July, 2017

Every single thing we do, or don’t do, during our work day can affect the company’s bottom line.  There is a measureable hard cost in time expended, tools used, and resources consumed. These direct cost elements are a form of metrics that can be relatively easy to collect and then measure their performance over time. This is especially true in organizations like product development, engineering, logistics, quality control, and customer service that are accustomed to handling lots of product-related data with fanatic precision. For these groups understanding the value of configuration management metrics is an important role for the CM professional.

Why the subject of metrics is so knotty for most organizations…

It is not uncommon that performance data and cost metrics receive scant attention until some urgent financial justification or ROI calculation is required by management to support the next big initiative. As many of us have experienced, all too often an entire organization stops work to become consumed in the exercise of identifying and gathering up metrics data as if no one had ever done this before.

Not surprising, the excavation and ensuing debate on metrics often unfolds at a painful pace. Little time is left at the end of the process for appreciating that these numbers, which should have been easier to arrive at, can be dwarfed by longer term costs which are harder to identify and even more difficult to measure as a performance metric. These longer term, indirect or soft costs, are plentiful and can include the cost of lost opportunities, financial risks incurred, repair or recall expenses, management attention diverted, and other path-dependent surprises that await downstream hiding in the lifecycle of a product line.

Indirect, or soft, costs are often discounted in the financial analysis as being unknowable or rationalized as unimportant, until they emerge and eat your lunch. That is particularly unfortunate for the configuration management profession because these costs, or their avoidance, can be very substantial in valuing the total benefit of CM outside of engineering and across the entire business including the supply/service chains. An example is a high-tech manufacturer who courageously admitted that when a recent new product line was being launched they had budgeted in a cost of $1M for the purchase of incorrect, faulty, or obsolete parts.

Managers don’t want an education in Configuration Management...

In this first CMsights post on metrics we begin by sharing how CM professionals should approach managers and executives – who are not CM literate and don’t want to be – with understanding the financial benefits from having a solid CM strategy, plan, and solution. This approach requires a command of the metrics used to construct an initial baseline. In the next post on metrics we will examine the economic cost and business risks of not having CM under control, or an underperforming deployment because there was no baseline metrics to measure it against.

CM professionals who have been trained and certified in the standards and best practices of configuration management can recite the many benefits of CM in their sleep. It’s easy for them to present alarming scenarios where the lack of configuration management and change control can kill off a whole product line, disrupt customer operations, and contaminate a company’s market reputation.  But when asked by their managers to defend the benefits claimed or risks averted with real numbers, very few can respond with credible metrics. It’s not surprising because rarely do product development cycles or project plans provide the additional time to perform a forensic analysis of the work just performed when there was barely enough time to do the work.

Inevitably, conversations at CM conferences always include comments that “few executives in my organization truly understand the importance of CM….nor do they allow me the time to document the benefits.” It’s not surprising that managers don’t want to invest time to understand CM because quite frankly they don’t care about CM! In CMstat’s experience from over twenty years we have found arguments to get executives to care about CM, or investing in CM for CM’s sake, will be futile unless they are put in context of what business executives already care about.  That means starting at the end of the lifecycle with the desired results using financial data, and not starting at the beginning of the lifecycle with configuration and product data!

In doing so, the challenge for CM practitioners becomes framing the conversation in context and terms that non-CM’ers will immediately understand without having to educate them about configuration management. That means producing a financial analysis of the cost of doing business, the cost of day-to-day processes, the cost of potential risks incurred, and how effective CM processes can save money now or avert downstream costs.

Establishing a baseline of core metrics is step one…

As a first step, just as in a CM plan where a product configuration baseline is needed, for CM metrics a performance or cost baseline must also be established. How can any ROI calculation be taken seriously if there are no metrics that measure the existing process baseline? All the large-dollar ROI predictions in the world will not have a dime’s value to management without a credible baseline. Capturing, as metrics data, the true cost of each existing task in the “as-is” baseline will provide the context for all those non-CM experts, who will always be non-experts.

For years companies have looked to their investments in IT and software applications to improve, transform, and even disrupt their baseline processes.  The cited “to-be” benefits featured processes performed more efficiently and accurately through greater levels of standardization, automation, integration, optimization, collaboration, and innovation. Yet, some manufacturers have struggled to justify these investments by believing the configuration management of their products, systems, or assets is so unique and unwieldy that it is always a one-off process which must be reinvented from scratch each time.

This mistakenly gave the impression that the notion of using baselines was not relevant or so situationally fluid that they were impossible to capture. We have heard our customers state many times that their products, processes, organizations, gaps, challenges, and requirements are all so different from anyone else – and different even within their own business – that resorting to the use of metrics and definition of baselines was just not applicable to them.

All so often we see these same organizations working their day-to-day tasks without any measurement metrics defined, much less collected and tracked. Yet their management continually seeks to learn how much progress they have made once implementation of new processes or tools are up and running.  That is like spending a year watching your diet, working out, and giving up dessert without weighing yourself at the start.  How will you ever know if you’re getting the results you want without an initial baseline measurement?

A preview of important metrics to come…

In CM there are numerous metrics that can be gathered in support of the baseline definition. With few exceptions they are independent of a larger CM strategy or the specific CM software solution used. A few of these metrics include:

  • throughput of change per week, month, year
  • how many document releases made
  • issues resulting from design deficiencies
  • time necessary to find correct versions of documents, BOMs, As-Built configuration records
  • length of time needed to route changes for review
  • man-hours required to install safety-related mod kits to fielded items

In CMstat’s presentation at CMPIC’s CM Trends 2017 in Orlando on August 28, we will discuss these metrics and more. In a CMsights post to follow after the conference we will share the details with readers who were unable to attend this most valuable annual CM conference.  We hope to see you there, and until then we have a simple four-question survey that we will report back on during the conference.

Does your organization capture metrics for the following: