Improving Enterprise Search: Workgrid Announces Launch Partnership with Amazon Kendra

To say that AWS re:Invent was busy would be a colossal understatement. In addition to 65,000 attendees, more than 2,500 sessions and hundreds of vendors, Amazon also announced more than 77 product launches, feature releases or new services. Included in the flurry was news focused on improving enterprise search.

Amazon Kendra, “a new, highly accurate and easy to use enterprise search service powered by machine learning,” touts “a more intuitive way to search, using natural language, and returns more accurate answers so your end users can discover information stored within the vast amount of content spread across your company.” It also aims “to help the company’s cloud customers incorporate functionality such as natural language processing, but without long waits often associated with AI-related projects,” according to a recent Wall Street Journal article that quotes our own Head of AI and Cloud Engineering, Gillian McCann.

This announcement is significant on a number of levels. For one, Workgrid is a proud launch partner of Kendra, which opens up a whole world of opportunity for the Workgrid chatbot. Because Kendra makes it possible to easily abstract specific answers that would normally be hidden deep within disparate and impossible to index resources like manuals, research reports, FAQs, and PDF documents, the chatbot will now be able to deliver personalized information to employees faster and more accurately than ever before.

“Kendra does the heavy lifting of providing the needed AI tools and connecting data sources.”
--Gillian McCann, Head of Cloud Engineering & AI, Workgrid Software


The enterprise search of today functions through the use of keyword searches. Users type a word related to their inquiry and the engine spits back a list of resources that contain that word. There’s no intelligence involved in the process and because of that, the results that are delivered are, more often than not, terrible.

It’s a huge problem because employees spend 36% of every average work day looking for and compiling information they need to do their jobs – and 44% of the time they fail to find what they need. The costs of such inefficiency are staggering, with losses starting in the millions for even small organizations with only 1,000 employees.

Amazon Kendra takes a step in the direction of putting an end to all of that by delivering on the goal of “answers, not links.”

The power of this promise was previewed at re:Invent by Amazon Web Services Sr. Product Manager Jean-Pierre Dodel who explained that responses to user inquiries with Kendra will be broken into three different categories:

  1. Kendra’s suggested answer, which will provide a straightforward answer to the question based on the most reliable information found within the indexed files.

  2. FAQs, so users can explore alternate versions of their question

  3. A list of related documents, arranged in order of relevance by a semantic deep learning model

The Kendra search console will also provide type ahead suggestions for users, so they can understand what types of inquiries have previously been done.


For Workgrid, Kendra is an exciting development because we foresee conversational interfaces playing a large role in enterprise search in the coming decade. We’re already well on our way to making that vision a reality with the Workgrid Chatbot. Launched in October, the Workgrid chatbot provides a self-service interface that makes it easy for knowledge owners (who don’t need any coding experience) to populate the bot with the content it needs to effectively answer employee questions, rather than just provide a list of links.

The Workgrid chatbot also includes a self-service Q&A builder content authors can use to train the chatbot to respond to the wide variety of questions employees typically ask. Backed by the power of Amazon Kendra, the Workgrid chatbot will become an even more valuable tool, with the ability to extract knowledge from the vast quantities of information that exist across multiple repositories (FAQ, Wiki, PDF Documents, guides, etc).

“With Amazon Kendra, we’re excited about the possibility of our customers getting the answers they need quickly and efficiently. Amazon Kendra makes it possible to extract answers directly from unstructured data across multiple repositories and has the potential to fast track our delivery of accurate, better-than-before answers to our customers.”
- Gillian McCann, Head of Cloud Engineering & AI, Workgrid Software

For more information on how Kendra will advance the goal of improving enterprise search, check out the session Workgrid’s Gillian McCann did at re:Invent with AWS’s Director of AI Product Management, Vikram Anbazhagan and Senior Product Manager Jean-Pierre Dodel, “Transform the Way You Search and Interact with Enterprise Data Using AI.”