Speech Analytics Gets Real (Time)
"It's a good idea to look at real-time capabilities but fix critical issues while they're happening post-call instead of in real time."
Ziv's recommendation means looking at the results from rules that have been set up in post-call. Predefined rules can determine what is relevant and what would be a relevant response to make. "You have to test predefined rules, and the only way to do that is in post-call, so, in hindsight, you can analyze different scenarios and see what agents say or don't say," he says. "Then, when that is combined with other data, you can determine what needs to be looked for in real time."
How to Get the Green Light
By this time, you may be sold on real-time speech analytics and are itching to get it deployed as soon as possible, but where will the money come from? Making a business case to upper management may not be all that difficult.
"There's a lot of confusion in the market about what real-time actually is," Napierski says. "Our product—Eureka Live—and all other real-time products would be more accurately described as real-time monitoring. An automated supervisor can alert agents based on what's happening on a live call. Additionally, Napierski believes that most vendors "blur" the lines by calling monitoring products analytics. "As true post-call analytics continue to get faster and more efficient, that line will get blurred even further," he says.
"The return on investment [comes from identifying] things that are happening in real time [and] that could increase the percentage of saves. You have the ability to proactively solve customer problems before they happen [or] before they get to the point where they want to leave," Pellet says. Another way to convince management to invest in real-time is to talk about customers' buying habits. There might be things that a consumer says that indicates he is interested in buying something. "Companies can use real-time speech analytics to understand behavior elements. Agents can receive suggestions from the system and pass those on to customers. We've seen significant uptick on the selling side by being able to use the conversation to drive what types of offers to give customers," Pellet says.
Before You Buy
After an organization has done its homework and decided to invest in real-time speech analytics, the experts recommend taking caution when evaluating solutions, especially considering that this is not a one-size-fits-all situation.
"All products that predominately do keyword searches and phrases are not equal, Fluss says. "That's very different than tools that are designed to do forensic analysis and trends that you didn't ask the system to look for, things that you didn't even know to ask about. There are very big differences in solutions, which buyers should be aware of when evaluating a purchase."
There are not only differences in task performance, but disagreement about what real-time solutions actually do as well.
"Over the next couple of years, real-time speech analytics is going to be a requirement for interactions," Pellet says. "In addition to the technical advancements that are becoming available, [Nexidia thinks] that real-time needs to be pointed at the right problems to be effective versus overwhelming. Companies must first quantitatively understand what problems they are trying to solve and how they want to solve them, prior to automating the resolution process through real-time."
Hope or Hype?
Many people in the speech analytics market feel that there is a lot of confusion and that the solutions that consistently perform will bubble to the top.
"Real-time speech analytics is being thrown around by a lot of companies, [Nexidia] included," Pellet states. "As the cream rises to the top and companies succeed in real-time deployments, there's going to be a significant shift in the market. A lot of solutions are claiming things but aren't able to deliver, and that's the area that we're targeting."
Gallino agrees. "There's a lot of hype in the real-time speech analytics market. It's easy to sell and therefore it's easy to believe. Over time, the marketplace will make its choice about post-call analytics based on results. I also think that this will happen in real time—we're already seeing it."
Senior Editor Michele Masterson can be reached at mmasterson@infotoday.com.
Related Articles
Enhancements provide visual data analysis and improved tools for review and workflow for the legal market.
02 Feb 2015
Nexidia AudioFinder will help the law firm and collections agency ensure compliance.
27 Jan 2015
Increased targeting and precision make an upgrade worthwhile.
10 Jan 2015
Real-Time Speech Analytics can provide agents and supervisors immediate feedback during the call.
17 Oct 2014