How Apps Drive Change in Analytics
Analytics could change the way people work and live.
As industries adapt to the application economy, analytics in particular is experiencing its own type of transformation.
Analytics is defined as the process of capturing, analyzing and presenting data to generate insights for the purpose of making informed, data-driven decisions and actions. This definition has not changed for decades. Also, the process to get from an event to a data-driven decision and to action has not changed. But what has started to change is the shift away from generic analytics technologies, such as data warehousing, reporting and dashboard tools, to specific analytics applications, such as fraud detection, food production and real-time news generation.
The role that latency plays in the analytics process is critical to understand why this change is happening.
Analytics-driven apps help humans conserve resources, increase production, understand what is happening across the globe and feel safer about online security.”
— Joav Bally, Vice President, Analytics, CA Technologies
Why Latency Matters
Latency may have little impact in some situations, but latency negatively impacts achieving the full value of analytics. Latency, from the onset of an event through analysis, cause determination and corrective action, is often lengthy. So why is high latency an issue? Because, to capture an opportunity one has only a finite period of time. If the latency is higher than that period of time, the opportunity is lost.
Take online retail, for example. In a DDoS (Distributed Denial of Service) attack, speed is of the essence. Hackers can quickly overwhelm the site with requests so that legitimate business is curtailed. Analytics will likely “see” the attack right away because of the extreme number of incoming messages amid other anomalies. If action is taken immediately, the attack can be stopped before serious impact to the business. If action is delayed, then real business will be lost.
One of the biggest contributors to high latency is, of course, the human factor. Wherever in the analytics process human beings are involved, the latency goes up dramatically. If a human being had to make the decision to react to the attack and shut down messages from specific sources, the associated latency would have assured the hackers’ success. Not only do people take time to make a decision, but also their availability is not guaranteed. However, if the messages from suspect sources were shut down automatically, the resolution could be in effect within minutes or even seconds, rather than hours. Business could then be transacted as usual.
Over the years, analytics technology has successfully reduced some sources of latency, but it could not eliminate the human factor completely. However, now might be the time when the human factor could be removed from the analytics process. There are four key drivers making this possible:
Open Source—All analytics technologies are available from open source (for example, Hadoop, no-SQL DBs).
Infrastructure as a Service (IaaS)—There is no limit to computational power.
Data science—Advancements in data science have made it possible to extract and deliver actual insights from huge amounts of data.
Democratization of data—There is a lot of data available for everyone to use for analytics purposes.
What happens if the human factor is removed from the process entirely so that the decision is made and automated action is taken entirely by the analytics engine itself? A door is opened to an entirely new market.
The app economy represents a welcoming environment for software analytics. Analytics can find a home with any app and is making apps smarter, opening an enormous market with huge potential. Let’s look at some of the apps available today that are benefiting diverse markets.
Banjo is a company that—according to its website—instantly organizes the world’s social and digital signals by location, giving an unprecedented level of understanding of what’s happening anywhere in the world, in real time. With a vast store of data, organized by location, they have created a picture of what is normal anywhere. They analyze everything to create that baseline—visual and audio—the people, landscape and language. Their analytics are deep, the number of calculations enormous. Any change to normal triggers immediate activity to understand what is happening.
AgSmarts’ tagline is “smarter fields, stronger yields”. But that is not the whole story. As the population grows so does the food production need to grow, and water is the essential at-risk component. With moisture-sensing sensors throughout a field, AgSmarts uses predictive analytics and automated equipment to water when needed, thereby increasing production while sustainably conserving water. Their analytics take advantage of multiple sources of information, such as ever-improving weather forecasts, to optimize the yield while conserving resources.
Most people don’t trust organizations to keep their identities safe or their personal information private, according to a survey performed by Feedzai, a data science company that uses big data analysis and machine-based learning to accurately analyze data in real-time to prevent fraud and to keep customers’ personal information and transactions safe. Feedzai provides go/no-go decisions in real-time, within the normal application flow, and over any channel—in-store, online, mobile—keeping commerce moving safely—and that’s good for businesses.
Analytics-driven apps help humans conserve resources, increase production, understand what is happening across the globe and feel safer about online security. Analytics-driven apps have the power to change the way people work and live.