5 Ways Speech Analytics Can Improve Knowledge Management


Frank Sherlock of CallMiner promotes the use of speech analytics in the contact centre by sharing five ways in which the technology can improve knowledge management.

1. Gain a Full Picture of Customer Query Types

If you want to improve advisor performance and effectiveness, there is no substitute than to be evaluating them on 100% of their customer interactions.

In order to do that, you need to capture and analyse engagement across all the channels they use. Anything less means that any training you carry out will be based on a guess at what is needed for coaching, rather than a full picture of what is happening.

Using an imperfect picture, based on an incomplete knowledge base, will also demotivate your employees because they will think it is not a fair reflection of their true performance. The net result will be a waste of time, effort and money – and unhappy advisors.

So, make sure you capture every interaction so that speech and text can be analysed in a single platform.

2. Create Analysis Categories That Drive the Right Behaviour

It’s great to have a sea of data, but you can drown in it if you don’t create the analysis categories that drive the right advisor behaviour.

One CallMiner customer wanted to up-sell insurance policy indexation to all their customers. So a category was built to identify “up-selling language” relating to indexation on every customer call. This was then applied against a call universe of ‘Sold Policies’.

A scorecard was then created to monitor who was and who wasn’t attempting to up-sell. This allowed the supervisors to identify best-practice approaches used by some agents and coach other agents who needed help to improve.

Also, it helped to identify gaps in knowledge that the learning and development department filled by creating simple workshops that enabled agents to understand how indexation works and how to sell it effectively.

The contact centre saw a 43% increase in the penetration of indexation in just 11 months.

3. Provide Advisors With Personal Scorecards

In my experience, agents want to feel like they are experts in solving customer problems or creating the perfect solution for customers.

In short, I think agents want to feel like they send customers away happy, so they welcome being able to see where they perform well and where their colleagues might be performing better.

Providing agents with a personal scorecard that shows in a very simple and visual way how well they have performed on a shift satisfies that need. It also identifies where they may need targeted coaching to help them raise their game.

The great thing about providing personal scores that are based on 100% of interactions, is that agents see this as a totally fair assessment of their performance. They therefore welcome the insight so that they can be in control of their own improvement.

For example, if you are targeting an improvement in the use of compliments and an agent sees that they are ranking seventh out of a team of 10, they can decide to work on improving that on their next shift. They can also ask for coaching to accelerate their improvement.

4. Use Artificial Intelligence to Turn Your Knowledge Base Into Predictive Insights

The brilliant thing about capturing every interaction in a single knowledge base is that the value of it grows over time – especially if you have artificial intelligence-fuelled analytics. This enables the analytics program to begin to predict outcomes from behaviours.

For example, by analysing every interaction over time, you will be able to identify words and phrases, and the tone with which they are delivered, that deliver positive or negative outcomes.

Also, you will be able to predict the reaction of customers to a new marketing initiative by your own company or that of a competitor.

On the one hand, this will enable your marketing team to create marketing campaigns that drive the desired customer or prospect behaviour and, on the other hand, it will enable your advisors to outsmart the competitor campaign by responding to calls in a way that creates loyalty.

5. Give Your Agents Real-Time Access to the Right Information

If you want your agents to be most effective, it makes sense to give them help and knowledge when they are on a call – so they can send every customer away happy.

Speech or interaction analytics solutions can provide real-time call monitoring and performance feedback to agents during the call.

Thumbnail image of Frank Sherlock

Frank Sherlock

For example, if the analysis identifies a change in the acoustics of a customer’s voice that suggests the call is deteriorating, it can connect your agents with back-office experts who can guide them or intervene if needed. The system can also provide ‘next best action’ advice in real time, so that the right response is given to the customer.

Driving improvement after a call is great, but driving effective behaviour on a call is even better.

To find out more about CallMiner, visit: callminer.com

Published On: 5th Sep 2019 - Last modified: 10th Sep 2019
Read more about - Industry Insights,


Get the latest exciting call centre reports, specialist whitepapers, interesting case-studies and industry events straight to your inbox.