Every person on Earth creates about 1.7MB of data every second. Buying a coffee. Reading a blog post. Liking a photo. Imagine a constant flow of ones and zeros emanating from everyone all the time. And businesses are trying to catch it.
Why? For insights, of course. Business and customer data can inform marketing strategies, inventory logistics, sales opportunities, and more. The ability to pull and report on key metrics on an ad-hoc, or as-needed, basis can significantly boost performance, improve efficiency, and drive growth. But data comes in different structures from multiple sources. It’s complex. It’s sensitive. Collecting, organizing, and reporting on it is no easy task.
A BI solution measures critical business metrics and displays them in charts and graphs on a dashboard. Solutions with ad hoc reporting capabilities enable the user to navigate through those metrics and dig deeper into the data to answer specific questions. And the benefits are so high that the BI market has grown to a $17 billion industry.
But not all BI solutions are easy to use. Many require technical knowledge, like how to write SQL queries. For a business, that means it requires skilled employees, like data scientists who have the technical expertise to create reports and analyze the data. But the demand for data scientists is high. So high that supply can’t keep up.
And that’s a real problem not only for businesses but for BI companies as well. So in the BI market, there’s been a shift towards more comfortable user experience, and a focus on simplicity. And Natural Language Query is the next step. But if you look at recent trends, that’s what we’ve wanted all along.
Small Talk, Big Results
We expect a lot out of our machines. Wake us up. Keep track of our appointments. Avoid traffic. Smart devices help organize and improve the lives of millions of people. But in recent years, there’s been a subtle change in how we interact with our tools. We do it every day and probably don’t realize it (or understand the significance).
Now, I don’t mean we ask them about their day or what they did over the weekend. I’m talking about the structure of the language we use to interact with them. It’s natural as if we were talking to a friend.
For example, a few years ago if you wanted to bake a cake, you may have searched Google for “best cake recipes.” But now, you may search, “what are the best cake recipes?”
Even though both queries return the same results, why are we using more extended, more complex language? This trend is more recent than you may think. According to Google, searches beginning with “Do I” and “Should I” increased by 65% over the past two years. Searches that start with “Can I” increased by 85%.
The demand to be heard
In spite of Orwellian naysayers and technophobes, we’ve always wanted to speak to machines like they’re humans. Launched in 1997, popular search engine AskJeeves prompted the user to ask a question in natural language. While native language processing was still in its early stages (and relied heavily on human editors), the engine’s initial popularity marked a desire for users and machines to interact more naturally. Keyword combinations and trial and error weren’t cutting it. Why should we have to bend our language to accommodate our devices? Shouldn’t the tool help us?
Sixteen years after Jeeves first prompted us to “ask,” Google released the Hummingbird update for its search algorithm. Rather than merely determining the meaning of a query, Hummingbird could evaluate the intent behind a query, resulting in more accurate (and personal) results. So we adapted to using natural language with our devices. And then came virtual assistants like Siri and Alexa. Now we can talk to our tools like a human, and they can respond like a human. The world’s information and knowledge are now accessible by one of our earliest learned skills: speech.
So for the data scientist problem, BI solutions had, well, a solution. Apply natural language queries to their platform so any user can navigate and draw insight from data as directly as they use Google.
For example, if the marketing department for a nationwide movie chain wanted to make a report on candy sales for a specific year. But they also wanted to know how movie genres can affect candy sales. Anyone in the department can go into the application and type “what were the candy sales for horror movies in 2011?” Just like Google, the BI solution dissects the question by searching for filter terms: “candy,” “sales,” “horror movies,” “2011.” It collects the data and presents it in a way the solution “thinks” is the best way to show that data, like a bar graph or pie chart.
Boom. That Marketing employee who can hardly make an Excel spreadsheet is now a data scientist. No need to devote IT resources for ad hoc reporting. No need to hire an expensive data scientist. Both the technologically inept and the highly-skilled programmer can create ad hoc reports with the same ease. It’s data democratization at its finest. And it’s the future of BI.
But not all BI solutions are on board. Several top companies still lack an intuitive and user-friendly interface. But others are leading the charge, focusing on empowering every user to gain insight through data analysis – regardless of technical skill.
Technology only improves on itself over time. Just as the push for natural language queries resulted in more user-friendly business intelligence solutions, the growing number of “citizen data scientists” will result in more data intelligence.
1.7MB of data for every human every day. And that number is only growing. FitBits monitoring steps. GPS tracking locations Shopping habits. Tagged pictures on social media. It’s all collected. And for many in the youngest generation, they haven’t gone a day without creating some data. So imagine a solution, like a BI solution, for your life. Personal intelligence.
Imagine being able to search through your life’s data. You could ask things like, “How many times have I been to the movies this year?” “How much money did I spend last week on food?” Just as a business makes data-driven decisions to improve operations, people can do the same to improve their lives. Better health. Better finances. Better life.
Big data is still a relatively new thing on the technological timeline. And it’ll take some time before an individual creates enough data to warrant a BI-type solution. But in the future, a single person will generate more data in one day than I do in a year. And we’ll want a way to aggregate and analyze that data for personal insights. Natural language query is just another step towards that.
About Izenda: Izenda is an application-based intelligence provider that brings critical data insights to end-users across industries. Thanks to Izenda’s embedded BI solutions, businesses can natively integrate analytics into their applications while conserving development resources and reducing time to market. Learn more about Izenda at https://www.izenda.com.