The Power of Federated Search: One Path to Explore Multiple Galaxies
We all have seen and tried to gather information only to be left short of the desired results. This is often due to the search needs to find, aggregate and respond to the request from multiple data sources. If the input is only searching on a single data source, the results will be limited.
At our Hawksearch Forum, Lance Bohrnell, Sr Project Manager and Stephanie Brudvik, Sr Business Analyst presented on how Hawksearch delivers on results from federated sources. In this blog, we will discuss:
- Why Federated Search – what it is and why it is used
- Federated Search Display – understand how presenting search results to the end user
- Relevancy & Architecture – keys to successful implementation
- Putting Federated Search to work for you – and what it looks like
Why Federated Search?
First, it is important to understand the term federated and what it means. Across your busines ecosystem, the organization runs across multiple systems of records or data bases in the environment. These data sources represent silos of data or information for a specific reason or result. However, when understand the complete view of a business, often times, more than one data source needs to be reviewed. This is where federated search comes in.
These multiple data sources must be considered when searching for information. If the search engine does not have the capability to search, gather and aggregate the information, the results displayed will only be part of the story. Once the data is collected, then displaying the federated search results needs to be presented to the end user.
Federated Search Display
Once the data is collected, displaying a collection of links is not the best way to provide the response. The response needs to be organized and easy to use otherwise the data results may get lost. As we discussed in a recent blog post outlines why Search is key to any digital experience but also explains why the presentation of data is key to success.
Organizing data can also be challenging if not done correctly. There are often challenges with multiple data types. This can include different data sources that have attributes for display, attributes for search, facets, sorting options and layout. These factors must be taken into consideration when addressing the display of the federated results.
So, what are the options? When reviewing the data is presented, the results need to include certain configurable options, so the end user has the best possible journey when searching for information.
Tabbed results are effective when looking and responding to data across federated sources. For example, if there are three systems of records, there should be three tabs in the search results.
Then each of the tabs would be able to display different attributes. These attributes would provide the content from the different data sources. These data sources would then be able to use tabs that can have different facets, sorting and pagination.
Once the data is presented, using merged results can be a good way to view the content. Merged results allow for multiple data sources to be combined and aggregated to provide a comprehensive result to the searcher. When using merged results, labels can be applied to indicate what type of content is being presented and provided to the end user.
Another consideration is understanding scoped search when presented findings to the end user. Scoped search allows users to limit their search to a section or type of content on a website instead of searching everything in one go. Typically, it is implemented in two ways: drop-down scope selection and autocomplete scope suggestions, as discussed in a previous post and also appears in the below example or within the search box.
Combining different techniques is also a good way to present data back from federated sources. Using merged and tabs to display the results can also provide a way to present results but allow for multiple ways to access the information. In this example, you can see how using multiple strategies provides the best results.
With autocomplete and the recommendation widget provides an accurate way to display results but also provide the recommendations based on the request.
Relevancy & Architecture – Keys to a successful implementation
When designing and implementing federated search, it is important to understand the relevancy, architecture and performance. These three factors are the keys to a successful implementation.
Relevancy is important because it allows you to understand how users are searching for data. This request then can tune the search by putting the top on site keywords first. It is important to review relevancy for each data type.
Architecture is another factor to consider with a federated search deployment. One data feed is built from different sources. These different sets of data feeds come from different sources. The need to account for scheduling index then can take place after the feeds are complete.
Performance is necessary because returning the accurate, best results in a timely fashion is important. If the information is not return within a reasonable timeframe, the user may leave your site.
Performance characteristics to consider would be single query performed on an index. This enhances the speed of the results. The tabs allow for aggregation of data optimizing the query time. When reviewing the architecture, making a scalable search response will allow to add data feeds without impacting the performance.
Putting Federated Search to Work for You!
Now that you have a better understanding on federated search, it is also key to understand areas outside of main data sets. These could be:
- EOL products
- Help desk data
- Website content / other websites
- User Manuals
- Warranty info
- Multimedia assets
- Salesforce Community
- And other data
Federated search will enhance your customer’s experience. Customers can access your rich data but also find everything they are looking for in one place. This also means provide results to the customer that helps them find additional information they may not know they needed. This will increase engagement, increase conversion and enhance the customer journey.
Hawksearch provides easy to manage results by using one search application for all data. The tools can be applied globally across your business ecosystem.
Hawksearch allows for expandability and scalability to integrate new data sources. The flexibility allows for a phased approach to grow as your business needs.
We are excited about federated search and the Hawksearch platform. To learn more about the power of the Hawksearch platform, you can contact us at [email protected] or learn more by seeing a demo of the functionality discussed in this article.