Today’s enterprise search users need more than answers. They want context, information, and guidance along the way.

We are experiencing longer-than-usual wait times. Thank you for your patience.

This message became the standard refrain from many support centres during the pandemic. Teams and customers moved home and demand for self-service information jumped dramatically. Unfortunately for many, it shined a light on how disconnected enterprise data can be.

Like the companies that use them, enterprise search tools also need to evolve to meet the more sophisticated expectations of both customers and the organization.

Today most product development in the intelligent search market is focused primarily on developing solutions to narrow results AFTER the search action is complete. SearchBlox has a different approach: we’re moving user decision-making upstream to help users get what they need faster. This approach to innovation in enterprise search is an idea we’re calling “DirectIntent” and we think about it in two ways:

1: Anticipating the searcher’s needs:

Leveraging AI to identify as much context as possible, as soon as possible, about the searcher and their situation so we can create the straightest line to the content they’re looking for, while still protecting privacy.

2: Guiding the searcher’s experience:

Providing the user with relevant options as they go through the search experience to help guide, and even influence, their decision-making as it’s happening. Then, learning from their behaviour to support the next search activity.

Recently, one of our customers leveraged DirectIntent to optimize earlier stages of intelligent search. The results dramatically decreased the search journey, generating a 2x increase in conversions on their site in the first month after launch. Whether serving data to team members searching the intranet or customers looking to buy, it works because users find what they need, or what they didn’t even know they need, faster.

Before we talk about what’s new, let’s look at “traditional” enterprise search.

With traditional search, a user types their term or phrase, hits enter, and is shown an unwieldy list of possible matches to wade through. The search engine’s job is done. In an effort to help users make faster decisions about the results, sorting and filtering options were added to traditional search tools (e.g., filter by price or sort by file type). But, that still puts all the narrowing criteria after typing and hitting enter.

In contrast, the modern intelligent search engine doesn’t stop working when a user enters a search term. Instead, artificial intelligence allows the platform to monitor how the user interacts with the results. Then it applies what it “learns” to the next search experience by adjusting future results before a single keystroke is entered.

Identifying ways to narrow relevant results sooner requires a new way to think about the search experience.

Rather than looking at the search experience as a single step - enter search term - the SearchBlox team thinks about enterprise search in four distinct stages. With DirectIntent, each stage is enhanced by the continuous learning feedback loop that artificial intelligence (AI) makes possible.

At each stage of a search, users are making decisions and taking action.

1: Before Typing:

With intelligent search, by simply clicking the mouse into the search bar, the platform offers possible search criteria to help users decide where to look for information. The user hasn’t exerted any effort yet. Still, based on organizational goals (promotional), trending topics (popularity), or individual predictive analysis (personalization), it’s possible to offer up options that serve their needs. Suggesting topics of interest can narrow or broaden ideas about what the user may be, or should be, considering. For example, the search platform could dynamically populate details about the top trending products for the day or the most relevant topics for the season to help call centre teams or customers choose how to direct their search.

2: During Typing:

Have you ever found yourself adjusting your search terms as you see the suggestions the search engine is offering? Without much effort on the user’s part, search engines powered by AI help users make choices just as they’re beginning to articulate their questions. Like the first stage, the suggestions can come from promotional, popularity, and personalization goals. It also includes the learnings from previous user search experiences. For example, a physician or patient researching medications with hard-to-spell names can count on autocomplete to make the right suggestion with just a few keystrokes.

3: After Typing:

This is the stage where traditional search starts and stops. However, an intelligent search platform that has optimized the first two stages is sure to deliver highly relevant material on the first try. In addition to a list of refined results, AI makes it possible to offer specific answers quickly in other formats. For example, the tools can automatically pull an answer out of a document and provide that specific snippet of text as a search result. The user gets what they need without having to click any links at all. AI-enabled selection filters and sort ranking tools also dramatically narrow the list behind the scenes based on user-driven options, administration tools, and real-time analytics.

4: After Selecting:

During the fourth stage, AI monitors and gathers data about how humans are interacting with the technology. Then, it feeds information back into the system so the search platform can learn and adjust, creating a continuous improvement loop.

Traditional enterprise search was limited by an IT team’s ability to guess all the ways people might articulate a question. Intelligent search avoids that by capturing all the contextual data surrounding a search - different terminology, how questions are naturally asked, what pages the user just looked at, whether they clicked the back button - to make adjustments for the next search experience. Activity data also produces insightful reports that help data managers adjust search results manually, for example, determining the exact placement of a document ranking if needed.

The goal of enterprise search has always been to get the best result in the shortest amount of time. By breaking the model down into these four stages, we can define “best” in smaller and smaller increments, narrowing relevant results at each stage of the search process and combine those gains to improve engagement from start to finish.

https://www.searchblox.com