18 Dec 6 Best Practices for B2B Site Search
Over the last few years, I’ve done a lot of work with eCommerce site search, both via consultancy relationships (directly with merchants) and also with Klevu. In that time, I’ve seen a clear shift in B2B retail, with the focus becoming more in line with B2C retail around customer experience. Search is a key part of the buying process and one that B2B merchants shouldn’t overlook.
In my experience, search-led conversion rates (based on only users who have completed a search) are usually three to six times higher than journeys without a search. This is also usually higher for B2B stores due to the nature of the buying process. There’s an argument that this is based on intent, but I’ve done two projects in the last year where we drove a higher proportion of people to use search and we achieved an incremental uplift in revenue on both occasions.
Here’s how B2B merchants can harness the power of search:
1. Serve Customer-Specific Results
One of the most important B2B features is the ability to assign pricing and product visibility to specific customers and groups of customers. On the pricing side of things, this means that the price displayed to customers will differ from all products or just specific products, which will inherently need to be adapted for search. The same principle applies to customer-specific catalogs and you can show products that are set to be available to a specific customer in search.
2. Support Tiered Pricing
Another feature often used by B2B merchants is tiered pricing, such as providing a discount based on the quantity of the item being purchased. You can display the price based on the lowest possible price of the item, as per the below screenshot from tattoo supply merchant Painful Pleasures.
3. Provide a Faster, Richer Auto-Suggest Experience
Being able to provide a fast and seamless auto-suggestion experience is another key requirement for lots of merchants, which can improve the overall customer experience and increase the speed of selecting products and ultimately completing a purchase.
In my experience, allowing the customer to filter in this initial overlay adds a lot of value, as most B2B stores have very large catalogs. Another feature that can add value here and is often requested is a ‘quick add to cart’ button.
4. Use Advanced Product Boosting & Merchandising Capabilities
With B2B stores generally having very large product catalogs, being able to create rules to promote specific products and groups of products in certain ways is hugely important. Your search tool should be able to:
Boost the visibility of a specific SKU globally
Boost the visibility of a specific SKU
Boost a specific selection of products globally (based on product attributes)
Boost specific items for specific queries
Assign hero SKUs for specific queries and groups of queries
Depending on how complex your customer base and catalog are, you may also need the ability to apply different boosting logic for different users.
A lot of these requirements would also be the same for filtering. For example, the ability to assign the order of filtering for specific queries or users, as well as allowing these to be machine-learned based on how filters support the effectiveness of queries and overall search performance. At a minimum, filtering would need to be adapted based on the selection of products being returned.
Another feature that I think can be very valuable for both B2B merchants is the ability to serve banners and ad blocks as part of the search experience so both on the results page and in the overlay. Ideally, this would be manageable at a query and customer level and would support scheduling.
Users are also likely to search by SKU on B2B stores, so this is something that needs to be supported, as per the below screenshot.
5. Get Reporting
Reporting is a big part of search for all retailers, as it can provide a lot of insight into what users are actually looking for, as well as what’s performing. In order to maintain a strong search solution, it’s important that you’re able to monitor what’s working and not working, here are some of the baseline reports you should be using:
Search term reporting – Keywords driving clicks and revenue (ideally by device and location)
Use of filtering – Reporting on which filters are being used and the values being selected
Refinements – Ability to drill down by geographic location, device etc
Zero-result search terms – Showcasing errors that need to be handled via synonyms, custom messaging or redirects
Device – Reporting on search performance across device types
Your web analytics platform will drive more granular data, including things like where searches are being performed on the site, behavior on search results pages, further segmented performance, and usage.
6. Take Advantage of Machine Learning & NLP
As I’ve briefly touched on earlier in the article, machine learning and natural language processing are capable of adding a lot of value to B2B merchants, by automatically optimizing results based on performance and user behavior (machine learning) and being able to extract more context from the query (natural language processing).
The core benefit of machine learning in search is that the results are updated either frequently or in real-time based on how users are interacting with the products. For example, boosting the products that are converting better or getting the most clicks. The benefits of natural language processing are being able to serve far stronger results by understanding more about the query and what the user is looking for and not relying solely on the relationship between the product and the terms being used.
Both are must-haves in modern-day search and they’re likely to reduce the manual overhead for your eCommerce and merchandising teams considerably. In the same vein, catalog enrichment is another capability that can help drive more accurate results, as well as process more complex search queries.
While talking about features supporting better long-tail results, another key requirement would be indexing more data and information, with examples being attributes such as size and price, and things like product reviews and Q&A content.
A few other features
Beyond the core requirements, here are a few other areas that may be of interest depending on your business and team:
Content search – Serve CMS pages and integrate other content feeds into the results
Personalized search – Personalize results at a 1:1 level based on the behavior of a specific user
Custom product recommendations for error queries – Serve custom product recommendations on 0-result error queries
Category merchandising – Enable category merchandising via the same system and business logic. For more advice on merchandising, I wrote this piece which provides more detail on merchandising in Magento.
About Paul Rogers
Paul Rogers is an experienced eCommerce Consultant at Vervaunt and works primarily with the Magento Commerce platform. Paul has previously worked with a host of well-known retailers, including Dr. Martens, O’Neills, Agent Provocateur, Nestle and The V&A, as well as working for Magento agencies such as GPMD and Inviqa.