Discover six ways to optimize data in the contact center—and avoid information overload.
Posted January 22, 2019
In the age of IoT, personalization and convenience, prevailing wisdom says big data is king when it comes to enabling brands to understand and cater to their customers’ wants and needs.
And indeed, there is truth to that wisdom. Big data sets are excellent for finding aggregate level trends, and for identifying ways to raise overall performance against common key performance indicators (KPIs) like CSAT, NPS and others.
With artificial intelligence (AI) and machine-learning capabilities, we have more means than ever to collect terabytes of information on just about anything. But many companies have fallen into the trap of collecting data just because they can, and are now finding that big data isn’t a cure all for what ails them.
Unfortunately, the rapid rise in our ability to collect data hasn’t been matched by our ability to support, filter and manage it. “The presumption is that you need massive amounts of data to come up with valid insights, but the reality is that you don’t. What you need is smart data, and smart data can originate from ‘big’ or ‘small’ places,” says Bob O’Donnell, president of TECHnalysis Research, a tech market research and consulting firm based in the Bay Area.
Data analytics is growing but not without challenges
The support for data and analytics in the contact center is growing by leaps and bounds. Dimension Data’s Global Contact Center Benchmarking Report found that the personalization of services supported by data analytics will be the top trend in the next three years, with the industry projected to reach $56 billion by 2020.
While big data holds the potential to transform operational effectiveness and deliver improvements in customer service, its implementation can be challenging. A recent McKinsey survey noted that a majority of business executives felt their data analytics goals were not being met. CIOs and CMOs, too, see data as part obstacle and part opportunity.
“The best analytics are worth nothing with bad data,” according to McKinsey. “The importance of understanding and working on all components of the insights value chain is mission critical.”
O’Donnell argues you don’t need sophisticated algorithms to make significant improvements. He says simple pattern recognition and basic data analysis can often move the needle more than complex means. “Try and look for patterns that you can correlate together: call patterns, types of questions, and indicators like time of day,” says O’Donnell. “It doesn’t have to be a super complex algorithm. Just try to tie data points together, and you recognize something you never realized.”
Keys for maximizing both big and small data
Here are some key ways to improving your data analytics operation in the contact center, and to make sure your company isn’t suffering from info obesity, or “infobesity.”
1. Collect the right type and amount of data. When collecting data, companies need to start with a specific question they want answered, rather than sifting through terabytes of unfiltered data to find a problem. “You have huge amounts of data. People know something’s buried in there somewhere and they’re looking to find it,” says O’Donnell. Focus on the question first, and then pull the data you think can help answer it, not the other way around.
2. Don’t forget small data. While acquiring big data can be cheap, effectively analyzing it can be expensive and challenging. “The challenge with big data is that it’s only useful if you can take the massive amounts of information being produced and turn it into actionable results,” says TELUS International CIO, Michael Ringman. That often requires an investment in AI-powered technologies and analytics tools.
“Small data, on the other hand, is readily available, having been derived directly from interactions with customers,” says Ringman. More often than not, consumers are looking for increasingly personalized service. Small data can help brands deliver on this by capturing individual customers’ likes, dislikes, preferences and values.
3. Ensure your agents have the data they need. Utilizing integrated Customer Relationship Management (CRM) technology provides your team with access to the personalized customer information they need in real-time. If an agent can’t readily see that a customer has called support five times in the past day, then they are not going to be able to meet or exceed expectations regardless of what any sophisticated algorithm has to say about them. Having a record of their previous purchases, support issues and preferences by support channel will give your team a clear picture of exactly how they should approach every interaction.
4. Enable customers to self-serve. By using data to build a deep and detailed FAQ, and even supplementing a standard FAQ with Conversational Bots, customers can make their own choices about how they want to receive service. In the end, the data that matters most is what the customer needs right now. Self-service is personalization 1.0, but it works!
5. Measure customer emotions through speech analytics and sentiment analysis.Speech analytics can help tag certain language patterns to help identify cases requiring immediate attention. Meanwhile, sentiment analysis can identify which customers are fuming right now. Giving them a little extra TLC in the moment could mean the difference between retaining or losing them as a customer.
6. Identify key interactions to find “root causes” of problems. Stop analyzing all calls, and instead focus your efforts on identifying the most challenging ones and spending more time with them. Speech analytics and sentiment analysis can help identify a starting point. Use these calls to drill down into the root causes of certain problems.
In recent years, big data has promised a lot of businesses untold insights into their customers. But in order to harness its potential, you must have a clear strategy from the start, focusing on quality over quantity in order to break the big-data addiction and delight your customers in the process.