How Process Analytics Can Improve the Customer Experience

Untitled Document

Few would argue that the customer is king. However, these kings are even more powerful as providers focus on market share to ride out the recession, then ARPU gains to achieve better-than-average market growth in the recovery period.

Throughout this transition, providers are making strategic investments to improve the customer experience and therefore better position themselves to achieve short-term and long-term market and financial performance gains. These gains include reducing operational costs and adding customers in near-saturated markets, avoiding churn (particularly for high customer lifetime value customers), accelerating penetration across product lines, driving increased revenue per customer and customer lifetime value.

In an attempt to support these goals, providers have focused on two areas of investment: business intelligence software to provide stakeholders and executives analyses, dashboards, and reports that show the baseline realities and subsequent progress on customer KPIs, and predictive analytics that use complex algorithms to predict customer behavior to help guide marketing and other programs. Both of these investments have proven their value over time and can provide a meaningful analytic backbone.

However, these capabilities miss a critical factor in what truly shapes the customer experience and actions in the first place. Organizations must first understand how customers transact and traverse across the large, complex order-to-cash process, from marketing programs through to billing, customer care and field service. This is where the customer lives, and where the strength -- or lack of strength -- of the provider's entire order-to-cash process can either enhance or erode the customer experience.

Process analytics, in this case customer-based process analytics, deliver transaction-by-transaction, root-cause based analytics to determine if the sub-processes and systems that comprise the order-to-cash process are operating optimally and whether they are supporting or inhibiting the overarching customer experience goals. They can also provide a more complete view of the implications and value of upstream marketing programs.

More specifically, customer-based process analytics focus on three core sets of issues:

1. Errors, delays or confusion in the order-to-cash process (e.g., order-to-activation delays, billing errors, or unclear marketing programs) that can frustrate customers and cause a negative impact to customer experience.

2. Avoidable calls to the call center and sub-optimal call handling processes that can unnecessarily drive up costs while further frustrating customers.

3. Sub-optimal revenue management in terms of errors that delay revenue and/or collections, call handling issues prompting excessive customer credits, and marketing programs that negatively impact margins by driving up call volumes without the associated ARPU gains.

For example, plenty of attention is given to customer service center benchmarks, such as time-to-answer and time-on-call, as they are proven determinants of operational costs and proxies for measuring customer satisfaction. But why are these customers calling in the first place?

Beyond the desirable reasons (adding services), is it also for avoidable reasons such as billing disputes or complaints for late or incorrect services? The calls could be related to marketing inquiries, because customers cannot evaluate a new program or their eligibility.

Or, are the follow-up calls because the first call didn't resolve or perhaps even compounded their initial issue increased their frustration? So while managing service level agreements (SLAs) are vital and good, identifying and attacking the avoidable call volumes and the sources of process inefficiency are what will ultimately enable providers to reduce costs, reduce churn, and enhance the customer experience.

Clearly, there is a powerful business case for conducting customer-based process analytics in complement to existing BI and predictive analytic investments, so why has investment in this area paled in comparison and lagged in terms of time? There are probably several reasons.

First, until recently, the ability to capture data and the associated business logic has been relegated to a few very specialized areas of analysis, such as revenue assurance and fraud detection. Secondly, for some providers, solving the challenge to acquire, enrich, correlate, and analyze the customer data across all of the operational sources is simply too challenging.

Some providers -- for many good reasons -- will seek to leverage prior efforts that relied on enterprise data warehouses to create a portfolio of customer analytics. Unfortunately, the "leverage" opportunity is limited as BI and predictive solutions were not designed for this type of analytics.

As such, the provider can report on results that are delivered through formal BI queries, or understand patterns and likely outcomes or behaviors from the reported results. But they still cannot effectively determine root cause, drill down to atomic-level data, and systematically detect and eliminate the defects causing issues.

Nor will they have the flexibly to tap into the different operational systems as they learn about their business. On top of these limitations, the leverage opportunity is hampered by the inherent waterfall development process that typically requires 6-9 months to move from curiosity or need to a workable query that can produce initial results.

So what is possible? Process analytics enables providers to acquire and analyze data and the associated process logic in the relevant order-to-cash systems and sources. It also establishes and analyzes causal relationships by understanding how the customer transacts and traverses across the process, and identifies atomic-level defects or find new relationships among the data to enable providers to attack root cause issues that impact the customer experience.

For example, providers investigating customer complaints will likely encounter a practical reality: the complaint data is only as good as the customer service representative inputting the data and often is different than the operational reality. To understand and address complaints, the actual operational data is the only reliable "system" of record and is uniquely able to show what is actually happening to the customer (e.g., errors in the order, system-driven delays to activation, provisioning errors, multiple on-site visits, etc). Without process analytics, the provider can determine that something is happening, but not know why it is happening or how to resolve it.

One example clearly showcases the power and focus of customer-based process analytics. As part of a critical product launch that was central to their Triple Play strategy, a leading North American telecommunications provider was experiencing a significantly faster rate of growth in truck rolls and on-site times than actual customer adds.

This was creating immediate risk to customer acquisition and churn in their business, as well as risking the full portfolio of services offered to those customers. It was also significantly driving up call volumes and overall cost. The provider attempted unsuccessfully, using existing tools and manual efforts, to correlate and analyze the range of data and rules that resided in order management, billing, CRM, dispatch, and trouble management systems, among others.

The operational complexity outstripped the capabilities of the tools, which could not help identify root cause or enable a continuous improvement program. The customer then applied customer-based process analytics and was able to rapidly identify and correct the systemic and individual issues related to products, delivery models, technical skills, and customer segments that were causing excessive truck rolls, lengthy time on-site and diminished customer satisfaction.

Winning the customer battle secures revenue in today's economic climate and is a central lever to driving higher-than-market growth as economies recover. Providers can quickly and cost-effectively add process analytics to identify and resolve defects before they erode the customer experience and impact market share or customer lifetime value.

More to the point, the absence of these types of analytics leaves providers unnecessarily vulnerable. Process analytics leverages and complements existing customer service and analytic investments and provides significant operational visibility, customer value, and financial return.

About the Author

Victor Milligan is Chief Strategy Officer at Martin Dawes Analytics, a leading global process analytics software provider. For more information, contact the author at or visit

More by Victor Milligan