“How to implement speech analytics” is one of the question we wished we were asked more often! However, currently the focus is still on technology itself; the issues of security, speed or accuracy. What many businesses do not realise is that the success of their voice analysing venture largely depends on the onboarding process and how much preparation has been put into receiving the new technology, new data and the new findings.
In this article we have collected advice on speech analytics implementation from the industry’s top specialists. We felt it was important to capture many different persepctives, which is why we’ve reached out to a CX guru, an SA specialist, a provider, a recent SA onboarding champion, a SAS Account Executive, and a tech know-how master.
Speech Analytics can be a powerful tool. The goal is to “listen” to the voice of a customer and not only hear but get a feel for the sentiment of the conversation. You not only hear words, you get the emotion, feeling and meaning behind them. Used correctly, this is a powerful opportunity to understand customers and train front-line reps on how to handle various customer interactions. Mismanaged and you waste time, money and potentially alienate customers through misunderstandings.
When bringing a Speech Analytics program to your support center, consider the size of the customer base. You can’t get a real understanding with just a handful of customers. It takes thousands of conversations and lots of data to get the full picture. One of the biggest mistakes an organization can make is to start using the data and information too soon. Yet, used the right way, Speech Analytics can help drive higher customer satisfaction scores, help employees better understand customers, and more.
You should be prepared to establish a long-term relationship with these people. Your ability to properly acquire valuable metadata as well as a steady ingestion of audio and video heavily depends on your dynamics with these departments. I always advise new SA customers that they will need to start making metadata requests with BI at least 6-8 weeks out prior to launch. You also want to include these folks early in the onboarding process so that you can be on the same page about what you will need from each other.
Speech Analytics is a powerful ally but it requires proper support and understanding to benefit from its full utilization. Gaining “insights” does not occur automatically – there is a fair amount of trial and error involved. Get dirty and experiment in the tool. Become comfortable with the data and soon those insights will come to you.
Do not be afraid to reach out to your speech analytics software vendor. Often SA software companies offer “think tanks” or mass conference calls with other users of the software for you to bounce ideas off. Most of us in the SA community are very open and willing to offer strategies and advice to anyone who reaches out. We have all been in your shoes before, so if you need something, say something!
Pick two or three key business challenges to go after first. Conversation Analytics can do a whole lot, so don’t be tempted to boil the ocean straight away!
Ensure senior leadership buy in and make your core team a part of your analytics insights reporting program. Some customers’ best practice is to set up a regular monthly or quarterly session with Senior Sponsors to take them through the insights, actions and progress.
Be ready for new insights and information your conversation analytics data presents you. Sometimes, you may need to manage your internal stakeholder relationships carefully given some of the data and insights you will uncover.
You won’t get the desired results if you don’t have the right team. Hire at least two dedicated full-time resources who are naturally curious problem solvers that possess a balance between left-brain (logical) thinking and right-brain (creative) thinking. Invest in training before deployment and sign up to Customer Communities and User Groups for continuous learning.
Before deployment, review a sizeable sample of calls and associated transcriptions (approx. 300 to 500 random). This allows you to gain a sense of how accurate the transcription is, and helps identify what keywords or phrases need to be added or tweaked. This is not a one-time activity! Commit to ongoing, periodic auditing and tuning.
Be specific about what you expect from Speech Analytics. Make sure you have baselines defined prior to implementation to allow you to measure change & ROI. Getting executive leadership buy-in from the onset will help ensure the insights get acted on.
Integrating speech data with other metadata (Customer ID, NPS, Demographic, Sales, Products etc.) helps maximize the value and usability of insights.
Understand and clearly document the use cases that the outcomes will be used for. Decide which Business Units have responsibility for delivery of the outcomes (it also helps calculate the ROI value of the business case).
Document the cultural/process changes that SA is likely to drive. Clearly map the order of execution. Employ a transformation management team that will be responsible for implementing the changes following the Speech Analytics program findings.
Socialize your Speech Analytics projects and its documentation prior to kicking off the project so that everyone understands the role that they will play in delivery.
A speech tool is not a plugand-play technology. Your organization must be prepared to dedicate the proper resources that will ensure its success.
Remember that superior speech analytics results don’t come baked into a tool. Ensuring the highest levels of accuracy requires multiple iterations of testing. This painstaking, meticulous work involves auditing search results over and over again, adjusting confidence levels and recommending tweaks until you reach the sweet spot of accuracy that’s thoroughly aligned with your business objectives.
Think in advance of the actions you’ll take as a result of knowing that certain phrases, patterns of phrases and trends were detected. What will you do with the significant amount of data you’ll be generating? Have you appointed one or more change agents with responsibility for identifying and capitalizing on opportunities for change? Make sure you put processes in place that enable you to get ahead and stay ahead of the technology.
Emphasize to your agents that this tool isn’t intended as a “gotcha” device. Your agents will likely “sit up straighter” knowing that all calls are being analyzed, even while they harbor suspicions about the new speech tool. Take pains to assure agents that speech analytics will impact them positively—both in identifying “champagne moments” to celebrate and helping build cultural trust within your organization. Consider building the first speech project based on a positive business issue while identifying other process and performance issues for future projects that you can roll out once trust is established. If at some point constructive criticism is necessary, be sure to deliver it respectfully to maintain morale and head count.
Choosing a Speech Analytics vendor is just a prelude to voice technology onboarding. The key to a successful Speech Analytics implementation lies within the willingness and ability of the business to change. We hope that the above tips will help you prepare for that change and address the transformation challenges.