Active value curves illustrate the impact of response times, Ohio State research shows.

Features

Feature

Learning Curves

Active value curves illustrate the impact of response times, Ohio State research shows.

Time is money.

That truism is behind operational business intelligence (BI)—the idea of connecting front-line workers, and even customers, directly to an active data warehouse. In theory, providing faster access to up-to-date status, product offers and service remedies creates value for your business.

But does operational BI really pay off? Does the speed of interactions truly matter?

The answer is a resounding yes, according to Teradata-sponsored student research at the Initiative for Managed Services at The Ohio State University Fisher College of Business. Conducted in early 2010, the investigation focused on the value of speed, with a particular focus on service remedies.

During the customer relationship life cycle, each interaction with a customer creates or destroys actual or potential value. This is known as a “customer cardiogram,” where a sequence of positive and negative events has financial consequences for the business. For every individual, value begins at a negative point from the company’s perspective because of the cost of acquiring the customer, e.g., marketing costs. Value rises as a person purchases products or services and decreases whenever a failure occurs.

No Time to Waste

The research focused on service failures because of their potential for damage. For example, if an airline misconnects a passenger, what is the impact of an immediate apology versus one an hour later? How about one provided a day later or never?

By measuring the drop-off in value, over time, researchers developed active value curves to illustrate the potential impact of responding—or failing to respond—quickly to problems. Here are two examples—one relating to social media and the other to energy consumption—that demonstrate the relevance of differentiated response reaction times:

Figure 1: Tweet Customer Feedback

Click to enlarge


TWITTER TRACKING

Most organizations struggle to understand and manage social media. Often, they merely monitor and don’t try to shape the content actively. But what if a company could use social media to its advantage through public data? Suppose after receiving poor service from a rude waiter at a restaurant, an unhappy customer tweets about it immediately, using a hand-held device. (See figure 1.) Subsequently, the event could cost the company not only the future business of that person but also that of those who follow his or her recommendations on Twitter. Without a response, this would result in a negative impact on sales.

With 1,000 regular and 200 non-regular customers each week, an average check of $20, and assuming a 1% readership and influencer impact for regular customers and a 3% impact on new customers, this restaurant loses 10 regular and 6 non-regular customers and $320 in revenues in just the first week because of this tweet.

An “active” company could provide counter-evidence on Twitter, salvaging some value (and face), or engage in promotions in other social media outlets to compensate. Even better, it could encourage supporters to promote the business or rebut the negative comment, hopefully, within 24 hours. At the same time, with active technology monitoring social media, management can spot and correct such trends. In this case, improved training of new waitstaff could prevent future problems.

Figure 2: Home-Based Power Management

Click to enlarge


ENERGY MANAGEMENT

Smart energy management systems are expected to grow rapidly in the coming years. Demand management of electricity usage has the potential to save 5% during peak hours, potentially $3 billion annually in the US alone.

Such systems use sensors at customer locations to feed centralized data warehouses and processing systems, allowing utility companies to monitor and regulate energy consumption within homes and businesses. The purpose is to equalize overall demand, which reduces the need for new power plants to meet peak energy requirements. If power consumption is not managed properly, in time, spikes in demand can lead to brownouts or even blackouts for whole regions. (See figure 2.)

Traditionally, it’s difficult to track and forecast power usage in homes and offices. Energy data is not well integrated and problems cannot be mitigated in time to avert large-scale losses.

In the future, tracking demand during peak periods will lead to BI and analytics applications that instantaneously collate information from different locations together with power availability, usage and interruptions to drive small corrective actions. This will eliminate failures by ensuring that the grid is not overburdened.

Fast, automated decision making in a matter of a few minutes or seconds is the key to averting catastrophic declines in power supply. Eliminating costly interruptions and repair work results in cost savings and value to customers and utilities.

The Tip of the Iceberg

These are only two of the active value curve examples developed. Others focused on a variety of industries:

  • Financial—stock trading
  • Healthcare—drug trials, home healthcare
  • Travel—flight rescheduling, traffic grid management, agile rebooking during emergencies, revenue assurance
  • Retail—customer returns, geospatial marketing, stockout reductions, product recalls
  • Insurance—timely renewals, reactions to customer policy cancellations
  • Hospitality—hotel occupancy maximization

More work is yet to be done. While the research created the framework and can illustrate the value of going active, the next step is to validate these examples to gather the real return on investment (ROI) of better, faster actions on the front line.

Doing so will go beyond proving that time is money and move toward calculating time’s true value—perhaps to the day, hour or even minute.


Your Comment:
  
Your Rating:

Comments
 
It is good to see the article and findings are quite interesting.I am more interested to know what method has been followed as part of Ohio State research.

12/20/2010 2:12:02 AM
— Anonymous
EMC Q3-2014