Performance Improvement: The Behavioral Side of Improvement Work—A Walkthrough

In this third Column of our series on behavior and performance improvement, we shall walk you through an example of how we integrate behavioral analysis and design into a typical process improvement project, our aim being that you can see up close some of the techniques, tools and thinking we use. As a reminder, what we mean by behavior are those specific things that performers and leaders say and do. Of course there are many behaviors that occur in any process of even moderate complexity so our approach is to sort through them, find out which are most important to performance, and identify the ones that need to be carefully designed, guided and reinforced. As shown in Figure 1, we apply a “funneling” process to identify three types of behaviors:

  1. Behavioral Issues— Anything having a negative effect on performance that may be behavior-related and needs to be investigated. These may ultimately be traceable to one or more specific behaviors of performers or leaders, or to something in the performance environment.
  2. Key Behaviors—The set of behaviors having the greatest impact on process performance. How they are performed determines results so our improvement effort will focus on these.
  3. Critical Behaviors—These are behaviors that Performers and Leaders absolutely must do to ensure optimal performance. Some of these may have been identified earlier as Key Behaviors, while others are not formulated until the future-state design is being developed.

In short, we start with the many behaviors happening inside the process, find out those that may be related to performance deficiencies or improvement opportunities, sort those down to the ones that are really key to desired performance, and eventually focus on those deemed so critical they need the most attention and reinforcement from management. But to make these distinctions clear, let’s walk through the example. We’ll trace the typical phases and steps of a generic process improvement project. The project phases are Define, Analyze, Design, and Implement. As we move through each phase, we will apply behavioral analysis and design techniques whenever it makes sense to do so, but not necessarily at strictly controlled places, as you’ll see in the example. The process is “order fulfillment”, and specifically on the sales and order entry sub-processes at the front end.

Focus 1: Identification of Behavior Issues

In the Define Phase of this project, we are working with the sponsor and other senior leaders to identify the project purpose, problem statement, target process, improvement goals, timetable, current performance, team structure and numerous other elements. . In Figure 2 is a diagram of the order fulfillment process, showing its major phases along with other upstream or downstream processes. This diagram (which we call a Process Profile) is typically used early in a process improvement project to help define process boundaries and sub-processes. While developing the profile, we hear about possible process “disconnects”—i.e., things that are contributing to sub-optimal performance. Whether these are valid or significant issues remains to be determined, but some of these may very well be Behavior Issues. A process disconnect is a possible behavior issue when it is a statement about adverse process performance that can be tied to actions by a specific class of performers.

Three possible Behavior Issues we hear repeatedly are:

  • Orders from sales reps are inaccurate and incomplete
  • Order Entry does all clarification with customers
  • Information on process performance is limited and not consistently provided to sales leadership

So we annotate the profile with these Behavior Issues. At this point these issues should raise a host of questions for which we will want to find answers. We don’t yet know if either issue is even true, nor do we understand the magnitude and impact of each. On the issue of faulty orders, how many orders are incomplete? How many inaccurate? What’s wrong or missing from the orders? What effect does inaccuracy and incompleteness have on the rest of the process? We’ll want to dig further to get some verification and supporting data. On the issue of Order Entry’s doing all the clarifying of orders with customers, we might ask why is Order Entry performing this activity? Isn’t that typically the job of the sales rep? Might this not be confusing to customers, who are used to dealing with the sales reps? Might this delay processing of the order?

During this phase of the project, we also begin educating the organization’s leaders about the importance of behavior in performance management, using the simple behavioral model shown in Figure 3, which depicts a performance system as consisting of three elements: A is for Antecedents, which are all the things (training, supplies, equipment, instructions, policies, etc.) that precede the behavior; B is the Behavior in question; and C is for Consequences, which is what happens to the performer (positive or negative) as a result of performing. The model focuses on consequences as the most powerful influence on performance and the most likely predictor of future performance. That is, performance that results in some level of satisfaction is likely to be repeated while a behavior that results in something unpleasant to the performer is likely to be avoided.

We illustrate how leaders sometimes inadvertently establish consequences that are in conflict with each other—for example, a bank teller who is told to simultaneously “provide personalized customer service” and “keep the line moving”. Given this ambiguous scenario, tellers will decide what to do based on the immediate circumstances they face and what is reinforced. If they have the time, they may provide a lot of TLC to customers, the consequence being appreciation. If on the other hand, they have a supervisor hovering nearby and cracking the whip to “keep things moving,” you can bet the personalized service will dwindle rapidly, because the consequence of an unhappy boss tends to be more powerful than the consequence of a happy customer. We explain to the leaders that as we analyze the order fulfillment process we will be looking for behavior issues and seeking to understand the consequences and other drivers that may be causing those issues.

Focus 2: Key Behaviors and Causes of Poor Performance

In the Analyze phase of the improvement project we gather much more data about performance. We find out that over half of the orders are incomplete or inaccurate, necessitating corrections and calls to customers, so it’s pretty clear this is a serious issue. To go deeper into behavior issues, we often apply another model (shown in Figure 4) called the Human Performance System, (HPS) one used often by Geary Rummler to examine performance at the performer level. This model is similar to the A-B-C model described above but recognizes that while consequences are the most powerful variable affecting performance, there are numerous other variables to consider when analyzing why a given performance is sub-par. The HPS posits that every performer operates inside a “human performance system” consisting of the following variables that influence behavior:

  • The clarity and completeness of the output specifications (Do performers know what is expected?)
  • The triggering input and necessary resources (Do performers know when to perform and have everything required to perform?)
  • The consequences of performing (What happens to performers who do what is expected versus those performer who don’t? What are the immediate versus delayed consequences?)
  • Feedback to performers (Is there any feedback? If so, is it useful, relevant, and understandable? How quickly is it provided?)
  • The performers themselves (What are their skills, skills and willingness to perform?)
  • The design of the process itself (Is it logical and straightforward or complex, with multiple variations, decisions and participants)?

An HPS analysis can reveal that often a behavior issue has multiple causes and thus potentially a combination of solutions. In our order fulfillment case, we find out the following:

  • The paper order form turns out to be lengthy, complex, difficult and laborious to complete (i.e., lack of an effective Input/Resource).
  • The sooner a sales rep gets the order submitted, the sooner he or she gets a commission check and can also go on to the next sale (i.e., it’s a Consequence issue).
  • Many of the errors occur with orders for new customers and for new products to existing customers. In other words, the process has multiple variations and is more complex than performers and managers may recognize (i.e., it’s a Process Design issue).
  • There is little if any feedback given to sales reps on the accuracy and completeness of their orders; they may not realize the mistakes they are making (i.e., it’s a Feedback issue).

That issue of when commissions are paid is very interesting. At this point we might hypothesize that the commission policy is a major cause of some of the other behavior issues. (We hesitate to use the term “root cause” since it implies there is only one cause of poor performance and in this case there appear to be multiple causes.) But if sales reps get paid as soon as they submit the order (that is, regardless whether the order is accurate and whether the customer pays the bill and is ultimately satisfied), they are being allowed to wash their hands of further involvement in this process; they can leave all the problems to Order Entry and happily go on to the next order. Hmm, so a likely explanation for at least some of the performance problem of inaccurate and incomplete sales orders is a lack of accountability, consequences and feedback. And this might also explain why Order Entry is doing all that re-checking of orders with customers—because the sale reps are no longer involved.

By this time we have ascertained that the need for complete, accurate orders is urgently required for this process to perform as desired, so we have identified a Key Behavior. During the Analyze phase of the project we create a behaviorally annotated as-is process map (Figure 5) which includes Behavior Issues, their probable causes and Key Behaviors which by definition are required to optimize performance. Each Key Behavior is stated in terms of what should happen—that is, each sales rep should submit a perfect order every time.

We validate this process map by involving sales reps, order entry personnel and other SME’s who know and understand this process. We also teach the project team the ABC model, with emphasis on the power of consequences, using the example of paying commissions to sales reps without regard to order accuracy or completeness as an example of an unintended consequence.

But where does the commissions problem reside? Not actually in the order fulfillment process but instead in the management system, where policies are formulated and managerial practices are defined. To help us pinpoint where the causes of a Key Behavior may be emanating from, we often use another model, called the Effective Process Framework (Figure 6) which identifies the major components that can affect process performance, including:

  1. The process output
  2. The process design
  3. The underlying models and assumptions that govern how the process is designed and is performed
  4. Contributing processes and inputs
  5. Inputting processes
  6. The performer and management hierarchy
  7. Technology enablers
  8. The management system of policies, rules, goals, guidance and support

In this instance, it’s the commission policy—an element of the management system—that is affecting how sales reps treat orders. In all likelihood, changing the commission policy is outside the project team’s scope, but even if we can’t change the policy, we can point out to management what we think the consequence of it is on the quality of sales orders.

And we may still not know if all our suppositions are true, but we know enough to identify certain behavior issues as contributors to poor performance, single out probable causes of those behavior issues, and identify the key behaviors vital to process performance. As we conduct our next key leader and stakeholder meetings, we now have a more thorough analysis and data to support their understanding of as-is performance disconnects and the Behavior Issues and Key Behaviors that need to be addressed in the future-state business process.

Focus 3: Identification of Critical Performer and Leader Behavior

Now let’s fast forward to the Design Phase. In this phase the project team identifies Critical Behaviors and incorporates them into the future-state design. Critical Behaviors are of two types:

  1. Actions performers must take in order to achieve desired performance (some of these would have earlier been identified as Key Behaviors) but that require reinforcement by leaders. (Performers that require reinforcement are those that are difficult or complex to perform; and we have discovered during this project that some orders are complex, which is leading to a high error rate.)
  2. Actions leaders must take to provide the required reinforcement. To distinguish these behaviors from those that process performers must do, we designate these as “Leader Critical Behaviors”.

Figure 7 shows the behaviorally annotated future-state process map that was produced during the Design phase. Note the following changes and additions to the behavioral annotations:

  • The Key Behavior of “Orders from sale reps are complete and accurate” is now designated a Critical Behavior, meaning it needs to be reinforced by feedback and attention from leaders.
  • The Key Behavior of “Order status is sent to associated Sales Rep and Field Operations leader” has been added, which provides the feedback to sales reps and management attention that was missing in the current-state process. (You could argue that this is not really a behavior because a computer system, not a human being, performs this activity, but that’s a hair we won’t split.)
  • The Leader Critical Behavior of “Sales leaders discuss new customer/new product proposals and order completeness/accuracy metrics in weekly meetings with Sales Reps” builds in the opportunity for leader reinforcement (including compliments and corrections as needed).

The future-state design also deals with some of the other process deficiencies that were contributing to errors:

  • The paper-based order form has been replaced with an on-line form containing mistake-proofing features (e.g., pull-down menus, required fields) that virtually eliminate the most common errors.
  • The Order Entry role has been eliminated because the order is machine-processed when the sales rep completes the on-line form.
  • Information and feedback loops have been built in to provide greater visibility about order status and quality, enabling leadership to more systematically monitor and guide performance.

To close the loop on leader reinforcement, after the future-state design has been presented to management and approved, the Implementation phase includes workshops with leaders to help them develop coaching plans that support their adoption of the Leader Critical Behaviors assigned to them. For process performers, the Implementation phase will include training and other support to help them adopt their assigned Critical and Key Behaviors as well.


Our purpose in this Column was to give you a close-up view of how we integrate behavioral tools into an improvement project to more effectively engage leaders and design better future-state processes. We have tried to show how a focus on performer behavior begins right at the outset of a project, as we gather performance and other data and begin to formulate hypotheses about the problems we are told and the observations we make. Those hypotheses are gradually fleshed out into statements of Behavior Issues and are collected along with other types of issues inside and outside the process. As we analyze the issues and learn about their causes, we are able to identify Key Behaviors—the ones on which optimal performance is dependent. By the Design phase, we are able to identify Critical Behaviors—ones that performers must execute effectively and that leaders must reinforce.

While we used a process improvement project as the example in this Column, the approach we’ve described applies to any kind of improvement effort where behavioral change is required, including technology development projects. It is a hands-on approach to change management because it goes beyond stakeholder communications and change management training, and instead deals directly with the specific behaviors required of performers and leaders to achieve the desired changes, whatever they are.

Figure 1 Shaping the Behavioral Focus

Figure 1
Shaping the Behavioral Focus

Figure 2 
Order Fulfillment Process Profile

Figure 2
Order Fulfillment Process Profile

Figure 3 
ABC Model

Figure 3
ABC Model

Figure 4 
HPS Model

Figure 4
HPS Model

Figure 5 
Behaviorally Annotated Current-State Process Map

Figure 5
Behaviorally Annotated Current-State Process Map

Figure 6 
The Effective Process Framework

Figure 6
The Effective Process Framework

Behaviorally Annotated Future-State Process Map

Figure 7
Behaviorally Annotated Future-State Process Map

Alan Ramias & Paul Fjelsta

Alan Ramias & Paul Fjelsta

Alan Ramias is a Partner of the Performance Design Lab (PDL), a consulting, coaching and training company that specializes in organizational performance improvement and an instructor with the BPM Institute. He brings 30 years of consulting experience in the analysis, design and implementation with organizations in Asia, Europe and North America. He is a co-author of the books White Space Revisited: Creating Value through Process and Rediscovering Value: Leading the 3-D Enterprise to Sustainable Success. Before becoming a management consultant, Alan was an instructional designer, training manager and organizational development manager at Motorola. Alan joined The Rummler-Brache Group (RBG) in 1991, and led improvement projects, became a Partner and Managing Director of Consulting Services at RBG and was responsible for selecting, training and overseeing RBG’s consultant teams. Paul Fjelsta is a Founding Partner of accomplir where he specializes in developing integrated, next-generation process improvement frameworks that link Behavioral Leadership & Behavioral Change Management tools with Lean Sigma (DMAIC) and Rummler Process Methodology (RPM) toolsets to create "Sustainable Lean Sigma" and "Behavioral BPM". These blended toolsets enable client ownership of the organizational, business process, technology, and behavior changes required to bring about timely, sustainable and measurable business results.


Leave a Reply

Your email address will not be published. Required fields are marked *