Measuring the Results of CRM

As the initial wild enthusiasm for Customer Relationship Management begins to plateau, and companies are increasingly skeptical of inflated claims for success, the time is now more than ripe for a hard look at how CRM projects should be measured. What is success in CRM, anyway? How will you know it when you see it? When will your millions of dollars in CRM investments pay off?

Steve Diorio, president of IMT Strategies, recently interviewed CRM heads at 50 companies and was astounded at how few of them had any metrics in mind at all. “Ninety percent of them had no ROI model in place,” he notes. “They just took a leap of faith that they needed CRM.”

For companies already involved in CRM, you might say the barn door on success measures is open, and the horses are long gone. The right time to ask these questions is at the beginning of a CRM project. In fact, the best approach to defining the right metrics is to identify the benefits your company seeks from CRM. Out of those benefits will naturally flow the right strategic and tactical measures to keep you on track.

For example, new CRM software from Pivotal was installed nearly three years ago at Wind2 Software, a project-management software company in Fort Collins, Colorado. Dave Marvin, the Wind2 CEO, had as his primary objective reducing response time in his support center. But he also sought to capture the correct times incurred in fee-based tech support calls, a major revenue opportunity for the company. And finally, he hoped to root out any hidden customer dissatisfaction, by instituting tickler files to remind salespeople to stay in touch with their customers. Each of these objectives has a neat metric associated with it: a drop in inquiry response times, a customer service revenue increase of 12%, and a sizable reduction in customer attrition.

If solid metrics did not enter your early planning for CRM, it’s still better to define them later than not at all. You can step back at any time, and consider what you want from the new tools and new approaches. Most companies are defining their metrics in some combination of three arenas: cost savings, revenue enhancements or customer value improvements.

Cost savings
Pitching senior management on cost savings is a much easier sell than talking revenue increases. Whether it’s because they’ve been burned before, or because they’re suspicious of assigning any causality to revenue increases, C-level executives are often uncomfortable with hockey-stick revenue projections. The promise of operational efficiencies, on the other hand, is more tangible. Managers can actually be held accountable for cost savings.

If CRM is viewed across all marketing, sales and customer service functions, cost saving opportunities can be represented in a broad range of metrics:

  • Campaign expense reduction. When prospective targets are selected more effectively, campaign volumes decline, and expense-to-revenue ratios fall.
  • Internal process improvements. The percentage of customer orders that can be handled automatically, shorter sales cycle time, less rework–all these efficiencies go right to the bottom line.
    Service center efficiencies. When the duration of a transaction or a problem resolution call is shorter, that translates directly into lower headcount requirements.

Metropolitan Life Insurance is one company that placed no bets on topline growth when planning the company’s CRM efforts. “Revenue is too squishy to justify the investment,” says Allen Harris, Jr., the former VP of Tech Services at MetLife, who supervised the installation of Chordiant as an enterprise-wide CRM platform. “We focused on hard operational goals like consolidating service centers and eliminating redundancy. We were looking for permanent margin improvements.”

Harris’s team identified very specific areas where costs would disappear. “We kept track of how problems are resolved in our call centers. How much time resolution takes, and how many people are involved. If we saw an increase in the number of problems resolved at a single point of customer contact, we knew we were going in the right direction.”

Bob Miller, VP of Information Technology at Shop at Home, Inc., used a similar approach. “Over the two years since we installed Retek CRM in October 1999, we looked at the number of customer service calls as a percentage of all incoming calls, and saw a drop. We also experienced a welcome decline in the raw number of customer service issues raised, and an increase in those that could be resolved in one call.”

Revenue enhancement
Even if they sell their CRM projects based on cost reduction, most marketers are eagerly anticipate that revenue increases will follow their new efforts. Mike McIntyre, SVP at Hunter Business Group, a Minneapolis-based consultancy that specializes in business-to-business, puts it this way: “Businesses will demand either cost reduction measures or revenue increase measures. Automating marketing processes is usually about efficiency, meaning cost savings–and this is what finance people are more likely to believe. But increased revenue is where the greatest long-term value lies.”

Revenue increases from CRM can flow from a number of sources:

  • Campaign response improvements–more orders, more new customers.
  • Increased size of the average order size, or the number of products purchased in an order.
  • Competitive wins, like an increase in share of the customer’s wallet.
  • Cross selling and up-selling opportunities during sales and service contacts
  • Customer profitability–migrating customer groups from lower to higher value levels.

MetLife sold the CRM project to management based on hard cost savings, but they also recognized that the solution would bring additional benefits. “We were interested in satisfaction and lifetime value, but our primary measures were around reductions in the unit cost of a transaction in the call center,” said Allen Harris. “That the platform also offered campaign enhancements we considered a freebie.”

One reason revenue enhancements are problematic as CRM measures is that they can take so long to show up. “Revenue is dependent on sales cycles, the size of the organization, and the amount of behavioral change needed in the company,” notes John Eben, VP Corporate Services at Hunter. If companies are looking for quick hits, they may do better to hang their hats on cost savings measures.

The Dallas-based Credit Union of Texas, with nearly $1 billion in assets, comes in firmly on the revenue side of the argument. “We look at internal operational enhancements, but most of our focus is external, on new members, sales and marketing. We have already demonstrated some excellent campaign results from our new business intelligence systems,” notes Jerry Thompson, SVP and CIO.

The credit union’s first campaign objective was to raise the profitability of the 44,000 low-value members the bank was serving. These accounts carried low balances and subscribed to only one bank offering. With Unica’s Affinium Model, Thompson identified which of the institution’s 109 services were most likely to interest each low-value member. A personalized direct mail campaign, with three different offerings to each customer, dropped in April, and garnered a whopping 10.4% response rate. Of responders, 2272 customers signed up for additional services, resulting in $150,000 additional net profit for 2001 alone. Another 3600 customers decided to close their accounts, which saved the bank $73,000 in servicing costs. The campaign itself cost only $45,000, and the credit union plans follow-up mailings to offer a different set of services options to the non-responders, or the chance to close their accounts altogether. “You want numbers?” exalts Thompson. “This campaign’s net present value was over $335,000.”

Brad Snook is VP of Client Relationship Management at United Asset Coverage, which offers consolidated management of equipment maintenance contracts. Because his business is almost entirely dependent on renewals for its profits and growth, Snook justified his CRM project by expected revenue enhancements. “A mere 1% shift in renewal rates more than pays for our investment in the Oracle 11i suite. A small increase in our retention rate even outweighs our entire revenue from new customer acquisition.”

Andy Frawley, founder and chairman of Xchange, an analytical CRM solution provider, also believes that revenue metrics are the way to go. “The things you measure should be economic, not operational. Ultimately, profit is the number one metric.”

Activity measures
As cost measures and revenue measures can both present problems, some companies throw in the towel and settle for metrics based on sales and marketing activity. These can be metrics about the CRM project itself, or about specific elements of the sales, marketing and customer service process. “It can be tough to rely on results measures when the sales cycle is long,” notes Hunter’s Mike McIntyre. The Hunter group recommends such interim metrics as account penetration, number of inquiries, number of qualified leads, or the percent of completely populated profiles in the database. In the consumer world, you might use sales cycle time, the turnaround time on reporting, or the percentage of address updates gathered from NCOA.

In the extreme, activity-based metrics can be as simple as project management milestones. Notes IMT Strategies’s Diorio, “When I asked CRM heads about their success to date, many of them said something like, ‘Yes, it’s successful. I got the software installed, it didn’t crash, and I didn’t lose my job.’ “

Customer value improvements
By far, the best metrics revolve around customers themselves. Judy Kincaid, author of the forthcoming book from Prentice-Hall, Customer Relationship Management: Getting It Right, identifies the three important customer measures to track.

  1. Lifetime value (LTV). The single most effective measure, calculated as the revenues the customer will generate over the life of his relationship, net of the variable cost of achieving those sales–the cost of goods sold, and the cost to serve.
  2. Customer satisfaction. This can be measured by observing behavior, but is more likely to be captured by surveys. Satisfaction should be evaluated around the customer’s relationship with the entire brand, versus any single product or function.
  3. Customer attrition. Definitions will vary by industry, but loosely this means someone who has not purchased over a certain period. It might be five years in the aircraft business, or 18 months in office supplies. The company’s attrition, or churn, rates should then be compared to industry benchmarks. Attrition is best analyzed by segment, since customer value can vary widely.

But gathering and making use of these customer metrics is not easy, especially for companies that distribute through retail or other third parties. In such cases, says Kincaid, “You use common sense. Allow yourself to recognize that metrics aren’t absolutely perfect. You need to use approximations. As long as you do something, and repeat it, you can measure trends, which is a big step in the right direction.”

According to Ro King, EVP, Client Solutions, at the CRM consulting company Quaero, every CRM company should have customer measures in place. King looks for indicators like the breadth, depth and length of the customer’s relationship with the company, as well as the number of channels the customer uses. Each measurement category can be supported by a variety of directional proxies, such as the number of product types purchased or the number of dollars spent. “The point is to move from product measures to customer measures,” says King. “CRM forces a change in the conversation. And the change may require three years to take root.”

Customer measures, while quantifiable, can also be squishy. “A relationship entails emotions,” says Jim Barnes, author of Secrets of Customer Relationship Management and EVP at the Canadian consultancy, Bristol Group. “We need to understand the emotional connection as well. We need to ask why, and get at the true customer viewpoint. Beyond functional loyalty lives emotional loyalty, and this should be the goal of CRM.” Barnes recommends a 30-minute questionnaire methodology, which is then tied back to customer data to validate the impact that the emotional points have on spending and profit levels.

The myth of ROI
As companies struggle with short-term efficiency measures and longer-term revenue and customer measures, one thing is clear: the average CRM project is very tough to justify on a classic ROI basis. For many, the “leap of faith” identified by Steve Diorio is the only solution. It may be that moving to a customer-centric business, and installing tools and techniques to support it, is simply the meets-minimum requirement of doing business in today’s competitive environment.

“ROI is a slippery slope,” says Hunter Business Group’s John Eben. “CRM is becoming a core component of sales and marketing. It is critical piece of the overall value proposition. Customers expect it. The mistake is to isolate CRM and seek an ROI on it alone.”

According to Jon Diorio, who carries the very classy title of ROI Evangelist at E.piphany, a leading CRM software provider, the biggest ROI hurdle comes when companies approach CRM measures from the bottom up, instead of top down. “Looking at call center metrics is only going to confuse the issue,” he says. “A better approach is to set a strategic goal, evaluate what benefits you expect from that goal, and then choose the specific data points that will indicate whether those benefits are accruing.” For example, a strategic goal of improving the customer’s experience would likely result in the benefit of increased loyalty, which in turn could be measured by tangible metrics like higher repeat purchases and reduced customer attrition.

Measurement benchmarks
The jury is still out on the best benchmarks to use for comparative CRM results. Three options are available, each with its own pros and cons.

  1. Control groups. Some software solutions allow control groups to be established and tracked, tactic by tactic. Control groups provide powerful proof of results, but many companies hesitate to fence off any customers and reduce their access to optimal treatment.
  2. Before and after. Tracking results over time and measuring results against one’s own self is probably the best solution for the risk averse. The problem is that you can’t be sure what variable actually caused the change in performance.
  3. Industry benchmarking. An ideal approach, one that keeps you both honest and competitive. One caveat: industry stats may be hard to come by.

At some point, companies must agree that CRM is a business philosophy more than a set of processes, or a set of software tools. According to Bob Thompson, president of Frontline Solutions, and founder of CRMguru.com, “In some cases, the CRM benefit simply is not measurable. The mandate of CRM is to take care of the customer. But often, the value is realized mainly in the corporate culture. Sometimes, the most important thing is the struggle.”

Qualitative Measures of CRM Readiness
Quaero, the CRM consultancy, asks its clients to evaluate themselves against a 5-point checklist of readiness indicators. Companies that score high on readiness are more likely to succeed in their CRM projects.

  1. How many cross-functional teams operate in the company.
  2. How frequent is cooperation across divisional borders.
  3. Name the ways that IT and Marketing work together.
  4. What incentives are in place that might conflict with a customer-centric view (e.g., sales commissions based on product categories)
  5. How is customer data captured, accessed and used.

Why Is It So Hard to Measure CRM?
A number of factors can derail CRM measurements.

  • The company is not yet ready to migrate to a customer-centric strategy.
  • The business is fragmented, with no communications among departmental silos.
  • A single view of the customer proves impossible to assemble.
  • CRM is isolated, relegated to a certain function, like IT.
  • Revenue measures are suspect to senior finance people.
  • Expectations for short-term results overwhelm management patience for longer term benefits.

Rules for Successful CRM Measurements

  • Set strategic and tactical goals up front.
  • Get senior management agreement on the goals and benefits.
  • Limit tracking to a few key metrics: “Keep it simple.”
  • Select metrics that are big enough to matter to the business.
  • Don’t necessarily call it CRM. Call it whatever works for the culture.
  • View CRM as a change in culture and process, not just automating existing processes.
  • Set milestones for 3, 6 and 12 months out.
  • Tie employee and business partner compensation to customer metrics.
  • Don’t evaluate CRM in a vacuum.
  • Do the IT part last.

Back to Articles and Columns »