With retailers worldwide prepping for the end-of-year sales rush, now’s the time to start fueling your personalization game plan to end this year strong. Here’s a summary of improvements we’ve made to our Onsite Product Recommendations between June and September of 2018.
1. Onsite Product Recommendations Settings Page
We’ve added a dedicated Settings page which allows merchants to change the global behavior of Onsite Product Recommendations to better match their business. For example, if your store is fashion-focused, you might not want to recommend items that a consumer has bought before; but with businesses where repeated purchases are valuable and more common, like beauty and cosmetics, this is almost a must-have.
2. Improved Fallback Product Recommendations
Fallback product recommendations are a simple, yet powerful tool to produce unique experiences for new, returning and engaged customers – with the goal of reducing bounce rate and improving conversion rate.
Improved by the recent launch of our Segmentation & Insights and Onsite Content Personalization products, but available in all plans, our fallback product recommendations now allow for the capability to fill the primary recommendation slot – which only generated partial results with fallback products. For example, first two products in a recommendation slot of five products could be tied to a user’s browsing history, with the remaining three slots filled by best sellers following the fallback.
3. Relative Margin Filtering
Our improvements to inclusion and exclusion filtering now allow for the use of relative margin filtering instead of absolute margin filtering – all driving better sales profitability through Onsite Product Recommendations. Previously, you could set parameters such as recommending only products with at least a 20% margin, up to 100%. Relative margin filtering allows you to anchor the margin to the viewed product. For example, if a viewed product has a margin of 50%, you can set the minimum parameter between -20% lower margin, which would then recommend products with at least 40% margin, but not less.
4. Google Optimised Landing Page Recommendations
Landing page recommendations are an effective way to customize landing page elements and reduce bounce rate depending on the traffic source. Landing page recommendations previously took into consideration UTM tags from different campaigns/sources/mediums. In addition to these attributes, you can now also consider Google auto-tagging Gclid parameters, allowing you to customize landing page recommendations automatically for each Google Adwords ad campaign. Check out our revised Recommendation Type Glossary for a full list of algorithms.
5. Dynamic Client-Side Filtering (Category Faceting)
Nosto’s product recommendation engine now supports faceting, meaning that when a website visitor drills down into subcategories to view only a certain color or size option, Nosto’s product recommendations can be adjusted accordingly to match that user’s selection. For example, Nosto can display best sellers or best personal product picks in the selected color/size. Read more about faceting in our Category and Brand Tagging Github wiki.
6. Random Recommendation Type
This improvement is built for developers working on Nosto. Our recommendation types now have support for the display of completely random product recommendations in the desired product slot. This enables teams and developers building the setups to both build the design and quickly do first-line evaluation on problematic cases more easily. You can view a full list of possible recommendation types in our Recommendation Type Glossary.
7. Image Thumbnails for Both Alternative and SKU Images
We’ve added an optional support for the ability to leverage alternative images or SKU images (also as ‘thumbnails’) in recommendation template. If this is something needed on your Nosto account, please contact our support team by using our onsite messenger app or via email@example.com.
There’s plenty more to come, so stay tuned to this space for the latest improvement updates. Want a deeper dive into any of the improvements we mentioned? Give us a shout at firstname.lastname@example.org or use our onsite messenger app to chat one-on-one with one of our team members.