Last week I was on a mission. The snow was incoming and my wife and I were getting excited to spend our first few days of the ski season in the mountains skiing the fresh snow and enjoying the après. As we started to pack up our gear I realized that I was missing a critical piece of equipment, my ski poles!
Frantically searching the closet, the garage, and finally my wife’s ski bag, I came to the conclusion that I better figure it out quickly. I went to Google and typed in “telescoping ski poles”. I browsed through some of the shopping ads and find backcountry.com whom I’ve used in the past and had quick free shipping.
After landing on the product detail page (PDP) I quickly realized these poles were not telescoping. This is important to me as I like to adjust the poles based on the conditions and terrain we encounter. For the skiing we enjoy, this feature can actually be a safety feature, so it’s quite important. I head up to the search bar on backcountry.com and search for “telescoping ski poles.”
Ugh. You’ve got to be kidding me. So at this point I think, back to Google? Head over to Amazon, evo, or REI.com? I know that backcountry.com carries these. Back up to the search bar I go, simplifying the search to “ski poles”. The search engine gets me now, and retrieves plenty of ski poles. I scroll the page looking on the left for a filter to help me refine the large assortment down to the “adjustable length” poles. After looking at the 18 telescoping ski poles, I selected the right ones and luckily got them in time for the trip!
Turns out this experience is what millions of other shoppers go through. However, the problem is, a significant number of shoppers give up. They go back to Google or jump over to a competitor’s site. This number is quite scary when you put dollars to it. You can quantify this by identifying the number of sessions that exit from a zero results page. Multiply that by your conversion rate and AOV and you have the money you’re leaving on the table. Here’s an example:
So how do we fix this? One approach is to export all the zero results to a spreadsheet and have an analyst work through them one at a time, curating rules to retrieve products and hope it’s all of the products that should be returned for that query. Wrong.
Your Zero Results Report Is a Goldmine
Never Null, the AI-powered zero result solution from Lucidworks, is the answer. The dense vector approach is trained on two encoders which guide the model to relevant products in your catalog for customer searches—like “telescoping ski poles.” Encoder 1 is based on the successful searches that result in add-to-cart. Encoder 2 is based on shoppers like me, the persistent shopper, pushing through the friction to find the products in the catalog that deliver on my goals.
These shoppers are the ones we need to tip our hats to. We’ve come to realize that with the Never Null approach every site is sitting on a goldmine of data. And it’s easy to mine. All we have to do is craft the signal data into the model, iterate on a few approaches to relevancy, and voila, your zero results problem is no more and your shoppers are extending their shopping journey, adding more products to cart.
Instead of being greeted with “We couldn’t find any results for ‘telescoping ski poles’”, Never Null would deliver the relevant 18 results.
Persistent shoppers help solve for multiple zero result problem themes, including:
- In stock / Out of stock
- Vocabulary mismatch
- Products not carried (suggest replacement items)
Lucidworks implemented Never Null at a US Top 20 Retailer ahead of the 2020 Cyber 5 with these YoY results:
- Clickthrough rates from search up nearly 20%
- Search influenced orders up 30%
- Reduced null results by 91%
Shoppers Stay for Great Results
Keep shoppers on your site. Keep the journey going. Let Never Null dig into the goldmine of data your persistent shoppers are providing and get your team focusing on the next big problem, we’ve solved this one. Check out the full case study on Never Null results here.
New poles came in time!