Algolia sponsored the 2023 Ecommerce Site Search Trends report which was produced and written by Coleman Parkes Research. The report, which was based on a survey of 900 IT and business decision makers, speaks to how companies plan to invest in search-related technologies this year, including ecommerce personalization.

Respondents overwhelmingly recognized the importance of search for generating revenue. They also identified how it will impact personalization across many channels including text and email communications, social commerce, and shopping cart programs. 

Personalization is a key component of the customer experience but there are many more features that could be introduced to maximize the potential return. For example, nearly three quarters (73%) said they have implemented a ‘related products’ feature — this being the most common type of personalization. Beyond this, less than half have implemented any other types of personalization.

Many respondents told Coleman Parkes research that they’ve implemented some personalization capabilities, like offering their customers a way to design a personalized shopping profile to start to receive personalized offers, and 43% said they have product recommendations based on items searched for or purchased. 

Many respondents said they’re still just getting started. For example, 56% said they’re planning to implement a ‘saved items for later’ program and 42% said they’ll offer a text subscription program soon. 

This is only the tip of the iceberg for offering personalized experiences. The challenge for most ecommerce companies is leveraging the 1st party data they’re sitting on to build personalized programs. 

Personalization is powered by data

This isn’t an entirely new war cry — Gartner published its first personalization magic quadrant back in 2018, Amazon published a 2003 paper on recommendations, and Netflix lit a fire in 2009 with its $1 million bounty to write a better personalized recommendation algorithm. 

Personalization matters to consumers, too: 91% say they are “more likely to shop with brands that recognize, remember, and provide them with relevant offers and recommendations.” For sellers, personalization can drive up to a massive 25% revenue lift — not to mention higher customer and brand loyalty. 

What does search technology have to do with personalization? Plenty, actually.

• On-site search results can be personalized
• Same for recommendations, which can now be powered by your search engine provider
• Search-related data is the foundation for delivering personalized results
• Machine learning for site search is driving many innovations
• Search data can feed a user profile with highly relevant intent signals

Data-driven personalization

For any business that intends to invest in personalization, it starts with data.   

The more demographic and psychographic data you can collect about your customers and visitors, the more sophisticated your personalization can be. This includes data such as:

• Location (geo)
• Age
• Gender
• Past purchase history
• Site search history
• Pages viewed
• Social media likes
• Member status (if you have a membership or rewards program)

It can also include contextual data such as browser and viewing preferences, off-site information such as emails clicked, rewards program credits, ads clicked, and more.

Mid-to-large sized ecommerce companies have begun to invest in building data lakes to store, analyze, and leverage customer data across omnichannel marketing touch points for real-time personalization. Even if you are not investing in a data lake, many ecommerce platforms store enough data on their own which can be used for personalization. 

And there are other systems that have a lot of customer data already — your search solution for example! Your search engine is generating and storing heaps of valuable data which can be used for personalization and other programs. It includes things like recently viewed categories and products, recent searches, purchases, and more. 

You can leverage your data for to implement personalization in multiple ways to personalize:

• Search
• Recommendations
• Browsing
• Related products
• Dynamic re-ranking

The same personalization rules you set up for search can be implemented everywhere. For example, in Algolia you can use the product dashboard to assign value, or weight, for each event or attribute, which in turn boosts personalized results in the search rankings and on dynamic catalog and product listing pages (PLP). 

For example, shoppers on EyeBuyDirect’s website enjoy a consistent, personalized experience. 

When users browse to the “women’s glasses” category page, the results are boosted based on each user’s unique preferences, such as frame style/shape or materials. Preferences can be determined from past purchase or browsing history. It can optimize for any attribute, which will vary from use case to use case. Personalized results can be displayed as customers search, browse the site, or view a category. Set the rules once, deploy them everywhere.

You can also add personalization to recommendations. Recommendations could be one to many, one-to-one, or a mix of the two. For example, it could include “customer who bought x also bought y” or “because you bought x and y, we think you’ll also like z.” 

In the case of EyeBuyDirect, early results showed:

• 4% more revenue with search, increased browsing, and personalization
• 1 – 1.5% improvement in conversion rate with product recommendations

Another example: By using Algolia’s Recommend, a recommendation engine, customers were able to drive a $1.2 million uplift in revenue. 

New machine learning algorithms built into solutions like Algolia will also dynamically re-rank search query results and apply personalization rules. This happens in single digit milliseconds. As a user types a query, the search engine first looks for relevance and then it will apply re-ranking to push better results to the top — where better refers to a combination of personalization rules, merchandising, ranking rules, or other factors. 

Conclusion

From recommendations to personalized search results, there’s a huge opportunity for innovation of user experience to drive revenue and customer loyalty. Advances in machine learning have filled the sails — personalization solutions are accelerating and more powerful than ever. 

Behind these personalization algorithms are different machine learning techniques that can leverage data for multi-visit sessions or even single first-time user sessions, identify predicted intent, and deliver personalized results that are clearly measurable. If you want to learn more you can start by downloading your free copy of the 2023 Ecommerce Search Trends.