Whether it’s driving informed decisions, unearthing unexpected results, enabling effective continuous improvement, or empowering collaboration; real-time and accurate quality data should be the lifeblood of an effective quality management program. In fact, leveraging evidence to make decisions is a core quality principle which makes the current state of real-time quality metrics so surprising.
Click here to speak with Dan Jacob
It’s easy to see evidence-based decision making at work on the shop floor where inspectors and in-line monitors pass and fail components using empirical data. The shop floor is a stronghold for quality data; the result of a century of work developing math and techniques to improve conformance to standards. Data is available in fractions of a second to minutes, depending upon the level of automation. We find a different story when we include data from test labs, in-service maintenance, customer call centers, warranty data, supplier data, customer data, etc.
Much of this broader set of data is poorly connected, difficult to search, and slow to retrieve. Can you imagine waiting days or weeks to get the information needed to make decisions on the shop floor? These are reasons that 37% of 1,328 respondents state that their top roadblock to achieving quality objectives is the inability to effectively measure quality metrics.
Of course, generating data alone isn’t enough – the data and results must be timely and visible to those that need to make decisions. Let’s analyze LNS’s survey data to see how leaders compare to the rest of the market. The analysis shows that Innovation Leaders (Level 5 on LNS’s maturity model) have nearly three times the adoption rate of real-time quality metrics than the rest of the market. “All Others” comprise 80% of the market; 80% of the market have poor visibility to real-time metrics. For more discussion on Innovation Leaders, maturity, and related topics, download LNS’s Supplier Quality Management eBook here.
The 7 Reasons for an Executive to Insist Upon Accurate, Real-Time Quality Data.
1. Informed decision-making
Research shows that qualitative decisions are increasingly error prone as decisions become increasingly complex. It’s all too common to think that the problem du jour “always happens” or is “the most common problem”, or for a problem to become high profile because an important customer complains, or a problem surfaces to a top exec. While this happens, more frequent/costly problems are ignored. Data provides the perspective to balance these biases.
2. Unexpected results
Accurate data can confirm expectations, but in many cases provides new and interesting insights. Take the chart above as an example. Knowing that 37% of the market identifies poor metrics as a top quality challenge, it is no surprise that adoption rates of real-time metrics are low. However, the canyon separating leaders from the rest of the market is a little unexpected. It is surprising that real-time visibility of quality data in manufacturing is at the low end of the range. Why aren’t we sharing this source of rich real-time information with the broader enterprise or connecting data from the rest of the enterprise to manufacturing? Good data should challenge assumptions and provide new insights.
3. Metric Levers
Real-time metrics provide a method to quantify performance. By necessity, this means they are connected to the data that drives these metrics. While metrics quantify performance, their connection to process data exposes the levers that drive performance, which makes all the difference.
For instance, LNS has discussed Cost of Poor Quality (CoPQ) with many quality leaders. CoPQ is a valuable metric but often calculated manually and infrequently. Innovation Leaders know this metric 3.2x more often than Ad Hoc. The data is often compiled manually, which precludes the ability to understand levers, substantially reducing CoPQ’s utility as a decision-making tool.
4. Effective Continuous Improvement
Real-time metrics are critical to understanding the effectiveness of corrective actions and continuous improvement. Once a root cause is addressed, it should be possible to trace that event to an improvement in metrics and a decrease in root causes.
5. Culture of Quality
Quality strategy and execution impacts many operational and financial metrics which are valuable to the corporation. Communicating these metrics and data with other functions crystallizes the impact of quality on common goals and metrics. Once other functions begin to include quality metrics into their decision-making, it improves the perceived impact of quality management to day-to-day processes.
6. Executive Visibility
Serving real-time quality metrics to executive dashboards allows quality to be an integral part of executive decision making. Of course, it’s beneficial to get executive visibility in cases where quality is exceeding expectations, but it’s also valuable for executives to see where quality needs to improve. This allows quality to identify necessary additional resources.
7. Journey Towards Differentiation
Analytics are often separated into four categories: Descriptive, Diagnostic, Predictive, and Prescriptive Analytics. Descriptive analytics describe what happened, diagnostic analytics capture why it happened, Predictive Analytics identifies what is likely to happen, and prescriptive analytics defines what should be done. Metrics are most often descriptive analytics. Read this for more on quality management analytics.
Manufacturers are increasingly looking to Predictive Analytics to differentiate themselves. However, analytics maturity is a journey, and LNS recommends that manufacturers support real-time metrics first before advancing to Predictive Analytics. Real-time quality metrics are an important building block.
Art into Science
Interested in learning more? There are many resources in LNS’s Research Library that drill deeper into this topic. Additionally, metrics and performance are a common topic of interest in quality executive council roundtables.
Reference
1 - Shafir, E., & LeBoeuf, R. A. (2004). Context and conflict in multi-attribute choice. In D. Koehler & N. Harvey (Eds.), Handbook of judgment and decision making (pp. 341-359). Malden, MA: Blackwell.
