7 Common Mistakes to Avoid in Speech Analytics


Seven sounds like a big number, but speech analytics can be a big investment of time, money and people.  When done properly it can revolutionize a contact center and when done incorrectly, it can syphon off precious resources and erode internal C-level engagement.

Consider this a list of warning signs to watch for and react to, which will help you avoid common pitfalls:

  • 1. There’s no senior sponsor, no C-level buy-in. This is trouble, because an investment as significant as speech analytics, which is going to require cooperation from different departments, is going to need a champion who can run interference across boundary lines. The death of your speech analytics project is if it’s seen as “a contact center thing,” so find senior sponsorship outside of the contact center itself.
  • 2. The rest of the organization is unclear on the benefits. Speech analytics has concrete benefits for marketing, for plenty of other people in the organization outside of the contact center. Make sure they’re aware of that, and you’ll find cross-department cooperation goes much smoother. Your enterprise-wide steering committee, in fact, should include people from various departments.
  • 3. The technology & processes are expected to do too much at first. Take it easy. Donna Fluss recommends “do not try to do too much all at once with speech analytics. Allow the analysts to concentrate on two or three issues at a time.” And hey, allow for the growing pains – “This is important,” says Art Schoeller, principal analyst at Forrester to MyCustomer, because a lot of projects fail on this point: “I’ve seen companies do the proof of concept, implement the technology, get some results that they think are great, and then they get a little more into production and have a couple of instances where they’re doubting what the system is saying, or making changes based on findings and then find out that it isn’t really happening. So where I have seen systems fail is when they end up being viewed as being untrustworthy.” Allow for fine-tuning. “When you’re dealing with speech analytics, it is more unstructured data and so you’re dealing more in probabilities” than the black and white contact center people are used to, he says. “For example, the likelihood that we saw this phrase coming out of speech analytics 5,000 times is 88%. It’s not exacting.”
  • 4. Internal analysts are improperly prepared. As Fluss would agree, far too many businesses throw analysts on the job with far too little training. “Thoroughly train the speech analytics analysts to use the application and to understand your business,” she says, noting that “a good speech analytics analyst should be able to identify issues and provide a subtle and nuanced interpretation of the underlying business drivers.” She suggests having all speech analytics analysts listen to calls for a week before they even start formal training.
  • 5. All the solution does is crank out reports. Lori Bocklund, president of Strategic Contact, an independent consulting firm for customer contact technology and operations, told SearchCRM recently that reports “aren’t terribly meaningful to people. Action is the operative word.” If your speech analytics project doesn’t include action triggers, on an individual level and an organizational level, it’s not set up correctly: “Make that a routine part of the process. Don’t report on just ‘this is what the speech analytics is telling us,’ but ‘here’s what it’s telling us, here’s what that means to us, and here’s what we’re going to do about it’.”
  • 6. It works, but nobody cares. One surprisingly common mistake is not spending enough time to “find a meaty business problem before pulling the trigger,” according to Martin Hill-Wilson, director at Bridge House Consulting, who says when launching a speech analytics project, the thing to do is look for significant business pain so that “the insights uncovered by the analytics have instant relevance and appeal to the business.” Does wonders for resource prioritization. Picking a relevant business issue gives you “a wealth of low-hanging fruit for improving servicing costs, customer satisfaction scores, staff productivity and motivation,” he said, adding that “even the broader issue of front to back office efficiencies can be part of a quick wins program.”
  • 7.  You didn’t build in response systems. Again, you’d be surprised how common a mistake it is to fail to put in place systems for getting the data and insights your shiny new system is generating out to the parts of the business where it can do operational good. You simply have to establish some sort of formal reporting process where you’re not simply generating reports nobody reads and acts upon, but feeding the information into a closed-loop process, as Fluss says, “that measures improvements on an ongoing basis, with managers recognized and rewarded for tackling and correcting the issues identified by speech analytics. Producing pretty charts is not what speech analytics is all about, although it’s nice to get those. Disseminating the information and addressing the underlying issues and/or identifying sales opportunities is number one..

These are common symptoms of a speech analytics program sliding sideways.  Watch for them carefully to keep your program on the straight and narrow. 

Click Here to learn more about ZOOM Speech Analytics solutions:  ZOOM Int. Speech Analytics

Tell us what you think:


Captures and improves the omnichannel customer experience across calls, emails, chats, surveys, and more. Covers more platforms: Cisco, Broadsoft, Genesys, AudioCodes(radios), Microsoft Skype for Business, and Amazon Connect.

Learn More >>

Get a Demo

Are you ready to take a closer look at what ZOOM has to offer?


About ZOOM

At ZOOM, we give contact centers of all sizes the tools to improve the customer and agent experience while addressing back office compliance and risk. That's why we've made it our mission to lead by example, and continually aim to improve our tools and set new challenges for ourselves.

Since 1999, ZOOM has held a world class track-record of customer satisfaction, scoring 82%+ with Net Promoters. Our customers range from sub–100 agent contact centers to some the world’s largest companies—Finansbank, Cigna, Rostelecom, IBM, and Saudi Aramco.

ZOOM has solved some of the hardest call recording problems in the world, and the new ZOOM Omnichannel Search Engine is the first of its kind.

Learn more about ZOOM »