9 Lessons from 250,917,080 Ecommerce Site Searches

9 Lessons from 250,917,080 Ecommerce Site Searches
This is a special guest post from Miles Tinsley, Head of Sales at Klevu.

Ecommerce site search is perhaps the most overlooked (pun intended) element of online sales.

In an examination of 21 different stores, when visitors searched for a product, they were 1.8 times more likely to convert. In addition, visitors using search can generate as much as 13.8 percent of a site’s overall revenue.

Despite its importance, most ecommerce sites don’t support onsite search as they should. As Baymard highlights, “18% handle phonetic misspellings so poorly that users will have to pass a spelling test to be presented with results (e.g. 0 results for “Kitchen Aid Artysan” when looking for the “KitchenAid Artisan” mixer).”

Drawing from over 250,917,080 searches Klevu has powered, I wanted to share nine lessons to make your own onsite search more profitable:
  1. Anticipate Common Queries, Not Accurate Descriptions
  2. Consider the Order of Products and Attributes
  3. Personalize Results with Aggregate and Individual Data
  4. Accelerate via Predictive Autocomplete and Error Tolerance
  5. Make Mobile Search Simpler for Higher Conversions
  6. Understand Analytics by Testing Revenue Breakdowns
  7. Teach Visitors to Search and Offer Query Alternatives
  8. Index Not Just Your Products but Your Content
  9. Prepare Your Website for B2B Searches
box

Keep reading learn more

But, if you’d like to go behind the scenes and find out how onsite search is just one part of what helps the average Shopify Plus merchant grow between 126% and 274% YoY …

Then download the full recordings and slide decks from our two-part webinar event, Growing Your Entire Online Funnel.


Access the entire event here

1. Anticipate Common Queries, Not Accurate Descriptions

Despite all the effort you’ve put into naming and describing your products, the vast majority of visitors don’t know the product title nor the exact attributes of what they’re looking for. On a site like Chubbies Shorts, instead of typing in “THE KAUAI NOTS,” a far more likely query could be “crazy patterns.”

Other common search queries include:

  • Sizes (e.g. “medium black peacoat”)
  • Best or top-rated (e.g. “best LED 4k tv”)
  • Features (e.g. “waterproof iPhone case”)
  • Compatible (“wireless keyboard for iPad Mini”)

Natural language processing (NLP) shows accurate results to the visitor even when they don’t know how to describe what they’re looking for. In addition to product names or descriptions, NLP-powered search takes context and relevance into consideration such as pricing propositions or color variants.

Remember, visitors can’t become buyers or customers if they can’t find what they’re looking for.

2. Consider the Order of Products, Attributes, and Content

Each product resonates with visitors differently. Unfortunately, with many ecommerce technology platforms, there’s no way to change the ordering of the products in search results.

At your store, you might want to show off products most important to your business (be it in size or profit margins, known as “Hero SKUs”). Or, these products could be strategic to retail customers and meaningful to shoppers (e.g., setting a price anchor or creating a branding perception).

Showing these results up front, even (and especially) when they aren’t specifically asked for by the visitor, enables retailers to continue to run product-specific campaigns on the search level.

3. Personalize Results with Aggregate and Individual Data

Many third-party providers are starting to explore personalization as a way of accelerating the buying process. In on-site search, that means making the most of the search bar, even when the visitor hasn’t necessarily started typing yet.

Klevu does this by prompting the visitor with words from popular searches across all consumers, the top products, and then more importantly with recently viewed items from the individual.

This is what we call “zero character search,” the ability to create conversion possibilities without the user entering a word. As another example, Chubbies Shorts shares their popular “trending” items as a default search option as well.

4. Accelerate Searches with Predictive Autocomplete and Error Tolerance

A decade ago, Google started offering predictions in their search. In the following years, this experience trained people to expect support as they search anywhere else, including your ecommerce store.

Predictive autocomplete and error tolerance create a smoother experience for visitors searching your site.

No longer do visitors need to fully type a query. Predictive autocomplete removes friction by drawing them into your site and toward the product they want quicker.

Error tolerance also removes confusion from a visitor’s search query. Your on-site search can now understand a spelling mistake and still bring the item up, showing your product to a potential buyer faster and with less confusion.

Both these functions enable visitors to interact with search faster and more smoothly, leading to less drop-off and a higher conversion rate.

5. Make Mobile Search Simpler for Higher Conversions

In 2017, 64% of sales on Shopify during Cyber Monday Black Friday took place on a mobile device. Mobile is rapidly growing to be just as popular as desktop for purchases, which means you had to develop and design for this interface yesterday.

The Baymard’s Mobile Usability Study determined that on-site search, “was generally the preferred product finding strategy of the test subjects during the mobile e-commerce usability study, as they perceived it to be faster than category navigation. Unfortunately, search is almost entirely broken on most mobile e-commerce sites in practice.”

Ecommerce site search on mobile

6. Understand Analytics by Testing Revenue Breakdowns

When you’re looking for a third-party search solution, you want to see what’s under the hood to really understand the value they’re providing (as you would with any core premium solution).

As such, when you dive into analytics to check out the impact of your search solution, I suggest breaking down revenue by looking at the most regularly searched keywords, the individual clicks to conversion, the geographical locations all the way down to the individual products that were purchased from the individual query itself.

Understanding the data behind your search is the best way to know how it makes money, how visitors use it, and how you can make it better.

7. Teach Visitors to Search and Offer Query Alternatives

Simple and smooth processes mean your visitors will buy more easily, and your business will see a quicker conversion. There are a couple ways to do this with integrations …

Teaching your users to create a more detailed search query in order to find what they’re looking for will bring better results for them and your business. Prepopulating your search bar with popular, suggested, or campaign-related terms and visuals make this straightforward:

Another opportunity presents itself even when you don’t have the product that the user had searched for. If your search recommends similar products based on the users query, you can create more interest and hence greater conversion opportunities.

8. Index Not Just Your Product Catalog, but Your Content as Well

So far, most of our points have been focused on the product catalog. Now, let’s talk for a second about website content …

In order for your search to integrate with your brand experience, you’ve got to index relevant content at your retail site in addition to your product catalog. This can mean blog posts, your about pages, shipping or contact information, and more. Making these pages searchable helps to unify your brand to your customer creating a more user-friendly experience.

Sometimes, it must just be a visitor looking for terms and conditions, shipping, policies, or contact information.

But if someone is looking for content, or to find out more about the company, they really want to get learn more about your brand. That’s when a visitor turns into a loyal customer that keeps coming back for more.

Indexing also improves site performance, which means a smoother process, which again makes for higher conversions.

9. Prepare Your Website for For B2B Searches

You’ve seen all the mindblowing stats on B2B ecommerce. As such, you’d be crazy to not consider them in designing your on-site search. There are a couple things to keep in mind when optimizing search for B2B customers…

First, group products and pricing should be readily available for when businesses are buying in larger quantities.

Also, make SKU numbers searchable. B2B visitors likely know exactly what they are looking for. It’s also important for your on-site search to be able to deconstruct the codes so that even if your business doesn’t have the product, the B2B visitor is looking for, you’ll still bring up similar products to recommend.

Looking Ahead

We’re just starting to explore the future of search. As we continue to move forward, I hope to see the tool grow to better serve retailers and users. Here are some of the ways search will improve:

  • Deeper personalization: Creating a one-to-one level approach and considering data such as gender and age when recommending products and content.
  • Voice Search: 40% of millennials are already making search queries through a voice assistant. This feature will be extremely important to shoppers, especially via mobile.
  • Connected experience: Connect search with email campaigns and advertising campaigns in Facebook and Instagram. That experience really shows value for retailers.

There are many ways to make your search the best it can be, and the possibilities are likely to grow. The key to making search work for you is to understand that the faster the customer can find what they are looking for, the more likely they’ll buy from you.


About the Author

Miles Tinsley is the Head of US at Klevu. He is responsible for all sales, operations and partnerships within North America. Miles has a strong background in working with high-growth ecommerce startups.