Peter Drucker, legendary management consultant, once said: "What's measured, improves." Drucker espoused the benefits of metrics decades ago, yet in 2011, few companies are experts at defining and implementing metrics. What companies have instead is a preponderance of data. Over the years, some have said that companies and workers suffer from "information overload." A critical problem companies face is gaining actionable insight from all of this data.
Most companies have several different reporting systems and productivity tools to collect and display data related to sales, customer activity, marketing campaigns, financial performance and market trends. Some even have handy collaboration sites where employees can post charts and tables for others to view and edit. Yet all these spreadsheets, files and tools can lead to inertia. No one knows where to start, or how to make decisions from all these reports. Sadly, the finest of reports do not naturally lead to insight and action. There may not be an obvious "next step” for the business user. This gap is one reason why many BI projects to fail.
Closing the gap means moving away from data toward metrics. Metrics are actionable measures associated with core business processes, such as: on-time delivery for a manufacturing company, employee attrition for HR, and “time-to-conversion” for online marketers.
By focusing on the metrics that matter most to the performance and viability of a particular department, organizations can gain more value from the data they collect. Studies indicate that the timely, effective use of sales metrics by midsized companies can result in increased revenue and a higher sales close rate.
Below, we share how two companies have benefited from a metrics approach, and discuss the critical success factors in developing and deploying metrics. Many businesses are still in the early stages of developing a metrics strategy. Some, however, have made credible progress toward building a metrics-driven organization.
Timbuk2: Inventory metrics
Retail analytics is a hot area, and companies such as Timbuk2 are gaining quantifiable value through a sharp focus on metrics. Timbuk2 was founded in 1989 by a San Francisco bicycle messenger who wanted to create a bag rugged enough for professional messengers yet stylish enough for others simply craving a backpack alternative. The idea took off, and by 2006, the company was manufacturing several styles of messenger bags, travel bags, backpacks, laptop bags and accessories including made-to-order custom bags.
As the company grew, an emerging challenge was determining inventory availability. Whether an order was placed by a customer directly at the website or through a distributor, Timbuk2 needed confidence of filling the order on time. The company’s manual process of compiling and merging Excel spreadsheets, with no internal IT support, was making it tough to accurately plan and forecast inventory in a timely manner.
After adopting a SaaS-based BI system that integrates data from the company’s ERP system to produce real-time data through a Web portal, the company had the means to do faster and deeper analytics. Now the company has a highly accurate picture of inventory out to 120 days. Using the Web portal, sales reps and retailers can see exactly how much product will be available on a certain date before placing an order; precise forecasting has helped Timbuk2 double its reorder rate and improve customer service.
They have also added key metrics to help with the analysis. The first metric the company tracked was “on-time fill rate.” The business goal behind the metric was to avoid late shipments and incomplete orders. The company measures this metric on a monthly basis and has refined the metric over time to improve processes. Timbuk2 is now measuring several metrics related to sales performance and supply-chain processes.
It’s easy to get excited about metrics and jump in feet first by tracking and measuring a bunch of things at once. But if you take the approach of Timbuk2, which began with a few metrics that are meaningful to its business, you'll have better success. The roadmap below begins with a few definitions.
1. Metrics versus KPIs
Hurwitz & Associates defines a business metric as a measurement that is quantifiable and relates to a business activity. When properly defined, metrics should relate to each other and provide a framework for analysis.
Many companies also talk about KPIs, high-level metrics such as customer satisfaction, revenue growth and profitability. KPIs typically rely upon smaller metrics that relate to business processes. For instance, if an e-commerce company has a KPI of "Percentage of Repeat Customers," it will need to measure metrics related to customer complaints, returns, website performance, and shipping performance to understand the KPI measure. It's ideal if KPIs come from the top down, reflecting company -wide goals. But more typically, KPIs funnel up from the ground level; sometimes the act of measuring something in one department bubbles it up to the top.
2. Create a team
A team of individuals who are committed to developing and refining the metrics gives structure and accountability to the process. Often, a metrics program starts from a group of employees working on common goals. From the start, the team should have executive sponsorship and guidance on connecting the metrics to companywide objectives.
3. Define goals
Corporate objectives are not always clear or shared across the employee base. This lack of insight into the big picture can slow down the process, yet it's not worth measuring something if it's not important to business strategy. Ideally, a metrics program will incorporate top-level goals as well as more specific departmental or functional goals. For instance, if an HR department initiates a metrics program, its goals might include improving employee performance and reducing attrition. Those goals then link into a corporate-level KPI such as revenue generated per employee.
4. Define metrics and related processes
Next, the team needs to define one or more metrics to track. Expect to refine and even replace those metrics as business conditions change. The team will also need to know how frequently users want to view the metrics, how they will access them, and how they will incorporate the analysis into their work.
5. Document data sources
Once you’ve selected the metrics, determine which information systems will fuel the needed data. A small company with no IT staff may need the help of a specialist to conduct this analysis.
The Role of Technology
Metrics-driven organizations must consistently and accurately collect, integrate and share data. There is really no way to do this other than through automation. While some BI tools help support a metrics-driven approach, many are primarily designed to produce reports. Yet the thought process around metrics is different from simply generating reports: It’s more high level and strategic in nature. Here are a few considerations for the optimal metrics-driven system:
- Ability to navigate metrics: A user should be able to define a broader or more granular view of the data, depending on his perspective. If there are multiple metrics that relate to a particular business process, a user can link those metrics together for a high level view. A simple way to offer flexible analysis is through dashboards, which can display several metrics in one screen with drill-down features.
- Ease-of-use: This may sound obvious but BI systems, including spreadsheets, are too often designed for analysts or technical employees. Modern BI systems can be as intuitive as filling out a form on a website. The interface should be clean, crisp, and provide an easy way to explore data and change views. Labels should be business-friendly and relevant for the people who will use the system. A user should be able to refine or add metrics without training.
- Accessible by partners: Web-based solutions with strong security protections can easily extend key metrics to third parties with a need to know. For example, if a distributor can see that the number of orders for a part has been consistently growing in a certain region over a three-month period, it can work with warehouse partners to introduce measures that will meet the rising demand. This kind of capability is now possible from a growing cadre of SaaS BI providers.
DMA: Supply-chain metrics
Distribution Market Advantage (DMA) is a marketing cooperative for the food service industry, which provides a national order and information exchange system between distributors and restaurant customers. DMA’s data comes from customer invoice transactions through multiple distribution warehouses. A few years ago, the organization adopted a cloud-based BI solution from provider PivotLink to support supply chain analytics for its customers, who needed to understand spend and product movement and verify prices of their manufacturer contracts.
When the system was implemented, DMA embarked on a metrics program as a way to organize the high volume of data for customers. This metrics view includes a KPI scorecard containing customer and distributor-focused metrics.
- Customer-centric metrics include deliveries, frequency per week, and extra deliveries.
- Distributor related metrics include mispicks, damaged cases, and total error rates.
Through the Web interface, customers can view metrics in multiple ways. For instance, they can see extra deliveries from distributors at the company level, warehouse level, or at a store serviced by a warehouse. This deep supply-chain visibility is helping DMA's customers make better business decisions and negotiate more effectively with manufacturers.
The Future of Metrics
A metrics-driven organization is better equipped to compete in today's global, fast-paced economy. Having BI systems, reports and processes is a solid foundation, but to go the full distance organizations need to incorporate metrics. That means having a clear process for defining and managing metrics and applying metrics-based intelligence to business decisions. Determining the right metrics to track and how to best use this data to improve performance and customer relationships is tricky at best. There is no “right” way to go about this, although there is an emerging set of best practices that managers in any industry can apply to their business.
Here are a few closing tips:
- When implementing a BI initiative, think about whether the system will be able to support key metrics.
- Understand the relationship between metrics and KPIs.
- Spend time defining and linking corporate and department-level goals to metrics.
- Ensure that your system enables multi-view analysis of metrics and drill-down capabilities to support the needs of many different job roles and scenarios.
- Allow simple and secure access to data and self-service capabilities so that users can do analysis and make independent decisions without burdening IT.
- Stay focused on business needs. In the case of DMA, the customer need was to drive down costs and inefficiencies in the supply chain.
- Take time to understand your data and how granular the information needs to be in order to be effective.
- Consider starting small: Timbuk2 focused on one metric to start, and then slowly added additional metrics over time as it perfected its process and realized benefits.
Building a metrics-driven organization takes time and persistence. It also requires knowledgeable partners, such as IT and business process experts, if your organization lacks those skills. It requires flexible technology that is easy to manage and share. If you take the long view of metrics, and keep it simple, your efforts could pay off sooner than later.
About the author
Dr. Fern Halper is a partner at Hurwitz & Associates, a consulting, research and analyst firm that focuses on the customer benefits derived when advanced and emerging software technologies are used to solve business problems. Fern has over 20 years of experience in data analysis, business analysis, and strategy development. She served as SVP for enterprise applications and services for Hurwitz Group and has held key positions at AT&T Bell Laboratories and Lucent Technologies. She is also a co-author of Cloud Computing for Dummies.