1. Insights
  2. Digital Experience
  3. Article
  • Share on X
  • Share on Facebook
  • Share via email

Four steps for successfully adding data analytics into your customer service strategy

Posted January 16, 2018
A man and a woman reading a book in an office library

Tea leaves, crystal balls, tarot cards and more—for centuries people have sought the ability to predict the future. While entertaining, these methods have yet to be proven reliable or fruitful. Fortunately, today’s business leaders have data analytics to inform their decisions and suggest actions for optimal outcomes.

While data analytics holds the potential to transform operational effectiveness and deliver improvements in customer service, its implementation can be challenging. In a recent McKinsey survey, 85 percent of executives admitted their data analytics goals were not being met. Moreover, Gartner predicted that 60 percent of big data projects would fail in 2017.

Based on our research and experience we have found that a methodical approach, incorporating the following four steps, is essential to an effective data analytics implementation.

Step one: Identify the right opportunities

Business process outsourcing is ripe with opportunity for data analytics. In fact, the sheer amount of data available can be overwhelming, making it difficult to identify exactly where to start.

Everyone knows the trick of using a magnifying glass to focus sunlight on an object to make it burn. For a data analytics program to catch fire in your organization it needs to be similarly focused on the right opportunities.

The following list of criteria, which we refer to as the Data Analytics Opportunity Finder, can help identify where data analytics can be the most effective in your organization:

  • People: Take note of functions requiring a large pool of people. Higher numbers of people tend to indicate rich data.
  • Knowledge: Look for areas within your organization where there is an abundance of expertise; where knowledge management and decision making will benefit from enhanced speed and efficiency.
  • Digitization: Deploy technology to collect data in new ways through digitization. For example, using speech-to-text technology, valuable and data rich conversations with customers can be digitized, leading to additional actionable analytic insights capable of improving the customer experience.
  • Market demand: Leverage forecasting data, such as specific market supply/demand and trends, to provide accurate models to aid in decision-making.

For TELUS International, one of our earliest data and analytics projects was focused on employee attrition—meeting several of the Data Analytics Opportunity Finder criteria. We wanted to do a better job of retaining long-term high-performers, knowing that the overall increased tenure of our global team would reduce our costs and improve our customer service delivery.

Step two: Develop the data and analytics strategy

Using a rubric like the Data Analytics Opportunity Finder helps identify problems you want to solve, which leads to goals, and ultimately a strategic vision. What do you need to learn to enable you to take action? Where are the points at which you need to intervene? What are the reliable predictors of behaviors you want to minimize or encourage?

When implementing our own attrition analytics project, our strategy was to embed an analytics component directly into our retention management process. The overall goal was to reduce the cost of employee churn by predicting who (especially top performers) was likely to leave the company, and to deploy the best retention efforts where they would be most impactful.

A strategic approach keeps your data analytics initiative focused on a business problem and helps you avoid a ‘tail-wagging-the-dog’ scenario. An effective data analytics strategy includes the following six components:

  • A mission: Determine how you will use analytics to extract value from data and transform its prescriptive insights into revenue growth, reduced operating expenses, better customer experiences or a combination of results.
  • Service and capabilities: Identify and outline the skill sets and services needed to deliver all components of the Data Analytics Value Chain (see step three).
  • Use cases: Develop processes for using analytics to deliver value.
  • Sourcing and location: Find available data sources and denote those that are required across the organization.
  • Governance: Create a cross-functional steering committee that can provide access to resources across the organization and collaboratively drive the process design and execution of the overall data analysis strategy.
  • Data and technology: Design and implement the infrastructure needed to enable delivery on the use cases.

Step three: Follow the Data Analytics Value Chain

Gartner has defined four links in the Data Analytics Value Chain that move along a continuum from hindsight to foresight. As a company collects data, moving along the chain from descriptive to prescriptive, they gain greater insight and focus. The four analytic links are:

  • Descriptive analytics (What happened?)
  • Diagnostic analytics (Why did it happen?)
  • Predictive analytics (What will happen?)
  • Prescriptive analytics (How can we make it happen?)

To see how the data evolves along the chain, let’s go back to our example of solving attrition. In many organizations, data related to attrition is compiled manually, and analysis is backward-looking based on obsolete data (descriptive). When a risk of leaving is identified (diagnostic), the employee has already decided to go, making the organization a day late and an employee short.

The later links in the Data Analytics Value Chain allow organizations to learn which actions they can take to prevent employee loss (predictive), and ultimately determine the primary drivers of attrition and implement an automated solution (prescriptive).

Step four: Establish credibility with an early win

KPMG reports that less than half of firms are very confident in the insights derived from analytics and just over 20 percent trust the most transformative analytics technologies, including predictive modeling, machine learning and artificial intelligence (AI)-assisted cognitive development. This is shockingly low, particularly when compared with a Gallop poll that found 26 percent of Americans believe in the power of clairvoyance!

If you are finding it difficult to convince others to put down the tea leaves in favor of data analytics, demonstrating an early win can help. For TELUS International, our attrition analytics project proved incredibly successful and our business continues to have one of the lowest global attrition rates in the industry.

The promise of data analytics to transform operations, customer experience and the bottom line is real, but organizations must take carefully calibrated steps if they are to achieve success. With digitization speeding globalization, rapid developments in robotics, AI and machine learning, and an evolving labor market, companies must learn to quickly and correctly integrate data analytics into their core business processes to remain competitive. Following these four steps will put organizations well on their way to harnessing the full transformative power of data analytics.

Interested in speaking to an analytics solution leader? Connect with us.


Check out our solutions

Harness the power of Big Data and analytics to make informed business decisions.

Learn more