.

I asked ChatGPT why it selected those specific links and here’s what is said:

So while you’re building your brand reputation, here are some more granular steps to optimize online product discovery for ChatGPT. Let’s get started!

1. Detailed Product Data

When optimizing your E-commerce store for product discoverability on GPT-like search tools, you need to make sure your product data is rich and structured correctly.

Currently, ChatGPT only provides answers in text format (no images), so when thinking about how to make your products appear there, focus on the following:

Metadata and Attributes

You already know the importance of keeping your product data up-to-date. It’s also essential in this context.

To surface products on ChatGPT, make sure each product has comprehensive metadata: product details, availability, categories, tags, and all available attributes (for example: size, color, brand, material).

Be sure to include multiple high-resolution images with detailed alt text so AI systems can understand all the information on the product page.

Use Structured Data

Make sure your data is structed in a way that AI systems (likeChatGPT) can read. JSON-LD is a common schema markup.

Doing this helps ChatGPT understand all of the products’ details.

2. Natural Language Processing (NLP) Optimization

You need to optimize your product listings for natural language processing, too. Think about how people use generative AI tools and how conversational they are. This type of information, and style, should be mirrored in your product metadata.

Keyword Management

You can do this by adding natural language keywords and phrases to product categories or product tags. Think about what people are likely to search for in relation to this product. Even better, think about what they might be asking ChatGPT in order to for it to return this specific product.

If you need some inspiration, you can always ask ChatGPT for guidance and use that as direct input.

Synonym Management

As an online retailer, you already know the importance of synonym management. Relevant and thorough synonym work is also important when optimizing product discoverability for ChatGPT.

When thinking of what (natural) keywords to add to your product description and tags, be sure to add synonyms to have comprehensive coverage across a wide range of search intent.

Conversational Phrases

When creating product descriptions, write them in a conversational tone, so it fits in with ChatGPT’s style.

This gives you a better chance of being picked up and directly used by the algorithm.

If you already have comprehensive product descriptions and not many resources to re-write them, directly use ChatGPT to do so.

3. Content Strategy & Optimization

The more conversational-style text you have on your E-commerce store, the better for surfacing products for ai search engines.

Think back to the first example here, where I asked it for hiking pant recommendations. The final sentence in ChatGPT’s answer is most likely a summarization of user reviews and FAQ answers.

Conversational Content

FAQs and buying guides are great places to start when creating ChatGPT-favored content both about your brand and specifically about your products. You can also create blog posts and product comparisons to help ChatGPT and your shoppers decide which products to buy.

ChatGPT can use this conversational information to elaborate on your products in a wider range of conversations.

Product Reviews

You already know user reviews are important in helping shoppers decide on which products to buy.

These shopper reviews of products might also be one of the most important aspects to include in your E-commerce store as this content can be used to enhance product descriptions and answer questions shoppers ask ChatGPT. 

You can also leverage AI to summarize reviews on specific products, to condense them to 1-2 sentences including pros and cons, to make hundreds of reviews easily digestible for shoppers.

Consistent Formatting

Keep consistent formatting throughout your website so ChatGPT can predictably find relevant information. To do this, use clear headers and bullet points to break up content wherever possible.

4. API Integration

After you’ve put in so much work enriching your product data, getting shopper reviews, and creating conversational-style content, you need to make sure ChatGPT can actually access all that information

You need to provide an API that ChatGPT can access in order to pull all of this information in real-time to make sure it can reflect current availability and pricing.

5. SEO Best Practices

You’re already using SEO to improve product discovery on the web, the same principles apply in this context.

Make sure to focus on the following areas to make your products discoverable on ChatGPT:

On-site SEO: Make sure you optimize traditionally important on-site elements such as: titles, meta descriptions, headings, and URLs for relevant keywords

Link building: Cross-link your internal pages by connecting products and related content to improve navigation. Also, build backlinks from external websites with a high Domain Ranking. This helps build your site’s credibility with search engines.

Loading times and site optimization: Your site should load quickly in order to provide a great browsing experience in addition to improving your SEO. To do this, make sure you optimize your mobile site and follow search best practices for mobile devices.

6. Work with ChatGPT or AI Developers

Working with an expert in the field can help you uncover areas to optimize or help you solve more technical challenges.

These developers can help you figure out how ChatGPT interacts with your product data and which issues are most pressing. They can also help you provide anonymized data to help ChatGPT better understand your product catalog and how it relates directly to shopper queries.

Conclusion

Shoppers will soon use ChatGPT to discover products, so retailers need to start thinking about ChatGPT as another avenue of product discovery to lift customer experience and boost customer loyalty.

By focusing on the strategies outlined above, E-commerce retailers can make their product catalogs and content discoverable by ChatGPT and gain a competitive edge.

This is just the beginning of conversational commerce; retailers can ride the wave further by offering their own ChatGPT-like chat-bot on their online store. Start diving into conversational commerce here.

Role of Product Pages in E-commerce SEO?

There are three main ways leads find your website pages on Google:

1. Your blog articles are found from informational searches. These come from people only looking for information and don’t have a high purchase intent.
   
2. Your home page is found via navigational searches. These people are looking specifically for your brand and probably already know you.
   
3. Your product and category pages are found from transactional searches. These people don’t know about your business yet and want to buy something.

That’s why SEO for product pages is the most important aspect of search engine optimization for E-commerce stores. They attract new customers to your store. Since many of these shoppers are interested in making a purchase the product page can convert them with good visuals, descriptions, and user-friendly navigation.

Investing in product page SEO means you’ll get an influx of new leads who don’t need nurturing, who, in turn, convert on the spot and drive your sales.

SEO Product Page Essentials

Before you optimize product pages, make sure your technical optimization is up to Google’s standards. Check your website with the Lighthouse tool, improve Core Web Vitals if needed, and double-check crawlability and website security. Check out other online shopping problems that are easy to eliminate.

You’ll also have to do link-building by submitting your site to business directories, writing guest posts, and distributing your content. This will increase your site’s authority in Google’s eyes and increase the odds of your product pages ranking higher in SERP.

With that said, here are the nine things to enhance your product pages and gather more organic search traffic.

Keyword Research

The first step towards optimizing a product page for search engines is finding the right keywords. Keywords are phrases Google users typically use to find the products to purchase.

A single product page can be found with multiple keywords. For instance, a product page for a handheld fan can be found with the following keywords:

• Handheld fan
• Personal fan
• Portable fan
• Battery powered fan
• Mini fan
• Fan that runs on batteries
• Handheld electric fan
• Handheld cooling fan

Your job in this step is to find a list of keywords that would represent your product pages. The more relevant the keywords are, the more traffic it can potentially attract. How do you do that? The fastest way is to use a specialized SEO tool. For instance, the SE Ranking keyword search tool can provide thousands of suggested keywords along with important metrics like keyword difficulty, search volume, and search intent.

The last one is especially important—you want only commercial and transactional keywords to represent your product pages. Users with informational queries will be disappointed because they are searching for information, not a product.

If you have already done keyword research before and want to find more similar keywords for a specific product page, you can input the keyword manually and go from there. You can also import your keyword list into the tool to do it in bulk.

You can also get the list of keywords you can start working with by exporting a performance report from Google Search Console, or your product list from the CRM.

Once you have a list of keywords, group them and assign them to individual product pages.

Product Page Content Optimization

Once you have that list grouped, optimize product pages for these keywords. Look up which keyword in the group has the most search volume and use it as the main keyword.

Now, you’ll have to add your keywords to the following places on your product page.

• Product Title: contains the main keyword and combinations with other keywords with high search volume.
• Meta Title Tag: contains the main keyword and combinations with other keywords with high search volume.
• Meta Description Tag: contains the main keyword and is aimed towards convincing users to click on your link in SERP.
• Headings: contain the most prominent keywords, H1 contains the main keyword, and other headings contain supporting ones.
• URL Slug: contains the main keyword.
• Product Description: contains all keywords in the list in an authentic way.

The most important thing when it comes to optimizing the product description is not to overoptimize it.

First and foremost, you want your description to be informative for the user.

Add keywords only where they make sense, and don’t add more than a couple of instances of the main keyword. You can add other keywords just once

Also, add user-generated content to the pages. Having customer reviews on the product page adds unique content and updates the page, both of which are good for SEO.

It may be hard to optimize hundreds of product pages that an E-commerce site typically would have.

If you have to prioritize, start with the low-hanging fruit. Find pages with the potential to rank well—pages in the top 20 for a keyword with low keyword difficulty. 

Image Optimization

Keyword-optimize your image files. To do this, add the main keyword to the alt tag of the image and to the image file name. The alt tag should describe the image, so don’t stuff it with keywords.

There are other areas for optimization when it comes to images. First, keep the size of the images to a minimum. Large image files take a while to load and can slow down the page loading time. This negatively impacts both user experience and SEO for product pages.

Also, try to have the images fit the optimal aspect ratios for appearing on SERP. Your images can appear in rich results and the image search feature.

The best aspect ratio for product images is 1 to 1, with resolutions ranging from 300 x 300 to 1200 x 1200. Make sure the image still looks good when it’s resized in SERP.

Video Optimization

Sometimes, videos can appear in search results, too.

Mostly, videos will appear in informational or commercial searches—searches with the intent to find information on using a product.

The majority of videos linked here are hosted on YouTube. If you have a YouTube channel, make sure to add keywords to the title and description of the video.

If you’re hosting the video on your E-commerce site, compress the video for faster loading, add VideoObject Schema, and create a captivating thumbnail.

Good ideas for product page videos are unboxing and product guides.

Helping Features

Adding helpful widgets relevant to your product can help both SEO and user experience. For instance, a calculator for vitamin pills, an AI try-on feature for fashion or makeup products, or a size-fitting assistant.

The impact on the UX is obvious. In terms of E-commerce product page SEO, it can increase the number of links to your product pages because people love to share handy widgets like that.

Structured Data for Product Pages

SEO is not only about optimizing the page to appear higher in search. It also includes optimizing SERP design. This includes creating a title and description that entices users to click as well as creating rich results like these.

To have your product page display images of the product, rating, and price, use structured data or Schema markup. It’s a JSON-LD document that includes all the relevant information about your product.

There are several ways you can create structured data.

• Write the JSON code yourself.
• Use an AI tool to create it.
• Use Google’s Structured Data Markup Helper.
• Use a plug-in native to the E-commerce platform you use.

Make sure to check your product pages with Google’s Structured Data Testing Tool before you make the changes live.

Avoiding Cannibalization

Often, E-commerce stores have dozens of nearly identical products, like different color variations of the same handheld fan. If you create a different page for each of them, it won’t be good for product page SEO, as Google doesn’t like duplicate content. These pages are also likely to cannibalize each other—your pages will compete with each other in SERP.

To avoid this, don’t create unnecessary pages and group similar products on one page.

Sometimes, you have to, though. For instance, when there’s a sizeable search demand for “pink handheld fan” and “white handheld fan,” and you want to serve your audience by showing exactly what they want.

In this case, make sure to write unique product descriptions for these pages and point a canonical link to the main product page.

Mobile-Friendly Design

As over 50% of internet traffic is mobile, Google prefers to rank mobile-friendly pages higher in SERP, especially so for mobile searches. You can check how well your website works on different devices with the Chrome Inspect tool. It lets you change between different devices and aspect ratios.

Run a Lighthouse report on the page to see how well it loads on mobile platforms. It will show a brief analysis of performance problems.

To improve performance, minimize JavaScript execution time, remove time-blocking resources, and minimize JS and CSS on the page.

If you’re using an E-commerce platform like Shopify, try testing performance with and without plug-ins, especially older ones.

Internal linking

The only way Google can know a page exists is if a link points to it. You need to create internal links to your products to help them get discovered and indexed. A lot of internal links pointing to a specific page will signal to Google that the page is more important than others.

To help with crawlability, create a user-friendly linking system. Link to your product pages from:

• Navigation menu
• Sitemap
• Category pages
• Other product pages

The last one can also improve your site’s UX and average sale value, as they often are in the form of product suggestions.

Use a plug-in to create those product suggestions automatically for each product page.

If you want to improve the SEO of a specific page, for instance, a page that has the potential to rank well for its keywords, leave additional links from your website.

Your best options are adding a link from the homepage in the form of a banner or a navigation menu link, then adding links from your blog content. Prioritize it in product suggestion sections or add a bestsellers section.

Summary

Optimizing product pages for search doesn’t take as much work as off-page SEO. Follow this guide and see how the positions change in a few weeks when the page is indexed.

Change your approach as you track the key metrics and work closely to optimize your pages, starting with the ones that have the most potential. Once you find the right SEO approach to your specific market based on these pages, expand it to the rest of the site.

Author:

Understanding Seasonal Trends with Analytics

To prepare for the Black Friday period, understanding trends in shopper behavior is key. 

You can use information about seasonal trends to not only inform pricing decisions, but also to create marketing campaigns and merchandising strategies.

Start by analyzing customer behavior patterns to forecast future trends. For instance, identify products that start to gain traction as Black Friday approaches. Additionally, review the types of products and categories that were popular during the same period last year to anticipate this year’s demand. 

In addition to looking at historical data, you can leverage the Salesforce Holiday Predictions 2024 report to understand trends and shopper expectations for this holiday period. Here, Caila Schwartz, Research Insights Director, analyzes aggregated and anonymized data from Salesforce Commerce Cloud customers then uses a model to extrapolate it to the rest of the market. Using voluminous data like this can help inform how this season will be different from prior. 

Once the holiday period starts, continuously track customer preferences by monitoring search and purchase data to pinpoint popular products, ensuring you stock up on items that are likely to drive sales.

These data-driven tactics ensure help in marketing efforts and inventory planning to maximize revenue-generation during the holiday shopping season.

Improve Product Discoverability

After identifying the products to discount this holiday season, the next major decision is how to make them easily discoverable.

Site search is one of the most important areas of your online business because if shoppers can’t find what they’re looking for, they can’t buy it.

Based on data from Prefixbox, approximately 27% of online shoppers use the search function on a website and generate 71% of all online revenue. That’s because shoppers who use the search function have a high purchase intent because they are looking to buy something specific.

Even though search is so important, retailers often miss the mark because 78% of consumers reported that the site search function on retail websites sometimes provides irrelevant results. This has a huge, negative impact as 68% of shoppers report not being willing to shop on a site with a poor search experience.

That is a shame because retailers who leverage an AI-powered search solution can increase core KPIs by 25% (based on Prefixbox’s data).

In order to make products easily discoverable, you need to think about your search performance in addition to product recommendations.

There are essentially 3 main areas to handle when it comes to enhancing product discoverability.

1. Optimize your search. To do this, follow site search best practices. Additionally, make sure your product meta-data is optimized for discoverability. Not only does the product need to have the right attributes listed, but those attributes need to be analyzed in an intelligent way (think by relevance and popularity), so that the right product can be surfaced at the right time. 

2. Improve on-site navigation to highlight promoted products on as many pages as possible, using personalization, merchandising, and advanced product listing page features.
   
3. Drive traffic directly to promoted products’ pages. For this, use carousel-type social banners to showcase and link to specific product pages.

Power of Product Listing Pages

Retailers should create standalone custom Product Listing Pages (PLPs, or Category Page) for Black Friday, including only promoted items. These can be further segmented by category, including only discounted products in that category, as shown on the picture below.

These can serve as the landing pages for your holiday marketing campaigns, such as:

• An email campaign to loyal customers or subscribers offering early access or special discounts 
• A social carousel banner that drives traffic to specific product pages, with the main Black Friday page featured as the first or last slide of the carousel.

A dedicated page makes it easy to create audiences of Black Friday shoppers for retargeting in current or future campaigns. Additionally, featuring only the promoted products improves user experience and helps reduce churn.

Since these pages can leverage the same advanced features of Search Engine Result Pages, they have dynamic filters and rich product tiles that let shoppers quickly add products to their cart. 

It’s important to leverage personalization and AI ranking on Product Listing Pages because many products can appear here. 

In order to increase engagement on these pages, have them personalized to individual shoppers to highlight the brands they prefer or products in their sizes. 

Additionally, AI re-ranking is important here to make sure the most popular products appear at the top. This way, shoppers can quickly find what they’re looking for and proceed to check-out.

Merchandising

Merchandising is one of the most powerful tools to drive product discovery during the holiday season.

Merchandising tools enable retailers to promote the products they want to sell by creating banners, ads, and by giving them the power to override Search Engine Results Page product ranking to place promoted items in top positions, generating higher visibility.

When creating your merchandising strategy, don’t forget to factor in your inventory and continually monitor your campaign performance and stock levels. There’s nothing more frustrating than accidentally promoting an out of stock item, or under-promoting inventory that needs to be cleared out. 

On the topic of using merchandising to clear out stock, think about using clever, new ways to freshen old inventory in order to spur last minute checkouts. For example, you could curate a holiday look made from slower-moving inventory. This can help make the products relevant and desirable.

Homepage

Other than dedicated landing pages the marketing team drives traffic too, the homepage is one of the most frequented pages on the website, so it’s important to make the most of that real-estate.

Add banners or navigation tiles that can direct shoppers to different category pages like “biggest discounts”, “best deals”, most popular brands, newest products, products with the highest ratings.

PLP (Product Listing Pages)

Merchandising here can be useful to override the search results so retailers can pin products with the best margins to the top of the page or to create dedicated branded landing pages.

SERP (Search Engine Result Page)

The Search Engine Results page is a great place for merchandising to both promote the products you want to sell and to also streamline the shopping journey and help shoppers navigate your site easier.

Intent clarification tiles can be placed at the top of the SERP to help shoppers quickly narrow down results to simplify the product discovery process.

Use ads and promotions to pin products you want to highlight (either because of great margins or stock levels) to the top of, and above, the SERP so shoppers see them more frequently.

It’s important to select the products in the first two positions on the SERP because based on data from Prefixbox’s analytics portal, products listed in the first position have a 63% higher Engagement Rate than products ranked lower. And products in the second position have a 30% higher Engagement Rate than those ranked lower.

If there are products you really want to push this holiday season, make sure they’re ranking highly!

Promotional Campaigns

During the holiday period, marketing campaigns shouldn’t end at driving traffic to the website; marketing teams should think their campaigns through all the way to check-out. For example, teams should create a cohesive shopping journey leveraging personalized campaigns that drive shoppers to relevant category pages and retarget those who abandon their carts. 

By using marketing tactics like retargeting and audience segmentation to deliver specific messages to different user group profiles, retailers can deliver a seamless, personalized journey and will benefit from higher revenue, great conversion rates, and increased customer loyalty. 

To create effective promotional campaigns, make sure you leverage data:

• Using profile and behavior identifiers can help you evaluate whether to actually give that 10% off welcome coupon to a specific user.
• Targeting promotions based on likelihood to influence means that if a promotion won’t improve conversions, it can be banked for a different shopper. This drives promotion effectiveness and saves on cost.

Personalized Experience

72% of shoppers prefer shopping online mainly because it is more convenient than going to traditional brick-and-mortar stores. With personalization, retailers can make shopping even easier and enhance the overall user experience. At this point in E-commerce, providing a personalized shopping experience is expected.

When creating your E-commerce personalization strategy, think about using the data that’s already available in your Customer Data Platform (CDP). The user profiles in your CDP can not only be used for creating marketing campaigns, but also for enhancing site search functionality too. 

This includes creating personalized experiences on Product Listing Pages (like Black Friday collection pages) and highlighting preferred product types, or brands, based on historical purchases, for example.

Personalization is the driving expectation of today’s shopper. Knowing when to pull back or reach out is just as important. Adding in case suppression and identifying consumers likely to disengage in order to inform your messaging strategy can lead to a lower unsubscribe rate.

Cross-Selling and Bundles

Don’t forget about cross-selling and bundling products to increase revenue during the holiday period.

For example, combine promoted items or products with high inventory levels with your most popular items using product recommendation modules. These modules can be strategically placed on product detail pages and cart pages to encourage additional purchases.

Enhance visibility and attract more attention by adding badges to your Black Friday product bundles, making them stand out to shoppers. This not only increases product visibility but can also significantly boost conversion rates. 

Don’t forget to leverage data when crafting your bundling strategy. Consumer insight and market basket analysis could make bundles more effective as they are created around products frequently moving together. And bundling in products that have a lower attach rate could mean clearing them out faster.

Integrate Search Analytics and Merchandising

To make the most of the holiday period, retailers shouldn’t forget about data monitoring during and after the sales period. 

E-commerce specialists can only make informed, data-driven decisions if they follow the most important KPIs closely. To do this, set up automated reports or dashboards to continuously, and easily, stay up-to-date. 

Be sure to set up automatic alerts (e-mail or platform notifications) for business critical information. For example, to be updated immediately when an item goes out of stock, so you can stop, or switch, merchandising promotions.

Some of the most important KPIs to keep track of:

Product Revenue

Helps identify products sold the most, so you can adjust promotional campaigns or cross-selling strategies accordingly.

Search Click-Trough Rate

The proportion of searches with any products clicked on the Search Engine Results Pages. Helps evaluate the relevance of the site search engine, which can be improved with data-driven personalization.

Cart Abandonment Rate

Percentage of shoppers who add items to their cart but do not complete the purchase. These users can be retargeted through e-mail or other online campaigns, offering special bundles or discounts to encourage them to finish the process. 

In addition to retargeting, identifying reasons for cart abandonment could lead to better conversions. For example, adding more details to product descriptions or better photos, or 3D imaging could mean less hangups at checkout. Or if a shopper is held up by high shipping costs, offering a shipping promotion in the abandoned cart email can help them convert.

Banner Impressions and Clicks

Number of times merchandising banners are displayed to and clicked by shoppers. Essential for evaluating the performance of promotional campaigns.

Promotional Items Purchased

The number of purchases of items highlighted by merchandising tools. Evaluate campaign performance by comparing purchase numbers to a pervious period, or to non-promoted items.

Top Search Queries

Shows the most popular terms shoppers type into the search box. 

Zero Result Search Rate

Measures the proportion of searches that do not return any relevant results. Effectively reduce this by adding synonyms and setting up redirect rules. 

Since zero result pages can’t be completely eradicated, maximize them by adding Related Searches or Related Products so shoppers can easily continue their shopping journey. 

Search Success Rate

The percentage of product search sessions where users engaged with (clicked or carted) at least one product on the search engine result page.

A low SSR signal issues with search relevance and is a sign you should dig deeper into search performance.

In addition to the above, it’s important to bring together marketing and E-commerce data to help you identify top performing campaigns. 

You can take your analysis even further by checking out lesser-known metrics like Promotional ROI or Sale Velocity to get further insights. 

Conclusion

This holiday season, leverage the power of advanced site search tools such as Merchandising, Product Listing Pages and AI-driven personalization to significantly increase your E-commerce sales. 

Analyzing seasonal trends and customer behavior allows you to anticipate demand and stock up on popular items. 

Enhance product discoverability by creating dedicated landing pages and using merchandising strategies to highlight key products. 

Personalization further improves the shopping experience by displaying relevant items first based on user preferences. 

Monitor essential KPIs like product revenue, search click-through rate, and cart abandonment rate to make data-driven decisions and optimize your strategy in real-time to make the most of the holiday shopping season.

Find out more ways to get ready for the holidays in Salesforce’s comprehensive holiday preparation guide.

Authors

Headless Commerce: Flexibility and Performance

Headless solutions decouple the front-end (presentation layer) from the back-end (commerce functionality). This separation allows businesses to use any front-end technology to deliver content, leading to highly customized user experiences and customer journeys across different channels like their website or mobile app. Check out some of the most popular headless platforms.

Benefits of Headless Commerce Solutions

• Enhanced Customization: With no fixed front-end, businesses can create unique user interfaces tailored to their brand and customer demands.
• Improved Performance: Front-end changes can be made independently of back-end updates, ensuring faster load times and smoother user interactions. The front-end system is more scalable with this approach.
• Seamless Integrations: APIs facilitate integration with various third-party services, enabling businesses to add functionalities without overhauling the entire system.
• Predictable costs: while it might be more cost-intensive to get started with headless commerce architecture, the associated costs are more stable and predictable over the long-run.

Challenges of Headless Commerce Architectures

• Complex and Custom Development: Requires more sophisticated development skills to manage and integrate different components. This can also lead to higher overall maintenance costs.
• Higher Initial Costs: The need for custom front-end development can result in higher initial investments.
• Security and Performance Issues: managing multiple systems and APIs can lead to more risk exposure and if not properly optimized, your site can suffer from latency issues.

Composable Commerce: Modular and Agile

Composable approach takes headless commerce platforms a step further by allowing businesses to select and assemble best-of-breed commerce components (so called Packaged Business Capabilities – PBCs) to create a tailored ecosystem.

Packaged Business Capabilities (PBCs) are modular software components (a collection of microservices) that encapsulate specific business functions. Each PBC operates independently, enabling rapid innovation and customization, as companies can easily add, replace, or update individual capabilities without affecting the overall system.

PBCs and microservices a preferred over a monolithic architecture where everything is written in a single piece of code as microservices act like puzzle pieces where each section is responsible for 1 area.

Benefits of Composable Commerce Platforms

• Unmatched Flexibility: E-commerce businesses can choose specific components for functions like checkout and search, optimizing their technology stack and improving the overall customer experience. This lets you easily adapt your E-commerce set-up to your business’ unique needs.
• Rapid Innovation: Components can be easily added, replaced, or updated, allowing businesses to quickly adapt to market changes and add new technologies with this modular approach.
• Cost Efficiency: The pay-as-you-go business model enables E-commerce retailers to invest only in necessary components, reducing waste and optimizing budget use. It also removes the need to have a large development team as you don’t need to frequently make back-end changes.

Challenges of Composable Commerce Solutions

• Integration Complexity: Managing multiple vendors and ensuring compatibility between components can be challenging.
• Ongoing Maintenance: Each component may have its own update cycle, requiring continuous attention to maintain a seamless operation.

Key Differences between Composable and Headless Commerce

To wrap up, these are the key areas where the two technologies are most distinct:

• Scope of Modularity: While headless commerce focuses on separating front-end and back-end, composable commerce architecture involves selecting various independent components to build a complete system.

• Flexibility: Composable architectures offer greater flexibility with their modular approach, allowing online businesses to customize every aspect of their E-commerce solution.

• Implementation and Maintenance: Composable commerce platforms may require more extensive planning and integration efforts due to the variety of components involved, whereas headless commerce solutions primarily deal with front-end and back-end separation.

Summary

Choosing between headless and composable commerce solutions depends on your business needs, technical capabilities, and long-term goals. Headless architectures provide a robust foundation for custom front-end development, while composable commerce offers unparalleled flexibility and adaptability through its modular architecture. Both approaches can significantly enhance your E-commerce capabilities, ensuring a seamless, high-performing, and innovative online shopping experience.

For more in-depth insights, visit check out our separate detailed guides on headless commerce and composable commerce.

What is Composable Commerce?

Composable commerce is made possible by headless technology, but the two are not interchangeable.

Elastic Path describes composable commerce as the process of composing the best possible commerce components to create a cohesive solution that meets your unique business needs.

These components, often referred to as “composable building blocks” and Packaged Business Capabilities (PBCs), handle specific E-commerce functionalities such as product catalog management, search, checkout, payments, etc., and are connected via APIs.

Businesses can easily put together multi-vendor commerce solutions based on their preferences and business requirements.

This modular approach has allowed online businesses to gain more control over the customer experience across multiple touchpoints. Composable commerce platforms also enable retailers to quickly adapt to changing market demands or experiment with new features by easily swapping out or adding new components without disrupting the entire system.

MACH architecture

We can’t talk about composable commerce without mentioning MACH architecture.

The letters stand for Microservices, API-first, Cloud-native SaaS, and Headless. MACH is a set of technology principles that enable businesses to build best-of-breed, modern platforms, and is the first building block of composable commerce.

Let’s take a look at each of these “letters”:

M – Microservices: These are individual, independently developed and managed pieces of software. The composability of Packaged Business Capabilities (PBCs) is enabled by a microservice-based architecture, as PBCs are built up by one or several aggregated microservices.

A – API-first: The storefront is connected to the back-end system through APIs, enabling several applications or services to tie together.

C – Cloud-native SaaS: Software-as-a-Service that leverage cloud features, like automatic scaling and high availability. This software automatically updates, so you don’t need to worry about making upgrades.

H – Headless: Headless architecture is the base of the composable approach. As the front-end of the storefront is completely decoupled from the back-end, it allows retailers to easily update the user interface. The application can be tied to several endpoints, different channels or devices.

Salesforce is a proponent of composable commerce. It enables retailers to work with a composable storefront (“an industry-standard framework for creating innovative Progressive Web Apps”).

Salesforce recommends composable storefronts to all those businesses that:

• need more operational agility to stay ahead of customer expectations
• want flexibility and freedom in choosing which third-party solutions to integrate that match their objectives best
• strive to create unique digital experiences.

Benefits of Composable Commerce

Increased speed

By separating the back-end and front-end systems, you eliminate the need for requests to pass through both. With the help of headless solutions and/or a composable commerce approach, many requests can be handled on the front-end. This results in faster loading times, which corresponds to a better user experience for your shoppers.

Greater flexibility and agility

The modular approach of composable commerce architecture allows businesses to easily swap out or add components as needed, without interrupting the rest of the system.

It heavily relies on APIs (Application Programming Interfaces) to facilitate communication between different components, which defines how they interact with each other, allowing for seamless integration and interoperability. APIs make it easier to connect with external systems and third-party services, further enhancing flexibility and agility.

For example, if a business wants to change its payment processor or add a new feature to the checkout process, it can simply integrate a new component via APIs without needing to rebuild the entire platform.

Reduced cost

The pay-as-you-go model that the modular architecture of composable commerce enables means that retailers can choose to only pay for the components they need, avoiding overinvestment in features their business needs don’t require.

This is a significant advantage compared to monolithic E-commerce platforms.

More scalability

Traditional, non-composable E-commerce platforms often present challenges with scalability since they are usually originally built as a single, tightly integrated system. As a retailer’s business grows, they may encounter limitations in terms of system performance, capacity, and the ability to accommodate increasing traffic or transaction volumes. Scaling such monolithic architectures and systems typically requires significant investments in hardware, software, and infrastructure upgrades.

In contrast, composable commerce solutions enables retailers to scale their technology stack more dynamically and efficiently as they grow, minimizing unnecessary costs by this all-in-one platform approach.

They can scale

• horizontally by adding additional instances of specific components to handle increased loads
• vertically by upgrading individual components to handle higher demands.

Challenges and considerations for retailers

While composable architectures offer many advantages, it’s always beneficial to be aware of any challenges that may arise with adapting a new, modern approach.

Challenges rooted in integration

Integrating multiple software components from different vendors or providers requires careful planning and coordination. Each of these components may come with their own set of APIs, data formats, and protocols, especially if they are developed by different teams instead of a single vendor.

Testing the separate components together might be the best way to ensure that they’re 100% compatible and interoperable, which also requires effort and time.

Maintenance of individual components

It can happen that the different components of a system have different update cycles, dependencies, and compatibility requirements, which can increase the overall maintenance burden.

Dealing with several vendors can also add to the complexity and the cost of the processes, but an efficient coordination of communication and support helps a lot with ensuring the smooth operation of a composable system.

Does composable commerce impact your customers?

In short: yes!

Composable commerce can have several significant impacts on your online shoppers.

As 80% of shoppers agree that the experience a company (including their website) provides is as important as the products and services they offer and sell, now is the best time for brands to embrace innovation and stay on top of trends.

Now, let’s see how composable commerce can assist in this matter:

Faster and smoother customer journey

Increased speed is not only beneficial for your business operations, but contributes to higher customer satisfaction and an increased conversion rate. By decoupling the front-end and back-end systems through headless commerce architecture, composable commerce allows for faster loading times and smoother user interactions. Shoppers experience quicker page load times, seamless checkout processes, and reduced friction when navigating the website or app.

Personalized shopping experiences

With access to a wide range of specialized third-party tools and technologies (and by choosing those that offer data analytics and utilize AI algorithms), you can personalize the shopping experience for each customer, deliver product recommendations based on previous user behavior, display promotions and special offers.

Access to innovative features

AI-powered product recommendations are not the only innovative feature shoppers can enjoy access to.

If you decide to utilize other best-in-breed components in your store, your shoppers can have virtual try-ons, ask for assistance from interactive chatbots, or engage in an entire social commerce experience.

By prioritizing personalization, performance, and innovation, you can build loyalty, drive sales, and differentiate your store and brand in a competitive landscape.

Wrap up

In summary, composable commerce marks a significant evolution in E-commerce strategy, enabling businesses to swiftly adjust and innovate according to market changes.

By breaking down commerce functionalities into interchangeable modules, companies gain flexibility, scalability, and agility in their online endeavors. Embracing this method unlocks avenues for tailored solutions, seamless integrations, and enriched customer interactions, positioning businesses at the forefront of digital commerce innovation.

Prefixbox’s solution optimizes the search experience on all headless and composable storefronts. What’s more, we are deeply integrated into many of these platforms, including native integration with Salesforce Commerce Cloud.

Heard about conversational commerce, but not sure how to get started?

Don’t worry— we’ve got you covered.

This new trend, often called chat commerce, is reshaping the way businesses connect with their customers, and those who get a jump-start, will be poised to gain a competitive advantage.

For a comprehensive understanding of conversational commerce, check out our previous articles covering its fundamentals, use cases, and implementation strategies.

Leveraging any of the wider known messenger apps (e.g. Facebook Messenger, WhatsApp Business, Viber, etc.) starts by choosing a conversational commerce platform that helps you with this kind of implementation. Oftentimes, they are quite versatile and offer other features for E-commerce businesses, besides the opportunity to be in contact with their customers directly. 

Why to Use a Conversational Commerce Platform?

Let’s take a quick peek at the most important features of said platforms:

• automated conversations through chatbots

Powered by artificial intelligence, they can engage in real-time conversations with customers, answer their queries, and provide support.

• personalized experiences and recommendations

This is a standout feature of conversational commerce platforms, thanks to their ability to access customer data, such as purchase history and preferences. It enhances customer satisfaction, customer loyalty, and ultimately increase conversion rate. 

• omni-channel communication

When choosing a platform, you first need to make sure it’s compatible with the messaging channel you plan to use. To choose the best channel for your business, make sure you consider factors such as your target audience and geographical location, and the specific features offered by each to align with your business objectives.

An omni-channel communication platform is a great choice if you’d like to engage with your customers across multiple messaging channels, be it social media platforms, messaging apps, and ecommerce websites. Ensuring that your customers can interact with you in the way they prefer enhances their convenience and satisfaction.

• the use of NLP (natural language processing) technology

NLP enables the platform to understand and interpret user queries in a more human-like manner. By comprehending the context and intent behind customer messages, you can deliver more accurate and relevant responses to them. 

• advanced analytics and reporting capabilities

Thanks to this feature, you gain insights into customer interactions, track key metrics, and measure the effectiveness of your conversational marketing strategy. This can help you refine strategies, optimize customer journeys, and drive better outcomes.

Choosing Your Ideal Platform: A Strategic Decision

When selecting the ideal platform for your business, opt for one that suits your business requirements and objectives, plus, integrates seamlessly with your current systems.

Evaluate aspects such as scalability, customization options tailored to your unique requirements, and compatibility with your E-commerce platform to make an informed decision.

There are two major use cases of conversational commerce platforms that stand out from the rest: 

1. Support

Through the integration of chatbots and AI-driven systems, businesses can provide instant assistance to customers, offer 24/7 support, and address their concerns in real-time.

Intercom’s has long been a popular support platform and is now integrating AI chat capabilities to support conversational scenarios. Their chatbot, called Fin, is powered by GPT-4 and is available 24/7 and chats with your customers to resolve queries based on your support content. 

You can see a demo of Fin in action here.

2. Sales 

Conversational commerce tools can take the offline shopping experience, online by replicating the experience of sales assistants who engage potential customers in personalized conversations. They can recommend products, answer customer inquiries, and guide users through the purchasing process seamlessly. 

Chatfuel is a chat commerce platform that has ready-to-use templates that you can use to engage your customers on other channels. This can help you build workflows to automate cart abandonment reminders, track orders, and more. 

Types of Conversational Experience Platforms (and How Can You Benefit from Using Each)?

Each conversational commerce platform can provide your business with unique features and benefits, however, in order to choose one you need to clarify what your objectives are and what purposes you would use it for.

Let’s see the most common solutions that conversational platforms support:

Build-your-own AI Chatbot platform solutions

These messaging platforms enable you to build your own chatbot to leverage the power of AI in customer support and engagement. The communication channels are up to your (and your customers’) preferences.

Instant chatbot support is quite an attractive feature for online shoppers when it comes to getting help with a service.

💡Pro tip: to maintain a balance between automation and human touch, provide the option for users to connect with human agents when they feel their case requires it.

The next level of automated E-commerce messaging is when the entirety of the shopping journey is carried out with the help of chatbots.

AI Customer Engagement solutions (Marketing Automation)

The purpose of these solutions is to assist your business in creating different campaigns for increased customer engagement. They help you find the most relevant content for your target audience, alongside the best communication channels and the recommended time of publication. 

Furthermore, automation promotes speedier communication between customers and businesses, allowing for faster problem-resolution times at cheaper costs than human processes.

Cloud Contact Center solutions

As a 2021 study shows, 77% of contact centers that already used AI for multiple purposes back then, reduced their overall costs as a result. 

In the context of contact centers, AI-powered bots are able to handle first line of customer interactions and free up some time in your customer service team’s already busy phone line. 

Chat commerce solutions can help you connect different digital channels to ensure the same level of service, regardless of the way your online shoppers decided to get in touch. Usually, you can access all relevant channels from one interface, alongside the information related to your customer’s profile.

AI Customer Data platforms

These platforms let you gather, store, and analyze data about users to create customer profiles and enable you to personalize every interaction with them for enhanced user experience. With the help of AI, you can recommend your products and services to the right audience, based on individual profile data. 

As we mentioned at the beginning of this article, there are platforms that offer numerous different features for businesses instead of focusing on one of the developmental areas. These are called conversational experience solutions.

Case Study: Infobip

Infobip is a great example of an omnichannel conversational experience solution. They are a CPaaS (communication platform as a service) company that now also empowers businesses to integrate chat commerce seamlessly into their operations.

Infobip’s multifunctional services include all of the above-mentioned features and solutions. 

They enable you to communicate with your customers over their favorite channels. By expanding your communication solutions and channel portfolio, you can create new revenue streams and win new business.  

Infobip allows you to craft end-to-end customer journeys powered by AI and build your own Gen AI-based chatbot by your own rules. They also have a solution that lets you create customer profiles as well, for maximum personalization. 

“Creating connected customer experiences is at the heart of what we do“

To enable your shoppers to reach out directly to your business with inquiries, be it about general product recommendations, needing support with choosing the right product, or having trouble with payment, you can choose Infobip’s chatbot platform, and imitate the feel of a human conversation with many times fewer effort. 

Wrap up

In conclusion, conversational commerce platforms represent a transformative force in the E-commerce landscape. 

From instant AI-backed support to personalized sales assistance, these platforms redefine the dynamics of business-consumer relationships. In order to select the platform best for your business, start by identifying your goals and what you want to achieve.

As customer expectations continue to shift towards seamless and personalized interactions, leveraging these platforms becomes not just an option but a valuable addition to any E-commerce strategy.

We’ve recently covered the basics and the main benefits of vector search, an emerging technology that’s redefining how we search and shop online.

Check our mini video series on the topic too! Now let’s see the technical aspects of vector search.

Mathematical representation of information

The key difference between vector and keyword search lies in the way data is represented and compared.

Traditional search methods treat data as a collection of words, focusing on the presence or absence of specific terms. Vector search, on the other hand, uses vector embeddings to represent data points.

Vectors are essentially mathematical representations of objects, whether they be words, products, or images.

Each product in an E-commerce catalog is translated into a vector in a high-dimensional space. These dimensions can represent various product attributes. For example, movies might be represented as vectors where each dimension represents a movie feature (e.g., genre, director, actors, etc.).

By operating in high-dimensional spaces, the search algorithm can capture intricate relationships and similarities between products. For example, products with similar attributes will have vectors that are closer in this multi-dimensional space.

Vectorization methods

To create vector representations, you can use various techniques, such as Word2Vec, Doc2Vec or deep learning models like convolutional neural networks (CNNs) or recurrent neural networks (RNNs). The choice of machine learning method depends on the data and the problem you are trying to solve.

In the E-commerce setup, methods like word embeddings (for textual product descriptions), product embeddings (based on their attributes), session data embeddings (that adds up to a user profile), and hybrid models that combine multiple vectorization methods might be the most beneficial, providing the best search accuracy.

Indexing

Once you have vector representations for all your items, you need to index them for efficient retrieval.

This is often done using data structures like search trees or more advanced techniques like locality-sensitive hashing (LSH) or approximate nearest neighbor (ANN) indexes. These data structures enable fast nearest-neighbor searches.

Search Query Vector

When a user or system submits a query, it is also vectorized in the same manner as the items. The query vector represents the features or characteristics of the query, which then will be used as a base of similarity to find the relevant products in stock.

Similarity Metrics

To find similar items in vector databases, you need a way to measure the similarity between the search query vector and the vectors of items in the dataset. Common similarity metrics include cosine similarity, Euclidean distance, L2 distance and Jaccard similarity, among others.

Closeness in space means more vector similarity, whereas more distance means fewer common characteristics. Businesses using a vector search engine could perform nearest-neighbor searches to determine distance metrics to the closest query-related vectors in a space. 

Utilizing the joint power of keyword and vector search

When vector search and keyword search results are blended, it’s called a hybrid search solution.

A hybrid search engine that combines keyword-based and vector-based approaches offers a more robust and versatile solution, providing highly relevant search results across a wide range of scenarios.

But how exactly are keyword and vector search results blended?

There are multiple different ways, but here’s how we do it at Prefixbox:

In a vector-based search result, you can still have traditional keyword-based search scores, like

• the matching score, which shows how relevant the result is for a specific search query, and
• the popularity score, which shows how popular a specific search result is.

In addition to these scores, the vector-based search results have similarity scores, which are based on the mathematical representation of the keyword, and its distance from the result in the vector space.

Blending is the method of merging the result lists. This can be achieved by scaling and normalizing these scores by applying weights in order to get the best ranking. 

Summary

Vector search is a technique used to find similar items in a large dataset based on their vector representations. Applications span from content recommendation to image retrieval and E-commerce product recommendations.

It’s the critical element that empowers recommendation systems to deliver more insightful suggestions and fine-tunes search engines for superior performance.