Natural language query: 5 benefits of Guided NLQ

Natural language query: 5 benefits of Guided NLQ

As part of our series on natural language query (NLQ), this blog details 5 benefits of using Guided NLQ, and how it differs from search-based NLQ to bring users true self-service BI.

 

Many analytics vendors today offer search-based NLQ tools. To explore data and find insights, you must use free text, but you also have to know what, or how to pose a query.

The problem here is obvious: There is no guidance on what to ask your data, or how you can use the tool to ask questions and get helpful answers. The analytic burden is placed entirely on you to learn what question to ask. But what if you're not an analytics expert?

The search-based approach, more often than not, results in blank search field syndrome. You may not know what to type in, or are left wondering what to do next, leading to frustrating attempts and failures, and disillusionment in natural language altogether.

With Guided NLQ, however, there is no blank search field syndrome. You’re guided through the entire experience to ask the right question, and find the right answers.

 

Benefit #1 - Guided NLQ is a unique self-service BI experience

Yellowfin Guided NLQ provides immediate assistance on the question you want to ask, with no guesswork or technical knowledge required to get started with using the tool.

After selecting a dataset, you’re presented with a search box you can type in, but it’s not blank. Guided NLQ provides a list of options for possible questions, then guides you through each step in formulating the query. You can choose your own path through the question by typing what you want to ask, using your mouse to choose an option, or both.

 

What does natural language query look like

 

These add-on elements can help build your query, and lead toward a more relevant result than traditional free text search. You’re actively shown a list of options in simple drop-down menus, and prompted with suggestions that can help correctly state the question you mean to ask, such as ‘compare’, or ‘list’, using familiar terms, not technical jargon.

Once your query is built, Guided NLQ presents the ideal level of data you need to uncover the answer, delivered as a best practice data visualization (chart), which can also be viewed in tabular form. These answers highlight hidden patterns, trends and outliers or shifts in behavior that can reveal deep insights otherwise not seen in traditional analysis.

From here, you can do a number of things:

  • You can go back at any time to rearrange the question
  • Change your data view to find more answers from other datasets
  • Save your question for later
  • Add the answer to existing content in Yellowfin, such as Dashboards, Stories, or Presentations

 

This fully guided approach to natural language query means there’s no more need to worry about the right terms to ask, or the correct synonyms to type to get a result, as the tool itself quickly generates the most popular search dimensions to help you get started.

You can also easily click ‘show more’ to see all available options in a category. Because the NLQ feature itself effectively guides you through each step, everyone in the business can use it for answers, not just the experts.

 

Benefit #2 - Every question is understood by Guided NLQ

Traditional search-based NLQ solutions are harder to set up because they’re focused on fixing the wrong problem: semantics (language used in a question), rather than analytics.

With Yellowfin Guided NLQ, there is no need to set up synonyms and word dictionaries, or continuously train the solution to understand your users’ intent, because using the Yellowfin metadata layer bypasses this problem altogether.

How it does this is each piece of text in the query you build is known and understood, and by offering guided options to choose from, any ambiguity or misunderstanding in what you’re asking - a problem that limited NLQ adoption in the past - is eliminated.

At no point can an invalid question be asked; there’s no more "Search didn't understand what you meant" messages, because there’s no such thing as a ‘wrong’ question.

 

Benefit #3 - Guided NLQ makes it simple to ask complex questions

The questions you can ask search-based NLQ tools are often too basic because the vendor has spent all their effort in fixing the language problem, and their approach doesn’t support question complexity in the best way.

Guided NLQ approaches question complexity differently by implementing thousands of comprehensively modelled question types and sequences, which effectively enables anyone to ask questions of their data, and to deliver answers as best practice visualizations or tabular reports for every possible question combination you can think.

 

 

Some examples of the complexity supported with Yellowfin Guided NLQ include:

  • Tabular and cross-tab reports
  • Automatic highlighting of items on charts, such as outliers, values, trends
  • Complex filter construction
  • Set analysis comparison, ranking, calculations
  • SubQueries, including minus, intersect

 

Whether you have a complex question, such as finding accounts that had more sales this month vs. last month for specific product SKUs, or a basic question, such as a comparison of annual business performance from one year to the next, Yellowfin Guided NLQ has been specifically built to accommodate your query.

 

Benefit #4 - Guided NLQ is integrated throughout Yellowfin

Guided NLQ is designed to combine with existing features of Yellowfin for a powerful analytics experience that accommodates all users and self-service BI preferences.

It’s fully integrated with Yellowfin Dashboards, Stories, and Presentations, which makes it easy to generate and add new content, and any questions and results generated using Guided NLQ can be shared using existing Yellowfin collaboration functionality. It also contains multi-language support, the same security model, and is multi-tenant enabled.

 

 

Most of the output from other NLQ vendors, in comparison, are siloed in their tools; you can't really do much with it after. Whereas in Yellowfin, you can because it's integrated with other content and functionality, and can form part of your analysis workflow.

With Guided NLQ, you can ask an ad-hoc question and immediately drop that into other content that you're working on, or share it with other colleagues. If you were working on your own content already (such as Dashboards, Stories, etc), you can access Guided NLQ from those builders as well, and drop the answers into that with a seamless workflow.

Learn more: Yellowfin BI - The Full Platform

 

Benefit #5 - It's easy to embed Guided NLQ into your applications

Yellowfin Guided NLQ is designed from the ground up to be easily embedded.

What this all means is the feature can be used independently of the rest of the Yellowfin platform, plugged into any of your applications, and launched from anywhere you want, whether it’s a customer relationship management (CRM), human resources (HR) payroll, or finance system. It can even co-exist within Tableau and Power BI environments.

As a stand-alone module not tied to a user interface (dashboard, workbook), or single data set, you can curate a view and drop in NLQ capability for quick and easy self-service deployment, and it's API-enabled to provide fine-grained control and a customized experience. You can even allow users to ask questions of any dataset, or limit the scope of what can be asked to ensure relevance to wherever you decide to embed Guided NLQ.

What to use NLQ in analytics for

For independent software vendors, this level of flexibility can be leveraged to white-label Guided NLQ as an attractive feature that can help customers quickly create their own analysis without being a support burden, while further enhancing the product's value.

In enterprises, data analysts are usually the ones engaging in self-service analytics because it has a big learning curve, and non-technical business users don't have the necessary skills to perform it themselves, nor the time to build those skills. Guided NLQ gives these business users through the enterprise the ability to self-serve BI without having to rely on scarce data experts or analysts every time they want to explore data.

 

Why Guided NLQ is a differentiated, future-proof solution

Yellowfin’s approach with Guided NLQ is driven on analyst-backed evidence and customer feedback from using other search-based vendors that the current approach to NLQ (i.e. free text search) is not working. That approach has failed to meet user expectations, resulting in NLQ tools that are underused and unrealized.

Yellowfin Guided NLQ, in short, brings true self-service capability that democratizes analytics for non-technical BI users. Everyone can build a query and yield appropriate results much easier, and get relevant results faster without having to request help from analysts. It’s a feature we’re incredibly excited to see how users make the most of it.

See more: Guided NLQ in action

Check out Yellowfin Guided Analytics in our live demonstration video, and learn why it’s set to change the way your users engage with NLQ and perform self-service BI.

You may also like:

What is natural language query (NLQ)?

Natural language query (NLQ) is fast becoming a hot topic in business intelligence (BI) solutions. Learn what NLQ means for self-service analytics in this blog.