Collaborative Personalization

True personalization down to the individual.

Constructor Search uses collaborative personalization to make personalized search recommendations for every user with every query. Personalization is a proven strategy for improving customer loyalty and revenue. And Constructor customers are reporting up to 6% lift in “add-to-cart” transactions solely from collaborative personalization.

natural language processing

What is “personalization?”

Constructor believes that personalization centers on using customer data and algorithms to present results for a specific user as they are shopping.

Unfortunately, many ecommerce and search platforms, use less robust definitions of personalization. Some claim that using customer data records to show a pop-up that says, “Welcome back, Shirley” is personalization. Others claim that using  membership in a fixed demographic cohort to show items of interest to women, teens or Australians is personalization.

That’s not personalization to you or Constructor. We believe that personalization should improve the experience on your site for every individual. We believe personalization has the power to improve revenue and customer loyalty for site owners. Personalization is more than a buzzword, it’s a source of competitive advantage for your business and it’s what users expect today from search.


Personalized Search Recommendations

All personalization starts with knowing your customers’ preferences. Search personalization succeeds when it makes accurate, individualized recommendations.

e-commerce user personalization

Consider a customer who regularly purchases organic products on your site: milk, bananas, meats, etc. If the customer searches for “apple,” will they see “organic apple” at the top of the results? If your site uses Constructor Search, the answer is likely to be yes!

Constructor Search makes individualized recommendations using product affinity data from this customer and many others. We call this “Collaborative Personalization.” By filtering purchase histories of other apple buyers, and analyzing them against our customer’s organic-heavy purchase history, Collaborative Personalization algorithms will likely find and rank organic apples above other varieties. And this is without the word “organic” being in the original query.

What’s more, sites using Constructor Search report as high as 6% revenue lift from Collaborative Personalization-based recommendations alone!

Collaborative personalization features

Improve user experience by tailoring search results for each and every user on your site.

Behavioral data

Constructor Search tracks query terms and counts of click-throughs, conversions and purchases to improve future search outcomes. User information is hashed to ensure privacy.


An advanced collaborative filtering algorithm for predicting the purchase intent of a user by evaluating their preferences to the preferences of other users who have similar purchase intent.

Personalized search recommendations

Leveraging available personal data, Constructor returns personalized search recommendations. For example, If a user demonstrates purchase intent for “organic” items, “organic” items will will be ranked higher in future search results, regardless of her query.

Repeat purchase recommendations

Site owners can optionally configure Constructor search to prioritize repeat purchases. For example a customer who previously purchased a 64 ounce jar of Skippy peanut butter will see that exact product SKU among the search results for peanut butter.

In-session personalization

Collaborative personalization is computed automatically in real time, ensuring the best possible search recommendation in a current session and over time. No human intervention or batch processing is required.


Constructor Search values the privacy of your customers and honors your site’s privacy policy. Users remain anonymous within our systems to ensure privacy.

The advantage of true personalization

Personalization drives customer loyalty and revenue, and sets your organization on a path toward durable competitive advantage.

Personalized search recommendations represents an immediate opportunity for Constructor customers, because the feature is not available today at large ecommerce sites, including Amazon. Rather than fall into false traps like assuming that your current search is “good enough” or that you don’t have the capacity to tackle personalization, talk to Constructor about your path to personalized search.

Collaborative personalization works along with natural language processingmachine learning-enhanced reranking and merchant controls to provide an automated solution that improves conversions, revenue and customer loyalty.

Don’t just trust us. Make us prove it.

Let us quantify the value of Constructor’s ML-backed search and discovery on your site using your data. No contract required.