New product development
tools, tips and techniques
The New Product Development (NPD) process we have implemented at Harris Corporation, Broadcast Communications Division is based on the Stage-Gate TM (Cooper 2001) process. Using that as an outline we will review tools that range from portfolio management through project planning to performance tracking. The tools and project management process are closely coupled with the practices described in A Guide to the Project Management Body of Knowledge - PMBOK® (PMI 2000). These tools were developed to make the jobs of the Project Managers easier while providing more consistent look and feel for our projects.
These tools use mostly off the shelf standard applications such as Microsoft Excel, Word and are based around our project management software, Microsoft Project 2002. Visual Basic for Applications was used to help automate some of the tools, but many are simple templates, which can be modified for use in any project management process.
Project Managers of all project types and skill levels should find the material presented useful since it is focused on project management issues that affect all projects. These tools don't ensure success of our projects, but help make the job of executing projects more efficient which allows more time to be spent on the critical aspects of project management.
The only good tools are those that are used. With that in mind, our goal was that the tools be readily available and useable. Therefore, wherever possible, the tools were developed using off-the-shelf software packages on the project managers’ desktops. This limited our choices, but also ensured the tools we created could be used by anyone on the team. The basic software utilized in the creation of these tools consists of Microsoft Office products including Word, Excel, Project, and Visio. We were able to connect these software packages and automate some of the tools utilizing Microsoft Visual Basic for applications.
The basic formats and functions of the tools can be adapted to non Microsoft applications. Due to the broad use of these software packages in our office it made the most sense for us to stick with Microsoft.
Stage Gate Process and Organization
Two changes at Harris that have helped create a more project-oriented culture were the adoption of the Stage-Gate TM process and the reorganization of the development group into a matrixed organization.
The original development group was structured in a functional organization and divided by the major product groups Radio and Television. The development group was led by the Vice President of Engineering, but the two groups functioned independently. Radio projects were planned and executed by the radio team and TV projects by the TV team. There was little cross over of resources or skills. This organization structure is illustrated in Figure 1. This organization had some advantages but the inefficiencies of managing a global portfolio of projects limited our ability to complete the correct projects in a timely fashion. Instead of basing project priority on what made most sense for the business, the next project was based on which functional group was ready to start their next project. Staffing similar expertise in both areas was expensive and not an efficient use of resources.
The development organization was changed to a “weak” matrix organization, as shown in Figure 2. It is considered “weak” since there is not a strong Project Management organization reporting to upper management. Instead the project managers are picked from the various functional areas. The new structure enables more flexibility in project execution, which in turn, allows more projects to be completed and ensures we are working on the right projects. A structure was needed to effectively execute projects. Thus was born the New Product Development (NPD) process. The new process and organization requires stronger project management, which demanded our project managers receive the proper training. In many cases, project mangers went on to obtain their Project Management Professional (PMP) Certification.
The NPD process utilizes a 5 Gate, 6 Stage process. The process provides a structure to run projects and provides pre-determined checkpoints to review the projects to ensure the right projects are executed. The stages and gates are shown in Exhibit 3.
The tools discussed next are structured around the NPD process and cover Project Initiation through Project Execution with the primary emphasis during the planning and execution stages.
Exhibit 2: Matrix Organization
Exhibit 3: NPD Stages
What is the Right Project?
The first two Stages deal with choosing the correct projects based on market needs. Currently several tools are in place to determine both the correct project and how it can fit within the limited resources we have available. These tools are based on Excel spreadsheets and are relatively simple to create but provide a consistent way of looking at the portfolio of projects available and to determine the correct one to execute. The first tool utilized by the product managers combines all the proposed projects into a single spreadsheet and ranks the priority of project by product type (radio, TV) and overall importance to the organization. Metrics are used to determine priority, including ROI and projected revenue, as well as some softer criteria such as strategic needs and market demand. The benefit of this tool is that it combines all the projects into a single list and requires product line management to make the hard decisions on which projects are the best ones to work on. Exhibit 4 is a sample spreadsheet showing these features. The project portfolio list is reviewed on a biweekly basis to ensure that the next best projects from a market and revenue standpoint are at the top of the list.
Exhibit 4: Project Priority Spreadsheet
The project priority spreadsheet serves as the “wish list” of projects. The next step is to determine how these projects align with available resources in a given period. In order to estimate the required resource mix a first order estimate of work for each project is loaded into MS Project. These estimates are based on historical references and expert judgment. They are used to obtain a better idea of which projects can be executed with respect to our limited resources. These first order estimates are loaded into a master project along with active projects to get an overall project portfolio. This data is exported from Project into Excel for analysis.
Exhibit 5 shows the Portfolio Manager spreadsheet, which combines the projects (both in-process and proposed projects) and resource capacity. We use this tool to try to fit proposed projects within the resource constraints of the development group. This is an iterative process and requires the flexibility to move project priority level or estimated start based on resource capacity. This exercise is run on a quarterly basis, or as needed, to help choose the next best project to begin planning. The next Exhibit (6) shows the overall portfolio leveled for a given period. We use this graph as a sanity check to ensure the department is reasonably leveled after going through the capacity vs. projects exercise.
Figure 5: Portfolio Manager Spreadsheet
Planning the Project
Once the portfolio analysis work is completed and a reasonable priority of projects is determined, the job of planning the next project can begin. The next best project has a business case developed and is presented to the Gate 2 review board. Upon approval this gives the funding and authority to the project manager to begin the full planning process. During this stage, project managers must apply their full project management education and skills. The process we follow in planning a project aligns closely with the PMBoK. A project management guideline was created in MS Word to cover the various aspects in detail. It includes the following steps:
- Team Development
- Scope Definition
- Task Definition, Estimation, Sequencing
- Risk Assessment
- Schedule Creation
Figure 6: Overall Resource Loading Graph
The first step is to put together the planning team. This cross-functional team consists of subject matter experts, and includes the design group, marketing, manufacturing, service, and any other group with a vested interest in the project. Exhibit 7 shows a standard RAM, which is followed to show responsibility of each functional group to the stages in the NPD process. We also develop a project specific RAM for each design element in the project plan.
Figure 7: NPD RAM
Scope definition is one of the most important parts of the planning process. Poor definition will doom a project to failure, even before the first task is started. In order to have a consistent scope for projects, standard templates were developed to capture the “what” is needed for a product. The “how” to develop is left to the design team during the concept and planning stage. The definition is reviewed by the team to ensure it is clear, concise and measurable. The use of a matrix format reduces ambiguity sometimes found in narrative scope documents. Feature needs are balanced with the technical risk and cost of development to define the final product / project scope. This is an iterative process and is reviewed during the entire planning phase to ensure the right criteria are met for the appropriate risk and cost. Figure 8 shows a sample requirements matrix.
Figure 8: Requirements Matrix
Following the preliminary scope definition, work begins on developing the design concept, including block diagrams and sketches. Once that is complete, the Work Breakdown Structure is developed. We have developed a basic template for our projects, which uses the NPD stages within the WBS. Although technically not a PMBoK approved method, it allows alignment of our NPD process, accounting rules, and project schedule. We currently align the major deliverables, which are at level 3 of
our WBS, to our accounting structure for collecting charges.
Figure 9: WBS
We seldom create a graphic WBS or use the Post-It note type exercises. We have developed a MS Project template that provides our basic WBS structure and generic deliverables down to level 5. (Exhibit 9) This template contains all of our default settings to allow consistent views and reporting. This provides the starting point for our schedule creation.
Figure 10: Task Worksheet
We flesh out the details of the WBS using a variety of methods, including offsite project planning workshops where we get the entire team together out of the office to work on the plan. We then break up into single-subject groups to focus on the key deliverables. The team uses a MS Excel template to list tasks and provide time estimates. This is shown in figure 10. Duration estimates are usually based on expert judgment, but the teams are beginning to use some historical data to estimate standard activities found on all projects.
Standard activities include circuit board layout and fabrication time where we have good historical averages on durations of tasks. For the majority of tasks the team still relies heavily on past experiences and the judgment of key experts. We use team review of the task estimates as a sanity check to ensure that task times are not too inflated, or underestimated. The project manager takes these Excel-based estimates and loads them into MS Project. Sequencing is then done with the help of the team to ensure the work flow is correct. A standard template has been developed in MS Project to provide a basic structure aligned to the standard WBS. This template has all the customized views, macros, and calculations used at Harris.
Contingency Reserve and Risk Management
Once the schedule is loaded we run several exercises to ensure that the workflow and sequencing of tasks are correct, and the duration estimates are reasonable. We also review the risks associated with key activities and deliverables. These risks are tracked either within the MS Project schedule or using an Excel spreadsheet as shown in Figure 11. This sheet allows the team to score each risk on probability and impact; and to record mitigation plans.
Figure11: Risk Matrix
With the risk matrix, we use the 50/50 rule and a scaling spreadsheet to estimate an appropriate amount of contingency. This is dependant upon many factors, including risk and experience of the team (Weinberg, 1979). Figure 12 shows a sample scaling factor estimation tool. Typical project contingency reserves of 25%-35% are typical on NPD projects. Throughout the process the market requirements are considered with respect to schedule time, and trade-offs are made on scope, product costs, and project costs (resources). After several iterations engineering, product line, manufacturing, and all other groups involved sign off on the accepted project plan and the full business plan is presented to management for approval to implement.
Figure 12: Weinberg Task Scaling Estimator
In order to ensure that the requirements of a Gate review are met, a detailed checklist has been created for each gate. This checklist identifies all the requirements for that gate. If the checklist is complete going into the Gate Review, the project is virtually assured of receiving the coveted “Go” and is not “Recycled”. Figure 13 is an example of this matrix. It includes the specific element, who is responsible for completing the element, the requirement, and who is required to review and sign off before proceeding to the Gate meeting. This checklist facilitates the Gate review meeting and keeps it focused and on track.
Figure 13: Gate Matrix
Once successfully completing Gate 3, the project is ready to implement. In the current system, the core team is typically already applied to the project from the planning stage, and there is little interruption in work. The concepts developed earlier are in process of being further improved, and the project quickly moves into the development phase. Based on what is in the plan, resources join the project team and the full work begins. One previously-encountered problem was that all the project information was loaded into MS Project, but few people had the access or the inclination to run the tool. In order to facilitate communication and align the project schedule with actual work performed, another tool was developed.
This tool extracts the scheduled work for each resource for the next 30 days, and imports it to an Excel spreadsheet. The team members can than easily access this data to review what work is planned and when it is due. There is space available to provide a report at the end of each week regarding the work completed and any comments. This allows the project manager to update the schedule off-line and then review the updated schedule with the team at the weekly status meeting. Figure 14 shows an example status report.
Figure 14: Weekly Task List
Project Control and Monitoring
Our NPD process requires a monthly review of all projects with upper management. In order to facilitate this review a flash report was created to provide a quick view of each project with a minimum of paper. Figure 14 shows such a report. The report includes earned-value metrics on schedule and cost performance for each project. Project managers typically track earned-value on a weekly basis within MS Project to ensure their projects are meeting the plan.
On the flash report (Exhibit 15) any performance of less than 0.95 (SPI or CPI) requires attention. Once the performance drops below 0.90 the team needs to present a recovery plan to the management team. Early in the project some leeway is given, but if the trends do not show improvement then the team needs to address the issues and present recovery options. This flash report has reduced the time spent in the management reviews considerably, while at the same time allowing the management team to focus on the key issues of the project. In addition to earned value, product cost, risks, and marketing changes are all reviewed.
Figure 15: Flash Report
The tools outlined above represent a large sample of tools developed over the last few years to help the project managers succeed. They are constantly reviewed and improved to ensure they stay relevant and up to date. The next step in the NPD continuous improvement process is to increase the level of automation of these tools and incorporate them into an enterprise-wide project management tool. Work is in-process to review new software applications.
The following persons helped in preparing the original presentation this article is based on as well as developing the tools discussed.
|Dave Glidden||Director, Transmission Product Line, Harris Corporation|
|Tim Dittmer||Sr. Manager, DSP and RF Design, Harris Corporation|
|Tom Jones, PMP||Sr. Manager, Project Management, Harris Corporation|
|Jeff Malec, PMP||Sr. Manager, Transmitter Development, Harris Corporation|
|Randall Restle, PMP||Sr. Manager, Digital Design, Harris Corporation|
|Jason Rieg||Manager, TV Product Line, Harris Corporation|
|Frank Walker, PMP||TWG Project Management|
Project Management Institute. (2000) A guide to project management body of knowledge (PMBOK) (2000 ed) Newton Square, PA: Project Management Institute
Copper, R. (2001) Winning at new products, accelerating the process from idea to launch. Cambridge, MA: Perseus Publishing
Weinberg, V. (1979) Structured analysis. New York, NY: Prentice Hall
Walker, F. (2001) Project management using Microsoft project course workbook. Burlington, NJ: TWG
Proceedings of PMI® Global Congress 2003 – North America
Baltimore, Maryland, USA • 20-23 September 2003