I recently visited the website of a large industrial distributor. On a product page for a work glove, “related products” included thread locker, a replacement filter and duct tape.
There seemed to be no rhyme or reason for these recommendations, which were automated based on a product recommendation engine.
On another distributor’s website, related items for the same product are variations of the same glove, as well as a second complementary section featuring products gloves are typically bought with (safety glasses, particulate respirators and cleaning products).
These recommendations help customers find exactly the glove they need, and complements remind them of the other items they should be buying. The result: The distributor sells more, and the customer has a better experience. It’s a win-win.
The problem is that many distributors (such as the first above) are using product recommendation engines on their ecommerce websites originally designed for B2C. This isn’t sufficient for a distributor’s business model and won’t result in as substantial an ROI.
The logic that B2C engines apply to select and display product recommendations is based on online activity, or what people are browsing for. But to make accurate and useful recommendations in B2B, you need to pull from transaction data online and offline, attribute data, and other sources to train it to feed customers what they actually need and to account for the nuances of every end-user’s business.
Distributors sell through many different channels (ecommerce, customer service, field sales reps, EDI, branches and more) with far more SKUs than most B2C businesses; distributors need a product recommendation engine that analyzes all of that data for profitable upsells, cross sells and add-ons if they want to provide relevant and useful information to their customers online.
B2B customers’ purchasing behaviors are also different. B2B buyers need different recommendation types. While B2C buyers are typically purchasing for one-off situations, B2B buyers are more likely to make regular reorders or to buy items for specific projects. The timeframe for a B2B buyer is also usually more urgent. If you’re out of stock, your website needs to offer relevant suggestions for complementary or substitute products that are in stock.
Creating a cohesive and consultative cross-channel experience is a value-added strategy distributors can use to differentiate. AI-generated product recommendations bridge channel gaps and create a shopping experience that helps customers find what they need, presented as “recommended items,” “complete the cart” and “customers also bought,” to name a few. They pair the right content with the right audiences, including more targeted recommendations based on history such as “due to reorder” and “recently discussed with your rep.”
The Benefits of the Right Tool
All product recommendation engines are not created equal.
With a designed-for-B2B AI-powered solution, we’ve seen distributors yield triple- and even quadruple-digit ROI with profitable upsells, cross-sells and add-on purchases.
Before adopting AI, a lawn and garden parts distributor had a conventional ecommerce site. Their site worked as a passive ordering tool for customers to search and place orders. They ran a simple A/B test where the A site included AI recommendations that forecast which items a customer was most likely to buy, when customers will be due to reorder products, and what products are sold frequently together. The B site offered no such recommendations. The A site resulted in larger and more frequent orders that added up to a 21% difference in revenue per customer.
In another A/B test, a flooring distributor increased average revenue per customer online by 41% and average order value by 36% with a B2B AI-powered recommendation engine compared with no recommendations.
An effective product recommendation engine also boosts customer loyalty. An Accenture survey found that 91% of consumers are more likely to shop with brands who recognize, remember and provide relevant offers and recommendations.
Amazon’s success showcases the impact effective product recommendations can have in engaging customers and increasing revenues:
- An industry-low bounce rate (percent of visitors that leave the site after just one page) at 35%. Comparable competitors like Walmart and Target are at 50% and 45%, respectively.
- Average of 9 clicks per visit, 4 more than Walmart and Target.
- Amazon’s AI recommendation engine fuels 35% of customer purchases, or an estimated $50 billion in incremental sales.
Amazon’s recommendations don’t just directly increase sales, they deepen customers’ loyalty. While Amazon’s success has largely been in B2C markets, the implications of having great product recommendations also apply to distributors.
The work we’ve done with distributors has shown they can have similar results – stickier customer relationships and increased and more profitable orders – if they invest in the right engine designed for B2B.