How to put your machine data to work

Businesses are amassing terabytes of IoT data from their appliances and machines. CSRs cannot find the data they need. What’s the connection?


As consumer goods become increasingly commoditized and compete on features and pricing, many businesses are focusing their attention on delivering excellent customer service to build customer loyalty. For many businesses, this will require a reinvention of their approach to customer service. It can no longer be just a “support” function—it needs to evolve into a more proactive role in which the contact center agent is equipped with intelligent tools capable of actively diagnosing and fixing appliances and machines that are increasingly sophisticated and complex.

To make this possible, new systems and processes, employee training modules, and feedback systems will need to be incorporated into existing customer service solutions. But, also, it will require businesses to rethink data how they manage and process the information that powers their customer engagement channels. Fortunately, many businesses are already sitting on the data they need to accomplish this—they just don’t know how to apply it. I’ll explain how to do this, but first let’s examine where we’re going wrong today.

A Classic Day at a Call Center

When customers call a help line, they expect fast and accurate answers from the customer service representative (CSR). They don’t like to hear music during a long wait for a CSR to become available. And when they get put through, they expect their problem to be quickly diagnosed and fixed. However, in the case of complex home appliances and commercial equipment, so too does the information needed to identify the problem.

If a CSR is provided with excellent diagnostic tools and efficient access to relevant information, there’s a good chance the customer’s problem will be solved. Unfortunately, CSRs are not necessarily equipped with what they need. A recent survey by eGain asked CSRs to identify their biggest pain points while providing service when the customer is on the line. It found that agents are wrestling with several issues:

  • Finding the right answers to customer questions: 26%
  • Different systems/information sources give different answers: 25%
  • Hopping from one application/window to another: 20%
  • Hard to keep up with all the new information/changes I need to know about: 14%

What’s clear is that many CSRs are struggling with different types of information from disparate data sources stored on multiple databases. Accessing and then interpreting this information during a customer call lengthens the call time and potentially results in an even more frustrated customer.

How Many Databases Does It Take to Service One Customer?

The average CSR refers to or accesses numerous different types of data sources during a customer interaction. Here are some examples:

  • CRM SYSTEM contains personal details of customers such as name, registered number, email address, etc.
  • EQUIPMENT DATABASE contains details about the models, serial numbers and warranty details of every machine.
  • CASE-BASED REASONING TOOL (CBR)—an FAQ-styled database of common problems
  • for a variety of machines.
  • STANDARD EXCEL SPREADSHEETS are often used to record a problem described by a customer during the call.

What happens during a customer interaction can make or break a customer’s trust in a brand. The following is a typical interaction:

  • The CSR asks the customer to provide their name and telephone number.
  • CSR confirms the details in the (CRM) database.
  • If the customer details are not in the base, the CSR creates a new record for the customer.
  • The CSR asks the consumer to describe the problem.
  • If it’s a known issue, the call center agent may be able to use an FAQ database to find a quick DIY solution for the customer.
  • If the CSR is able to provide a suitable solution to the customer during the call, the agent has “saved the call” and does not need to schedule service.
  • However, if the problem cannot be solved over the phone, the CSR enters the problem description in the Excel spreadsheet, and adds details to the “notes” field of the dispatch software.
  • Then, the CSR schedules a service call for the customer.

Are you starting to see the problem? There’s a lot of data being transacted, and it’s stored in disparate locations. In addition to the CRM data, other kinds of data also are relevant for helping to solve a customer’s problem. Data that gets generated through various business cycles is unstructured, raw and in its native format. Often, this data resides in dozens of silos scattered across organizations. It includes customer data, engineering data, manufacturing data and field service data. And, the problem is about to get even more complex. New types of data are now flooding in—from the machines themselves! This data is being generated by IoT equipped appliances in both home and industrial settings. These machines are “phoning home” with all kinds of useful information gathered by their onboard sensors.

It’s Time to Orchestrate

Your Data To derive meaningful insights and drive relevant actions from this data, it must be organized, structured and made accessible—all on a single platform. A data orchestration platform can help organizations perform analytics on these new sources of data, derive useful insights from this data, and then use it to power various customer-facing service and support applications.

There are several ways this can immediately benefit organizations. Think how these use-cases might apply to your organization: predictive maintenance, remote asset monitoring, rapid diagnostics, customer support, field service, product-usage-based design and parts inventory optimization.

The Path from Reactive to Proactive to Predictive

Most organizations today are focused on delivering excellent reactive service—essentially what I outlined earlier in this article. By first orchestrating our data and then applying advanced analytics and the power of AI, we can deliver a range of powerful service-oriented applications for our customers, including predictive support, predictive service, predictive monitoring, intelligent chatbots, and knowledge-infused help for both customer self-service and agent dashboards. This data management problem for all service-driven organizations will only intensify. How will your organization organize this valuable resource and apply it to solving tomorrow’s customer service challenges?