Skip to main content

How Big Data and the Cloud are Changing Customer Service Solutions

By December 4, 2012Article

The customer service solutions space is moving quickly and will look radically different in the next two years. The biggest driver of change is consumers’ use of mobile devices. The current state of large enterprises’ customer service and sales operations still revolves around retail stores, 800 numbers and call centers. But consumers are living their lives in their mobile phones and want to do with their smartphones what they used to do through an 800 number. They are forcing change that enterprises can’t ignore if they want to provide best-in-class service and sales. The challenge for most enterprises today is that they’re relying on a set of disparate point solutions that no longer meet customer expectations in a digital world.

Customer expectations have changed rather substantially thanks to everything from Google to the iPhone and the intelligence that is delivered to customers through Facebook, Google maps, etc. There is a disconnect between what we have in our consumer lives versus how we’re forced to interact with large enterprises that are an important part of our lives.

Consumers are already losing their tolerance for transactions or interactions that are not “smart” and over the next two years will become even less tolerant. By “smart” I mean that an enterprise uses technology and the data it has about the consumer in order to provide a better customer experience and do so in real time.

For example, if I’ve been on a company’s website for 10 minutes and then need help and click to chat and they ask me “How can I help you?” I’m mad. They should already know how to assist me. If I call into a call center with an IVR system and I go through a phone tree before being connected to an agent, all that history should be there and the agent shouldn’t need to ask how to assist me. Why do I have to go online to pay a bill if I’m in an IVR system; why can’t somebody push me the bill so I can pay it from my mobile device? Consumers know that enterprises should be able to do these smart interactions today.

Over the next two years customer services will also need to rely much less on 800 numbers. Customer satisfaction is very low, consumers prefer to self-serve, and it’s a very expensive channel.

In addition, personalization — understanding the customer’s state (location and preferred device) and intent — will emerge as crucial to effective service. There are smart ways to understand these aspects of a customer interaction today with Big Data technology and weave that into the solution enterprises offer to their customers, but not in the status quo point solutions.

“Smart” solutions: real-time, predicting customer behavior, cloud based

[24]7 has focused for the past three years on understanding how changes in consumer expectations and technology enables us to provide better solutions. We’re an early mover in the customer service and sales space and created a single Big Data predictive platform for solving customer problems.

In today’s digital world, an effective “smart” customer service solution includes the following four aspects and components.

1. Customer understanding and prediction capabilities. At the heart of the software there should be an intent engine or real-time decisioning engine that can predict the intent of a consumer and attempt to provide more intelligence to the transaction independent of the channel.

The solution should enable tapping into data from a variety of sources that is sitting in customer service data repositories or embedded in enterprise websites to capture consumer behaviors when they try to purchase something online and then use that data to improve the customer experience or transaction.

An example is how [24]7 helped a major PC manufacturer dramatically change the customer experience for online buying. Less than five percent of their revenue for this particular business line was done online before we engaged; now 35 percent of their revenue goes through this channel. Revenue from sales of this product through the online channel improved from less than $1 million a month to $7 million a month in less than nine months. And customer satisfaction improved by nearly 10 percent.

When a consumer went to the website before and was offered a chat experience, it was based on agent availability and first-in, first-out. Today our Big Data platform is connected to their website and looks at behavior to understand the intent of that consumer. What is the consumer trying to buy, and is she going to find what she wants when she wants it before she leaves and buys from somebody else? So the software improved the company’s ability to prompt the consumer that is really struggling instead of users that aren’t struggling and don’t want the customer service prompt.

Through our Big Data platform’s prediction capabilities to anticipate what a consumer wants, we were able to simplify the online experience. The improved customer service solution transformed the online channel from an ineffective channel to a very effective, smart channel.

2. Connecting disparate channels and systems. The other piece that’s missing in the per-channel point solutions is the ability to connect the channels to enable transferring easily from one channel to the next, which is important in the digital world. The enterprise should be able, for instance, to push rich text to a smartphone if the customer is on the IVR system. But traditional IVRs and other systems weren’t designed for that.

For example, today if a consumer’s airline flight is cancelled, the airline sends an outbound notification of cancellation and asks the customer to call the airline to get rebooked. The customer is usually on the road at the time, and it’s not a good experience.

Today’s “smart” customer solutions connect systems and Big Data to a mobile interaction, which improves the customer experience. For example, we built a capability for one of our airline customers to ask customers if they want the cancellation notification sent to their smartphones. They can then push alternatives to the smartphone in real time. The customer can interact and say things such as, “give me more options” or “give me direct options” and can complete a new airline flight booking in real time in about a minute — including confirmation, choosing the seat assignment and changing a rental car reservation.

3. Machine learning. The solution also should have the capability on the back end to learn from every interaction. For example, the [24]7 engine can text-mine 100 percent of an enterprise’s consumer interactions and then cull out what worked well and what didn’t, which then is the basis for improving the consumer experience the next time. This automated machine learning should occur in real time and work much the same way as search engines do. But today’s “non-smart” solutions can only perform this task manually.

4. Flexibility at the application level. Finally, an effective customer service solution today must be tailored to verticals such as retail, travel, financial services, telecommunications. But it also must be customizable to include an application layer that goes beyond the vertical standardization to address an enterprise’s unique needs. These aspects of the solution are critical in reducing implementation time.

At [24]7 we’ve handled this aspect by working with many of our enterprise customers in various industries to map out the top problems that are easy wins for mobile, online or making the IVR a lot smarter. We’ve prepackaged these as kits that reduce the implementation time. These packages address about 40 problems that are common driver of customer calls and that cause customer-satisfaction issues across companies in most verticals (such as bill paying, flight booking and cancellation and fraud alerts).

But our platform is flexible enough to also provide an application layer that serves a need not included in the prepackaged solution. After brainstorming with a client, we can usually implement a solution for unique needs in two to three months.

Cost and complexity issues

The “non-smart” alternative solutions that are the status quo today are point solutions on a per-channel basis. Enterprises can build capabilities in house or buy analytic or business intelligence tools that help improve their understanding of a customer. But those tools are not real time and, by and large, they’re not designed for customer service operations or for sales operations. For online services, there are chat software and other point solutions. But they don’t connect to mobile, and they don’t have any prediction or real-time decisioning capabilities.

Addressing customer expectations today now requires more complex components than in the past. It’s too expensive for large enterprises to try to connect all the dots and build that infrastructure in house. And cobbling disparate components together on the back end is time-consuming. The [24]7 solution is in the cloud and can be deployed in four to six weeks with no CAPEX, and it’s a performance and outcome-based model.

The onus is on the enterprises and customer service providers to provide better, smarter solutions that meet customer expectations in helping them solve issues and that result in customer satisfaction. By improving the customer service perspectives for a digital world, enterprises will greatly improve their ability to sell more to consumers.

Kathy Juve has been with [24]7 since 2005, and served as chief sales officer before assuming her current role of chief marketing officer in 2011. Kathy has more than 25 years’ experience in the technology and customer service industries. Before [24]7, she was executive vice president of marketing and sales with Creditek Corporation. Earlier, she was vice president, Technology and Telecommunications, with Convergys Corporation. For more information, contact us at or

Copy link
Powered by Social Snap