Product

Introducing AI Personalization (đť›˝)
facebooklinkedintwittermail

Algolia is pleased to announce the public beta release of our next-gen personalization platform, Algolia AI Personalization.

Algolia’s AI Personalization will equip merchandisers and developers with the ability to present each online visitor with precisely tailored content, leveraging comprehensive insights from their behavior, preferences, feedback, and attributes — even in cases where no explicit information has been shared. This translates into enhanced performance throughout the user journey, delivering unique, individualized experiences while streamlining both setup and ongoing optimization efforts, saving considerable time and effort.

Personalization can mean different things depending on context.  From adding {first name} to an email subject, to manually segmenting users based on demographics and exposing them to bespoke content or adding a “recently viewed” carousel on a product detail page of an ecommerce site, the sheer range of strategies that fall under the broad topic of personalization is staggering.

While it’s not hard to grasp the idea of personalization conceptually, when it comes to execution we see many businesses still struggle, especially with the advent of the age of AI. And that is because personalization is not a tick-box exercise — you cannot just set it and forget it. Instead, personalization is an unremitting tuning and optimization process, accomplished today through a close collaboration between humans and AI.

For personalization to help meet business goals in a meaningful way at scale, it can no longer be one or the other — humans and AI both need to work seamlessly in tandem.

But there’s a catch: personalization is NOT (or no longer) just a marketing problem. It’s a product problem and it should be treated as such by widening the aperture to include the customer perspective, the business viability, as well as the technology feasibility. It’s a horizontal capability that in our case may start with search but extends into multiple touchpoints: browse, discovery, and beyond.

This requires harmonizing the end-user perspective (new visitors, returning visitors, authenticated users or buyers — what’s relevant to them) with the business perspective (does it generate sustainable return on investment?) and the technical perspective (is it fast, scalable, and highly available?).

Ultimately, similar to search, personalization is a continuous fine-tuning process with the additional complexity that it doesn’t apply only to search or only for users that interact with search. And it requires a hybrid approach: (1) state of the art AI/ML infrastructure and capabilities and (2) data-driven teams capable of going beyond the out-of-the-box personalization functionality into orchestrating sophisticated personalized experiences that maximize customer lifetime value.

And that’s the hybrid approach we had in mind when we built our end-to-end AI Personalization Platform: we offer commerce companies the ability to easily setup search and browse and discovery personalization. We also offer APIs and tools to build more sophisticated journey orchestration and measure performance.

End-to-end Personalization Platform Architecture

The building blocks of AI Personalization consist of:

  • The Algolia infrastructure known for speed, scalability, and ease of use.
  • The data source layer allows you to send events (views, clicks, conversions), send records (product catalog, items, articles, etc.) or even connect to third party platforms such as Segment, BigQuery, or Shopify to ingest all of the above from existing sources.
  • After processing the user behavior data (events and records), AI Personalization is generating user profiles that consist of affinities — attributes of products that each and every individual user has engaged with throughout multiple touchpoints (search, browse, discovery, etc.). This is where AI personalization excels, as it automatically adjusts event and facet weights, optimizing user profiles to boost business outcomes while maintaining consumer satisfaction.
  • These user profiles and their affinities are then applied at query time to re-rank results according to those preferences, enhancing relevance and ultimately improving key performance indicators (conversion rate, average order value, or revenue).

Here are just a few of the personalization experiences that you’re now able to build with AI Personalization:

Search and browse personalization

Personalized search tailors results by emphasizing content that is most relevant to the user, creating a more efficient and engaging browsing experience. By analyzing a user’s past behavior, preferences, and search history, personalized search algorithms can deliver results that are uniquely suited to the individual’s needs and interests. This approach not only saves time by reducing the need to sift through less relevant information but also enhances the overall user experience by presenting the most pertinent content upfront. Transparency can be improved by indicating personalization boosts.

Personalizing Autocomplete

Autocomplete results present a valuable opportunity to surface personalized content right from the start of the user’s search journey. By leveraging data on previous searches, user preferences, and behavior, autocomplete can suggest relevant results even before the user has finished typing. This not only enhances the efficiency of the search process but also significantly reduces the time it takes for users to find the information or items they are looking for. As a result, users enjoy a smoother and more intuitive search experience, with personalized suggestions guiding them swiftly towards their desired outcomes.

Facet reordering based on user preferences

 

Search facets can be personalized for each user to prominently feature their favorite brands or categories, creating a more tailored and efficient browsing experience. By analyzing user behavior and preferences, search algorithms can prioritize these preferred facets, ensuring that the most relevant options are easily accessible. This personalization allows users to quickly filter through vast amounts of information, honing in on the items or content that matter most to them. As a result, the search process becomes more intuitive and enjoyable, aligning closely with the user’s individual tastes and needs.

Personalized recommendations

Personalized recommendations enable users to discover a wealth of content that aligns closely with their individual needs and preferences. By leveraging user profiles built with AI Personalization, recommendation algorithms can curate a selection of content that is both relevant and engaging. This tailored approach not only enhances user satisfaction by presenting them with options they are likely to enjoy but also encourages deeper exploration within a platform. Consequently, users can uncover new products, services, or information that they might not have found otherwise, enriching their overall experience and increasing the likelihood of continued engagement.

Inline segmentation

Inline segmentation allows for the display of content specifically tailored to users who share similar interests, enhancing the relevance and impact of the information presented. By utilizing user profile affinities, personalized content such as banners, calls-to-action (CTAs), and other targeted elements can be seamlessly injected into the user experience. This approach ensures that the content resonates more deeply with the audience, as it aligns with their specific interests and preferences. As a result, inline segmentation not only improves user engagement and satisfaction but also increases the effectiveness of marketing strategies and content delivery, fostering a more personalized and compelling user journey.

Promotions by segment

Use ruleContexts to create rules to promote products based on segmented end-users, such as “new” or “returning,” “mobile” or “desktop,” “loyal” or “inactive.”

Exporting Algolia-enriched user profiles into 3rd party platforms to power custom/internal workflows can easily be accomplished with the affinity parameter  /users?affinity={affinity}. Learn more in the Algolia Docs.

AI Personalization is available starting today in public beta on high-availability infrastructure around the world. It’s deployed through a single API alongside our extensive libraries of popular coding languages and free community-supported connectors; or you can leverage it through our Merchandising Studio and powerful dashboard.

Learn more about Algolia AI Personalization and sign up for our public beta at https://www.algolia.com/products/features/personalization/

To learn more about all of our solutions, please reach out to one of our search experts.

About the authorCiprian Borodescu

Ciprian Borodescu

AI Product Manager | On a mission to help people succeed through the use of AI

Recommended Articles

Powered by Algolia AI Recommendations

How to master personalization with AI
AI

How to master personalization with AI

Ciprian Borodescu

Ciprian Borodescu

AI Product Manager | On a mission to help people succeed through the use of AI
Search personalization 101: how to capitalize on personalized search and discovery
UX

Search personalization 101: how to capitalize on personalized search and discovery

Catherine Dee

Catherine Dee

Search and Discovery writer
E-commerce personalization: pitfalls, tradeoffs, and solutions
E-commerce

E-commerce personalization: pitfalls, tradeoffs, and solutions

Eunice Lee

Eunice Lee

Matthew Foyle

Matthew Foyle

Solutions Engineer @ Algolia