Find Integration Guide

Overview

Find is a personalized search product that can be used as a replacement for existing search engines or as an added layer to existing search engines, bringing personalization to the results of a search based on information about the individual as well as groups of shoppers and how they interact with products and recommendations in real time.

Find is a search engine that uses shopping behavior and other strategies to adapt results, offering shoppers a faster path to the most relevant products to them.

Find lets the retailer choose the attributes that should be indexed by the search engine. The more attributes you choose, the longer it takes for the search engine to search across the index. Therefore, only choose attributes that have meaningful values that consumers will use to search.

Find generates the most relevant product results by combining static catalog information (product name, brand,etc), product availability, wisdom of crowds behavioral data, and individual personal affinities.

Integration

Integration is the process of preparing your site, implementing the code, setting up your rules for boosting and synonyms. When the process is complete, your site will go live with Find, and your customers will begin to see personalized results in their searches.

There are a number of steps involved in integration. You should work with your Algonomy team throughout the integration process to ensure that the process runs smoothly.

Core Integration

Before anything else can be done, the core Algonomy integration code must be installed. If you already use Recommend, you already have this code in place, however, your Algonomy team should review it with you to ensure that the code is complete and covers all of the things needed for Find.

Feeds

The next thing you and your Algonomy team needs to review or set up are your Feeds. Feeds are how we import your data into our products so that we can display your personalized recommendations, search results and more. Find may require that you add additional attributes to your feeds.

Current Search Solution

Whether Find is replacing your current search solution, or being used to sort the results from your current search solution, your Algonomy team should review your current solution to help them understand the level of results you have been getting, as well as to identify problem areas and expectations.

Another thing that can be beneficial to the team working on the integration is a list of the top 100 search terms used on your site. This provides the team with information on which attributes are heavily searched for so that they can make sure the instrumentation matches.

Staging Setup [optional]

If you are already live with other Algonomy solutions and you need to add new attributes to your feed, it might be necessary to set up a temporary staging environment to test and validate the new feed, the index builds and the API responses with the new attributes. This is because our integration environment uses the production catalog. If you do not require a new feed or you can deploy the feed without testing, this step can be skipped and the settings can be configured directly in the integration portal.

This will include some set up that only the Algonomy team can do.

Production

The final step of your Find integration is to configure the Setup in the Integration/Production environment. If you previously tested a new feed in your staging environment and the new feed has been deployed to production the settings can simply be duplicated to the Integration/Production environment.

The Algonomy team will set up the backend configurations.