Online retail is rising like a rocket. However, average e-commerce conversion rates still hover around 1-2% and with more competition than ever — from mom and pop drop shippers to large, well-funded retailers — there’s a good chance site visitors will abandon your store for a competitor.
You have 15 seconds on average to keep a customer engaged before they bounce off your ecommerce site. About 30% of visitors will use your search bar to find what they’re looking for, the rest will browse your website.
Visitors cannot buy products they can’t find. You may have exactly what a buyer is looking for, but perhaps they misspelled it or don’t know what category to look under. Or they simply do not know the exact name or description of what they’re looking for.
In this guide, we’re going to focus primarily on two things for improving on-site e-commerce conversion rates: search and discovery. These are the critical seconds after someone lands on your site.
By search, we’re referring to the search box on your website. When a user types in a query for “queen matress fitted bed sheet sale” (50% of queries have three or more keywords and somewhere between 10-25% of searches are misspelled) will they get good results?
Discovery refers to how users can browse sites by category or collection. Should you display the best sellers, top-rated, or on-sale items first? What if a best-selling item is out of stock? Where does personalization come in? Intelligent categorization and ordering can make a big difference in whether a visitor converts or bounces.
Search and discovery are two sides of the same coin. The same technology that powers intelligent search can also power intelligent categorization.
Let’s dig in to see how.
Conversion rate is simply the percentage of visitors to your website that buy something. If you have 100 visitors to your site and one person buys something, your conversion rate is 1%.
According to BigCommerce, average conversion rates across industries are 1 to 2%, and 2% should be the baseline goal.
While this article will focus on on-site, middle-of-the-funnel conversion optimization for after a visitor lands on your site, keep in mind that there are off-site factors as well such as advertising, seasonality, competitor campaigns, broad consumer trends, referral marketing, etc., that can inflate visits to your site without increasing transaction rate.
Google Analytics via on customer journey mapping
A visitor arrives on your site. Now what happens?
Onsite search and discovery begins with understanding the user journey to help you identify opportunities for improving the search and product discovery experience. Here are a few ways you can get insights into your customers’ journeys:
Whether you’re using Google Analytics or another analytics tool, this is a good place to start. Open the Behavior dropdown in GA to analyze your site’s stats for the following:
Heat mapping and visitor replay products like Hotjar or FullStory can show you metrics for exactly how people navigate through your site, when they abandon the site, how they interact with product pages, how far down each page they scroll, and what they’re clicking on.
Are people going straight to the search bar or did they click on the first product image they saw? A single journey won’t reveal much, but over dozens or hundreds of recordings you will begin to see trends and uncover areas for improvement.
Image via JourneyMapOperations
A comprehensive approach to understand visitor behavior and decision making is to build a user journey map. Journey maps start with the user — not your website. Define a persona (e.g., 45 year old mother of two) and track the user from the moment they decide to find a product all the way through purchase.
One of the big advantages to doing a journey mapping exercise is that your entire team gets involved, builds insights together, and owns the journey as a group. As a team, you will find deficiencies and opportunities, then prioritize and plan goals and key deliverables to shore up any shortcomings.
There’s enough content about journey mapping to fill up multiple guides. If you’re interested to learn more, check out both Shopify and Qualtrics for their write-ups on e-commerce journey mapping.
Mystery shoppers are well known for improving brick and mortar store operations, but they can also be hired for online commerce evaluation. You could hire online mystery shoppers to carry out a task such as “visit the site to purchase the best garden supplies for under $25” and gather feedback on their experience.
There is an entire industry for hiring mystery shoppers. The advantage to working with a specialized mystery shopping firm is their deep knowledge and expertise in crafting shopping experiences and gathering actionable intelligence.
However, if you already know what to ask and how to frame the questions, you could even go to usertesting.com to hire people directly to hear what shoppers are doing as they navigate from page to page.
Whether you choose to use analytics, heatmaps and visitor replays, journey maps, or mystery shoppers, or all of the above, you will have some good ideas of what to focus on to improve the product discovery experience.
The good news is that many of the issues you uncovered during your site analysis and journey mapping can be addressed using modern on-site e-commerce search and discovery platforms. Here’s a look at some of the ways you can improve visitor experience by activating features such as machine learning, dynamic facets, instant indexing, and more to deliver better and smarter results that convert more consistently.
“Face masks” meant something completely different in 2019 versus 2020. Prior to COVID-19, a buyer who was searching for “face mask” was probably looking for cosmetic makeup products. That all changed with the pandemic.
It’s impractical to manually adjust your catalog for every possible permutation. Merchants need automation to adjust results in real-time to changes in consumer behavior. The key here is machine learning.
Machine learning improves your e-commerce store’s ability to adjust search results automatically based on what users are searching for and buying.
A smart search engine should know when a user converts (registration, add-to cart, purchase, or something else) to boost similar results for future queries and page views.
Machine learning is not a nice-to-have anymore. It’s a must-have, especially for online stores with many products. Search engines can use reinforcement learning to score search results and user behavior to automatically improve results over time.
(If you’re looking for solutions with machine learning, one thing to note is that the terms AI or artificial intelligence can be used interchangeably with machine learning.)
Research from Nielsen and Norman has confirmed the necessity of faceted search; when users are presented with too many choices, they will overlook the products they’re looking for and will go elsewhere. Facets have become an integral part of shopping online to filter through results more quickly.
Allowing your customers to filter results by price, color, ratings, or other factors is vital — especially for sites that have hundreds or thousands of products. Your e-commerce search and discovery platform should support both search facets and filters in search results and on collection pages to help customers narrow down results to find exactly what they need.
What’s the difference between search facets and filters? Briefly:
Creating rich filters and facets can seem overwhelming, but fortunately, modern search solutions will make the task of specifying and implementing them a matter of minutes. Depending on what tools you’re using, filters and facets can be generated automatically based on your product attributes.
It’s also worth noting that filters and facets are not just useful on search results pages — they can also be used for category and collections pages to quickly narrow down results.
Engagement metrics such as click-through rate (CTR) provide some value in fine-tuning results, but conversion metrics like signups, registrations, or sales ultimately offer more value to the business.
When you tie search results to conversions and add machine learning, you can build a positive feedback loop that feeds into your success.
If a visitor searches for an item, clicks on a product result, and purchases it, your search solution should make the “aha” connection to automatically prioritize similar search results.
It can work on category pages, too; if customers continually purchase the third item from the top, the collection could re-prioritize items on the page.
When you add or edit a product to your online store directly in the CMS or via a product inventory management system (PIM), update pricing, change inventory, etc., it should be reflected in your search index immediately.
Unfortunately, for many search engines, it can take 24 hours or longer for these kinds of everyday changes to be updated. That’s way too long and could hurt conversions.
Many solutions advertise instant indexing but typically that means only the initial index is instant (depending on your catalog size, this could take seconds or minutes). Find a solution that offers instant product information indexing. It should also offer instant update functionality for product catalog additions, edits, updates, removal, etc., so you never lose an opportunity.
For example, in the case of Sajari, when you build a schema from your PIM or website, Sajari will instantly index the content, then re-index content immediately whenever a change is detected.
You want to push the best selling, most popular items to the top of your category or collections pages. Typically, categories and collections can be sorted by price, ratings, popularity, and other filters.
That’s fantastic, but even better would be rules that also intelligently boost products based on business criteria.
Brick and mortar retailers can set up displays to encourage potential customers to add more items to their shopping carts. These merchants have a good chance to make a sale or increase the order value because buyers can see a large selection of products.
It’s more complex for an e-commerce business. If a buyer searches for back-to-school backpacks, how can you also let them know about pencils, binders, or even kids shoes?
Merchandising is hard for most online businesses. Traditionally it required either a lot of manual work or a team of engineers using data science to discover the best times to promote related products or display “you might also be interested in” pop up.
The good news is that you can improve conversion rates using modern on-site search solutions without investing in a full-time search engineering team. Using data, newer e-commerce platforms can intelligently insert related items into search results to provide a merchandising experience. Products could be promoted this way based on:
Don’t let your customers abandon your site due to unavailability of stock or sell them an item that you can’t fulfill. Recommend options to keep them engaged.
Low inventory or out-of-stock items can be automatically demoted or hidden from search results or collections pages.
Conversely, high inventory or re-stocked items can be boosted in search results or collections pages. Both scenarios require a search and discovery system that can be connected to your inventory management system. You should be able to set rules to include/exclude or boost/hide items based on inventory levels.
If you know a customer has purchased Nike shoes in the past, on their next visit, you can recommend additional Nike products. The use of personalization can tremendously improve user experience and conversion rates.
BigCommerce found that personalization can lower bounce rates by as much as 20-30%. (Typical retail site bounce rates are around 30-55%)
Built-in personalization features and/or connecting to third-party personalization solutions should be on your list as must-have technology. Personalization creates contextual profiles for individual visitors so you can deliver custom, relevant search results and display content. Personalize results based on user preferences, location, gender, past purchase history, product category, and more.
Studies show that one-third of all searches contain 4+ or more keywords. In the context of site search, NLP is the process of analyzing queries to infer structure and meaning. Structure, in this case, is referring to information that is highly defined, for example, a category or a number. It can also represent relationships between things. Common examples include sizes, colors, places, names, times, entities, and intent, but there are many more.
If a user types in “low profile mattress bed sheets” or “bed sheets for queen mattress,” your search engine needs to parse this information to return high quality results. The use of NLP for e-commerce websites can impact your brand perception, user experience, and conversion rates.
An invisible and often overlooked UI element to e-commerce search is how to handle typos and misspellings. Somewhere between 20-30% of search terms can contain a misspelling!
There are a few different ways to manage typos and misspellings. A good site search and discovery tool will give you options, including a way to add “did you mean?” results in search result pages and/or auto-complete with typo tolerance.
Using typo tolerance of some flavor is a best practice. Don’t lose customers who think you don’t carry “bakcpacks” or “addidas.”
Amazon showed that just .001 second differences in returning results meant big losses. Your search tool or search provider load time should be clocking-in in milliseconds.
What if, for a given query, you changed the search results? Would you have better or worse click-throughs, conversions, sales, or user satisfaction? A/B testing can help answer that question with data.
Tests could be performed on everything from product terms, to how your data has been indexed, to the search results design. A modern search solution should allow for search A/B tests and provide guidance on what search algorithm helps your company improve its bottom-line results based on whatever criteria you’ve established.
We’ve discussed search and discovery, but let’s take a look at a few other on-site opportunities for improving the shopper experience. Here are some additional tips for site usability and user experience.
Reducing friction in the checkout process can improve conversion rates. This includes UX changes such as hiding additional fields on the checkout page unless customers need them — for example, if shipping and billing addresses are the same, there’s no need to display both. Check out (no pun intended) other tips from Rejoiner on improving checkout customer experience.
With an average of nearly 70% shopping cart abandonment rate, it’s no wonder merchants should spend time optimizing their shopping cart UX. About 58% of the time, online shoppers are merely browsing, but that other 42% should be optimized for success. For example, the same study showed that 20% of shoppers abandon their carts because of design-related issues. Improvements here, such as providing a way for customers to view the total cost of an order or displaying credit card payment options upfront, can positively increase sales and conversion rates and avoid the dreaded abandoned cart.
Offering the lowest prices, fastest shipping, or best customer service can certainly improve conversion rates, but your competitors can do that too. To stand out, consider your product descriptions. Moz explains how companies doubled conversion rates through better copywriting. Better product copy can increase the emotional appeal and sense of urgency of your products and help your company stand out from the pack.
Desktop users convert at a higher rate, but mobile users add items to their shopping cart at about the same rate. Mobile traffic to online retail sites is as much as 50% these days. All the search and discovery improvements we recommend above — setting rules, site speed, instant indexing, etc. — will work on mobile devices as well as desktops, but retailers need to pay special attention to building great UX for mobile devices.
People want to buy products that others love: 92% of potential buyers read customer reviews and testimonials when considering a purchase. Making customer reviews more visible, adding social share buttons with a counter, or even showing real-time purchases can all help increase conversion rate.
eBay Research published a paper showing that a “Visual image is a powerful channel to convey crucial information towards online shoppers and influence their choice.” In addition to including more images (and/or product videos) on your product pages, the researchers also found using larger photos allow for potential customers to more easily inspect the details of a product. That in turn increases the likelihood that someone will purchase the product. (source)
Your call-to-action (CTA) on your product and landing pages matters. Certain phrases — Buy Now, Join Free, Add to Cart — have been proven to be more effective than other phrases. Most retailers are using these phrases already. But there are other factors, too, that can improve your CTA including mobile optimization, colors, location on page, and more. Check out GrowCode’s full article for more ideas.
After helping hundreds of customers with search installations, one of the clearest insights we've learned is that most site search installations don’t work in an optimal way primarily because of insufficient meta information or poorly defined website structure. The work you do to improve on-site search is also the work you need to do for good search engine optimization (SEO) to drive visitors to your site. Optimized site structure, metatags, headings, canonicals, etc., help both on-site search and web search engines. Entire books have been written about SEO for driving new customers to your site, but for a good primer, check out Ahref’s guide to SEO for e-commerce.
Most e-commerce platforms such as Shopify, Magento, WooCommerce, and BigCommerce have very good design, good tooling for managing payments, inventory, and returns, and integration with a plethora of other systems.
But helping visitors find what they want is still incredibly tough for a variety of reasons, especially for sites with many products, categories, and collections. The good news is that new search and discovery engines for e-commerce can help. In particular, advances in machine learning make it much easier to intelligently deliver product results and automatically order and boost products to increase conversion rates.
If you’re in the market for a new solution to drive conversion rate up by making your products more visible, try Sajari for e-commerce search and discovery. Create a free account today or contact us for a personalized demo.