1. Insights
  2. Digital Experience
  3. Article
  • Share on X
  • Share on Facebook
  • Share via email

Six ways to make the most of your speech analytics program

Posted March 8, 2018

Despite the many digital support options now available to consumers, voice channels still command 50.8 percent of customer care activities according to Everest Group. But customer preference isn’t the only reason why voice remains relevant.

Consider that on average, there are 15 words per Tweet, roughly 22 words per sentence in a chat, but 1,000 words in a five-minute phone call. Assuming a call center of 250 seats, that is 500 million words per month!

Speech analytics make it possible to extract value from every one of those 500 million words by ensuring that every customer conversation is recorded, transcribed, indexed and analyzed.

But to derive the most value from speech analytics, you must apply its capabilities to the right high-value interactions. Here are six ways to optimize your speech analytics program to enhance the customer experience, boost operational efficiency and increase sales.

1. Improve Quality Assurance (QA)

Improving your customer service operation shouldn’t be as difficult as finding a needle in a haystack. But in practice, identifying solutions in the contact center to improve customer experience or operational effectiveness may be more difficult to find than that wayward needle. In fact, the random nature of traditional QA programs makes effective analysis unlikely. In practice, only one percent of calls are evaluated by customer experience management, leaving 99 percent unexamined. It’s like thrusting your arm into the haystack and relying on blind chance that the needle will be found in whatever hay you can clutch.

Grasping at straws is no way to manage customer experience. Speech analytics, however, can identify clear gaps in service delivery, helping to guide your action plan to improve customer conversations while enhancing processes.

2. Build predictive models

Traditional measures of customer satisfaction, based on surveys and NPS scores, are not able to provide a complete picture of the customer experience. The most disgruntled customers will likely not respond to a survey, which in any case, may not provide a full and clear explanation of the root cause driving your customer out the door.

At some point, these potential detractors likely told an agent in your contact center what was wrong. Speech analytics allows you to capture these negative sentiments and what led to them early on.  You can build predictive models with this knowledge, enabling your brand to take corrective action and reach out to potential detractors to rectify the situation before it escalates.

For example, if you want to hone in on why customers may be cancelling their accounts, speech analytics will provide key words, phrases or themes occurring on cancellation calls, so you can better predict and prevent a customer from leaving. Speech analytics will also detect emotion in a call, and can help analysts find the triggers for anger and frustration. Once discovered, you can create action plans for avoiding them in the first place.

3. Optimize self-help

When a customer calls your voice line it’s often because other touchpoints were unsuccessful. Speech-to-text analysis can tell you exactly why and what to do to improve. With speech analytics, you don’t have to connect the dots between disparate reports to find out where your digital self-help went wrong – your customers are telling you when they call.

For example, a TELUS International bankcard client was looking to improve service during card promotions. Through the use of speech analytics, we identified the top customer concerns and discovered that a third of them were already addressed somewhere on the promotional website. Turning this insight into action, the client added a pop-up to the site that included the answers to those top-of-mind consumer questions, deflecting 3,222 calls annually and saving an estimated $10,150 per year.

Knowing how your customers describe the problems they are trying to solve also makes it possible to align self-help content so that it is instantly recognizable and relevant. Speech analytics provides the method to learn that language, so that you can mimic customer words, phrases and the order in which they ask their questions. If you and your customers are speaking the same language, you maximize your chances of solving their problems quickly, and with minimal frustration.

4. Improve sales and up-selling techniques

Generating revenue from your contact center is a worthy goal, and one more easily realized with speech analytics. You can analyze the differences between every successful and non-successful sales call. What phrases were associated with the successful sale or upsell? Where in the call did the sale occur? What words or phrases are counter-productive? With speech analytics you can compare the terms, phrases, flow of the calls and timing of the ask between top-performing and low-performing sales agents, to unearth best practices.

For example, a banking client had instructed its agents to attempt an up-sell at the end of a service call, with the transitional phrase “can I have another moment of your time?” Speech analytics found this phrase was a sales detractor. Looking at the techniques of the top agents, an analysis showed that agents who told customers that they could be earning interest on their checking account, and then asking if they could set up that feature for them, nearly quadrupled their sales success.

5. Ensure compliance

For highly regulated sectors, how can you ensure your agents are complying 100 percent of the time? Likely the vast majority are, but with traditional monitoring and QA your chances of catching and correcting noncompliance are slim, leaving you unaware of any exposure.

By monitoring for certain keywords or phrases in every call, (or the absence of certain phrases), you can zero in on the calls most likely to be in non-compliance.

For example, one speech analytics client was able to flag eight calls out of 1,000 that were potential risks. Upon further examination, three of those eight were potential breaches and required action. Finding those three calls out of eight was more cost effective, rapid and realistic, than manually weeding through all 1,000.

Speech analytics is the only way to know if your agents are compliant 100 percent of the time, an ability certain to please and reassure regulators.

6. Maximize return with Lean Six Sigma

Speech analytics opens the door to a greater understanding of how brands can improve customer experience across every touchpoint. Knowing where to apply it is a key to deploying the technology successfully. Also important is developing the right process to realize speech analytics’ full potential. TELUS International takes a unique approach by linking speech analytic technology with Lean Six Sigma methodology to ensure maximum return on investment. Using the DMAIC approach (design, measure, analyze, improve, control) we can mitigate risks, improve team member performance, deliver cost savings and improve the customer experience.

Check out our solutions

Harness the power of next-gen technologies and digital accelerators to create the best high-tech, high-touch experiences.

Learn more