Introduction

In the current fiscal environment government organizations find themselves with fewer dollars and fewer staff—yet the demands for service continue to grow.  Faced with severe fiscal stress, organizations can do one of three things—raise revenue, reduce or eliminate services, or find more efficient ways to provide those services.  In the current political climate, raising revenue through increased taxes is not a popular option.  And since much of what government does is mandated by law and essential to the public good, there are significant legal, practical and political limits to government’s ability to reduce or eliminate services.  So is it possible to provide services more efficiently?  It may well be.  To find out, managers can use the tools, techniques and principles of process improvement to examine how they currently do things and to streamline processes where possible.  Obviously, streamlining the organization’s processes will not, in and of itself, make up for the magnitude of budget cuts faced by many organizations.  But it can help the organization maximize available resources by carrying out its work as efficiently as possible.

Process Improvement vs. Reengineering

First, it is important to differentiate process improvement from process reengineering.  Reengineering has been a very popular management tool particularly in the private sector.  Hammer and Champy (1993, p. 32) define reengineering as “the fundamental rethinking and radical redesign of business processes to achieve dramatic improvements in critical measures of performance such as cost, quality, and speed.”  Linden (1994, p. 75) defined reengineering as “challenging fundamental assumptions on which the organization is built by radically redesigning its processes, systems, and structure around desired outcomes.”  

Thus, reengineering is a tool for making radical change.   As Hammer and Champy (1993, p. 31) explain it, reengineering means literally starting over from a blank sheet of paper. 

It doesn’t mean tinkering with what already exists or making incremental changes that leave basic structures intact…It does mean abandoning long established procedures and looking afresh at the work required to create a company’s product or service and deliver value to the customer.  It means asking this question:  ‘If I were re-creating this company today, given what I know and given current technology, what would it look like?’  Reengineering a company means tossing aside old systems and starting over…and reinventing a better way of doing work. 

No doubt there are many public managers and employees who have wondered why their organization does things the way it does, or why it does them at all.  When asked that question, one of the most common responses is that “we’ve always done it that way.  It’s the way they told me to do it.”   Unfortunately, reengineering calls for a degree of radical change that is not always possible given the realities and constraints of managing a public organization.  There are very few instances in which public managers can simply toss aside old systems and start over, however desirable such revolutionary change may be.    

Paul Strassman (1994, p. 3) notes that:

To be effective, any way to improve the way people work must be evolutionary, not revolutionary.  The long history of miscarriages of centrally planned radical reforms, in addition to reengineering's as-yet-dismal track record… suggest that evolutionary approaches deliver better and more permanent improvements.  Evolutionary change stimulates morale and imagination, creating conditions for rewarding organizational learning and for inspiring employees to discover innovative way to deal with adversity and competitive challenges. 

Strassman calls for a more moderate approach he labels “business process improvement” or BPI.  BPI is a continuous process that calls for applying rigorous methods to charting, costing, and process-flow analysis of “as-is” conditions.   The objective is to create a learning environment in which renewal and gain will be an ongoing process instead of a one-time event.  In BPI, the people directly affected by the potential changes study the as-is conditions and propose improvements. 

Thus, process improvement can provide much the same benefit as the more radical reengineering, but has a number of distinct advantages. 

• It is more in-tune with the pace at which people understand and adapt to change.

• While most public managers do not have the authority to “start over” and make radical changes to mandated programs, most do have the ability to identify and propose “process improvements.”

• It taps into the knowledge and experience of those who carry out the process; thus, it can improve morale and job satisfaction.

The Language of Process Improvement: Some Terms and Definitions

In the following narrative, a number of key terms are defined as they apply to process improvement. This will be useful in understanding the process improvement model, which will be discussed in the next section of this article. 

Process – A set of business steps or activities that combine to produce a defined output (product or service) such as hiring a new employee.

Step – Actions taken by individuals or groups to execute an intermediate output of the process, e.g., interviewing candidates for the job.

Task – A discernable work behavior that, together with other tasks, make up a step.  A task is the most micro-level work that you will use in process mapping.  An example of a task would be setting interview appointment times for the job candidates.

Event/Trigger – A transaction or situation that initiates a process.  For example, an employee’s resignation may trigger the process of hiring a new employee.

Process Map – A graphic description of a process or activity showing the sequencing and dependencies of steps and tasks.

Bottleneck – A step in a process where work or information waits in line or queue to be acted upon.

Value-added Activity – A step in a process that results in a benefit for the end user.  Whether a step is value-added can only be determined by the end user.

Stakeholder – Any group or person who has a vested interest in a process.  Stakeholders would include end users (consumers, customers, constituents, clients) or people involved in the process (employees, suppliers, other agencies).

Parallel Steps – Activities in a process that can be done concurrently.

Sequential Steps – Activities in a process that have to be completed in a certain order, one before the other.

  A Model for Redesigning and Streamlining Processes

The process improvement model, as discussed in the literature, includes eight major steps.

1.      Identify the process or parts of the process to be streamlined.

2.      Establish and “charter” a design team.

3.      Map the current process.

4.      Identify the non-value added steps.

5.      Set a “stretch” objective.

6.      Design a new process.

7.      Identify implementation issues and strategies.

 We will look briefly at each of these steps.

Step 1 - Identify a process to streamline.

·        Core processes are those that touch an external customer, such as when an employee fills a customer’s order, responds to a customer complaint, or provides a service.  Core processes are the reason the organization exists. 

·        Support processes are internal to the organization and include such things as recruiting staff, purchasing, or issuing work orders.  Support processes enable the organization to carry out its mission (Linden, 1998, p. 9).

In selecting a process to improve, you should listen to your stakeholders and regard their input, complaints and suggestions as “free consulting.”  If you listen, they will clearly tell you where to start.  You must also consider both the impact of improving the process and the feasibility of improving the process. 

In terms of impact, there are several questions to consider:

1.      What major challenges are we likely to face in the next few years?

2.      Which processes, if improved, would give us the ability to meet those challenges?

3.      Which processes, if improved, would make the biggest difference to our current and future end users?

4.      Which processes, if improved, would have a significant impact on our key goals?

5.      Will there be a cost savings?

6.      Will it help us improve quality?

7.      Will it help us meet or exceed customer expectations?  

In terms of feasibility, questions include:

1.      What are the costs to make changes?

2.      What are the external constraints, e.g., statutory or regulatory requirements?

3.      What are the organizational values and culture?

4.      Is the organization adaptable to change, or resistant to change?

5.      What degree of support is there from process managers?

6.      What is the status of current and planned information technology?

Clearly it is desirable to start streamlining processes that you judge to have both a great impact and high feasibility. 

Step 2 - Establish and charter a design team.

Step 3 - Map the current process.  

In carrying out this step, it is critical that every effort is made by the design team to insure they truly understand the current process in great detail.  An accurate process map is essential.  Process mapping entails a number of tasks including:

1.      Identifying the people who perform the process.

2.      Identifying each of the steps and substeps anyone involved in the process has to complete.

3.      Determining the time and cost it takes to carry out each step and substep (including waiting time when nothing is being done to move the process to the next step), and total process cycle time.

4.      Creating a visual diagram or flow chart of the process (see Figure 1).

5.      Identifying and challenging the design assumptions underlying the current process.  

Figure 1 – Example of a simple process map.

Step 4 - Identify non-value adding activities.

By carefully studying the process map and challenging the underlying assumptions, the design team should be able to identify non-value added steps, bottlenecks, and sequential processes that could be completed in parallel.  Interviews with end users and process managers can provide critical insights.

Step 5 - Set a performance target.

The performance target, or “stretch” objective, is designed to challenge the status quo, to push employees to abandon the “way we have always done things” mentality, and to encourage them to think creatively about how things could be streamlined. 

Let’s assume, for example, that an organization’s current process for recruiting, screening, and hiring a new employee to fill a vacant position takes 90 days.  Through its interviews, the design team has learned that department heads find this unsatisfactory and it creates a number of problems throughout the organization.  It ultimately adds to customer dissatisfaction as well since positions that go unfilled for extended periods of time reduce the ability of the organization to provide service in a timely manner.  Indeed, it creates a vicious cycle by adding to the employee turnover rate since unfilled vacancies add to the overall workload of current staff. 

The users feel like it should take no more than 30 days to fill a vacant position.  They believe this allows sufficient time for the organization to do a fair process that meets legal and organizational requirements.  Thus 30 days becomes the stretch objective.  But this represents a significant reduction in process time.  How can this be accomplished?

Step 6 - Design the new process.  

In designing a new process, the design team must be guided by a number of principles (Linden, 1994, pp.75-120). These principles are enumerated below.

1.  Challenging assumptions.

First the team must surface and challenge the assumptions that serve as the rationale for how we do what we do.  Linden challenges a number of assumptions that typically underlie the design of governmental processes, including: 

·        Division of labor leads to efficiency.  In fact, it more often leads to narrowly defined, boring jobs that stifle employee energy, creativity and satisfaction.

·        Multiple layers of review and approval lead to accountability.  In fact, multiple layers often lead to less real accountability because people may not assume responsibility for accountability since (many) other people have to review and approve everything anyway.  When everyone is responsible, no one is responsible.

·        Employees are likely to cheat, make mistakes, and take shortcuts.  This is the classic “Theory  X” assumption that causes organizations to create numerous, sequential, non-value added steps for review and oversight.

·        The conditions that resulted in this rule, policy or regulation are still in existence.  It could be such rules are no longer relevant or necessary given current conditions.  

2.  Organize around outcomes.

Most organizations are awash with data about what they do, but much of it consists of input and output measures.  Input measures include such things as the number of requests  for service, or the number of dollars spent.  Output measures include such things as the number of people served, the number of arrests made, or the number of miles of road paved.  Outcomes measures, on the other hand, more closely indicate the extent to which the organization’s programs and services make a difference.  In our example above, one of the customers of the process is the department needing a new employee.  The outcome is the recruitment and retention of a capable staff.  Thus in streamlining the hiring process, it is critical to consider the needs of the hiring department rather than the needs or convenience of human resources staff.

3. Substitute parallel for sequential processes.

Let’s assume, in our example, that a request to fill a vacant position requires a number of approvals (a safe assumption in most bureaucracies).  Once a position becomes vacant the immediate supervisor makes a request to fill the position.  This goes to the department head for review and approval.  Then it is forwarded to the division head.  From there it goes to the administrator.  Once it receives the appropriate signatures, it goes to the human resources department who will then start the advertising and recruiting process.  Each of these steps is carried out in sequence, one after the other.  At each step, there is the potential for the request to lay in someone’s in-box for some period of time, perhaps days or weeks, before it is reviewed, signed and passed along to the next step.  Assuming that each of these approvals is necessary (and this is an assumption that the design team should seriously challenge), it may be possible to get these approvals in parallel—that is, send a copy of the request, perhaps electronically, to each level at the same time.  This change could save considerable processing time in- and of-itself. 

4. Bring downstream information upstream.

Simply stated, this principle suggests the need for sharing information among everyone involved in the process.  If the human resource staff member who does the advertising and recruiting does not have good communication with the end user department’s needs, they may not be doing the most effective screening of applicants.  If the job description is not kept current by the supervisor, the candidates they are presented to interview may not have the knowledge, skills and experience really needed to do the job.  If there is not good sharing and discussion of future workforce needs and trends, the staffing function will not be as efficient, effective and responsive as it needs to be. 

5. Provide a single point of contact for customers.

Most of us have experienced the lack of attention to this principle in our dealings with organizations.  The more employees a customer has to deal with to get something done, the less satisfied they will be and the longer the process will take.  This principle suggests replacing numerous fragmented processes with one simple integrated “one-stop shopping” type process in which an employee is trained and authorized to complete the process from start to finish. 

6.  Capture information once, at the source.

Consider how frustrating it can be for a customer who goes to an organization for service.  They have to fill out a number of forms with a variety of information—name,  address, date of birth, work history.  Once these documents are screened, they may be referred to another department within the organization where they have to repeat the process of filling out forms, generally with the same information.  This not only frustrates the customer, it adds unnecessary time to the process.  Why not capture this information once, at the outset, computerize it, and then it can be available at every step in the process where it may be needed?

7.  First, redesign the process, then automate it.

While computers and technology have the potential to help us streamline our processes, increase quality, and reduce cycle time, if we simply apply technology to a poorly designed process, we may get just the reverse.

Step 7 - Identify Implementation Issues and Strategies.

While some process changes can be relatively easy to implement, others may be much more challenging.  The team must consider a range of issues including getting required approvals for the changes; time required to implement the changes; employee training or retraining required; computer program changes; redesign and reprinting of forms; and all of the costs associated with these issues.  Success requires careful and thorough planning. 

Step 8 - Implement and Evaluate the New Process.

This step may be the most challenging one.  Because streamlining means change, it is wise to remember John Kotter’s rule of thumb.  “Never underestimate the magnitude of the forces that reinforce complacency and that help maintain the status quo” (1996, p. 42).  Indeed any design team would be well advised to read Kotter’s book Leading Change before embarking on a streamlining project.

Once the new process is put in place, it should be carefully evaluated to insure it meets its design goals and the stretch objective.  Adjustments can be made as indicated.  Employees must receive the training and support they need to carry out the new process.  Communication throughout the system is essential.  And don’t forget to celebrate any improvements in processing time or customer satisfaction.  In this way, the practice of constant process improvement can be embedded in the organizational culture.

Streamlined processes typically share a number of characteristics.  These include:    

1. Several jobs are combined into one.  

2. Workers have the knowledge and authority to make decisions.

3. Steps in the process are performed in a natural order.
4. Where possible, steps are performed in parallel rather than sequentially.
5. Non-value adding steps are eliminated or, if necessary, taken out of the main sequence and performed in parallel.
6. Work is performed where it makes the most sense.
7. Checks and controls are minimized.

8. Reconciliation is minimized.

9. The customer has a single point of contact.

Kotter, J. (1996).  Leading change.  Harvard, MA: Harvard Business School Press.

Linden, R. (1994).  Seamless government: A practical guide to re-engineering in the public sector.  San Francisco, CA: Jossey-Bass.

Linden, R. (1998).  Workbook for seamless government: A hands-on guide to implementing organizational change.  San Francisco, CA: Jossey-Bass.

Hammer, M. and Champy, J. (1993).  Reengineering the corporation: A manifesto for business revolution.  New York, NY: Harper Business.

Strassman, P. (1994).  “The hocus-pocus of reengineering.”  Retrieved April 13, 2004 from http://www.strassman.com.

Jon B. Pierce is the Associate Director of the Governmental Research and Service unit at the University of South Carolina’s Institute for Public Service and Policy Research.  He can be reached at jbpierce@gwm.sc.edu.