Your project is in trouble! Now what?
PMI® and the PMBOK® Guide give you a tremendous base of knowledge regarding how to run projects correctly, achieving on time, on budget, and high-quality deliveries. There is extensive training available in the field to include industry training, PMP® certification, master's certificates, graduate degrees and post-graduate degrees. The vast majority of this material addresses how to implement projects correctly. This is all well and good, but there are industry figures, showing that as many as 50% of all projects fail for one reason or another. The likelihood that you will face the recovery of a troubled project at some time in your career is virtually a certainty. Mistakes have been made and your project is now off the tried and true path to success, so what do you do? This paper will cover several case studies of projects that have been through the process of recovery. The projects were in different stages of the project life cycle and had different problems. This will be a discussion of what went wrong, corrective actions and the key lessons that can be learned from these experiences.
These projects all had management escalations take place. As a result of those escalations, detailed reviews and corrective interventions took place. For these reviews, a series of steps were followed:
1. A charter for the review was developed and communicated to stakeholders.
2. A detailed review schedule was established.
3. A kickoff meeting was held, answering questions and concerns.
4. An extensive documentation review was conducted.
5. Interviews were conducted with all key project personnel and a sampling of all project personnel.
6. Preliminary findings were discussed with key project personnel and agreed.
7. Recommendations were identified, discussed with key project personnel and agreed.
8. A review report was developed and distributed.
9. Follow-up review and mentoring needs were evaluated and scheduled.
10. Local management was briefed on findings and recommendations. Local oversight requirements were discussed and implemented.
A key element of such reviews is to accurately identify the root issues. There are nearly always failures at a number of key project management fundamentals. The list of which fundamentals that are lacking varies from project to project. Depending on the specific set of problems that are encountered, where the project is in the life cycle, and what the customer issues are will dictate prioritization. In projects with significant problems, especially nearing the end of a project it may not be possible or desirable to correct all problems. In some cases the cost of correction may outweigh the impact in just working through the issues that result until the end of the project. Only after such a diagnosis and prioritization can corrective actions be identified and implemented.
Project 1: Large Document Database for a German Customer
This project was an extremely technically challenging software development project. The solution was a very large-scale custom software development with some custom hardware and a nine-terabyte database. The existing databases involved data from many sources in enumerable formats with varying degrees of “dirty” data. The team executing the project was rather inexperienced in large-scale projects. Schedules were being missed, quality was suffering, data conversion was taking forever, work hours were long, project attrition was very high, team moral was terrible and there were non-negotiable end dates. The project was broken into two major phases to provide schedule relief and meet a political milestone. In order to meet this first delivery, quality reviews were skipped and testing was relaxed, creating a probable backlog in system and documentation deficiencies, which would have to be corrected later. There were significant disagreements with the largest subcontractor regarding change orders.
Project 1: Findings
1. Risk had been identified and estimated. The schedule and cost for those risks was simply included in the basic estimates, so there was no longer any visibility into what part of the budget was risk.
2. Project difficulties had led to significant attrition and problems in obtaining replacement staff.
3. Project moral was at an all time low and team members had lost confidence in their ability to succeed.
4. The largest subcontractor was continuing to work, but at reduced staff. They showed a plan that would succeed at that reduced staffing level, but they continually missed milestones.
5. The data conversion problems were enormous and had delayed the schedule by months. This was already absorbed in actual schedule and cost at the time of intervention.
6. There was a significant search data base product included in the solution. There was a contract with the vendor to provide required customization of the product, but no work was taking place on that customization. At this time they announced that they were being procured by another company.
7. There were substantial problems with some third-party products. The third party acknowledged the problem and indicated they were working on the problem under warranty, but would not commit to a schedule.
8. There were extensive communications at all levels between the project team and the customer. There was, however, no coordination or management of those communications creating problems in scope control and expectation management.
9. Project managers and team leaders were conducting reviews and exercising control at a level acceptable with the smaller and less-complex projects with which they had experience. This was not a level that would facilitate effective management and control of a project this size.
10. The Work Breakdown Structure and schedule were not at a sufficient level of detail to allow accurate forecasting or control.
11. Team communications and direction were lacking in structure and coordination.
12. Team personnel were well qualified over all, but most senior positions were staffed by people with too little experience.
Project 1: Corrective Actions
1. Unable to separate risk, which was buried in the budgets and schedules and since the team's motivation and trust were low, they couldn't be readily separated. Assigned a shorter delivery time and budget for all tasks uniformly, at 80% of plan, in order to manage to a shorter schedule and lower budget, effectively creating reserve.
2. Put team-building efforts in place, to improve moral and communications, including subcontractors and customers creating a more integrated team.
3. Briefed team on review results identifying the problems, but also stressing their successes and capabilities, in order to build confidence.
4. Worked with the customer on the known project problems and created an environment where the team could have a successful milestone and further build confidence.
5. Took steps to truly include the primary subcontractor as part of the team. Negotiated to postpone any litigation and maintained staffing levels at the required levels.
6. Worked the data issues with the customer. Disclosed the impact of past problems and establish change control around future data issues.
7. Met with the company acquiring the subcontractor. Documented the contractual obligations. Negotiated a mutually acceptable resolution.
8. Escalated product issues, but planned software workarounds were required, in order to maintain schedule.
9. Developed a consolidated communications plan and communicated the plan to the entire project team.
10. Established a regular schedule of meetings, reviews and communications. Trained project management and team leaders in appropriate requirements and techniques.
11. Examined every possible source for experienced personnel and paid a premium for key positions where required.
12. Developed detailed Work Breakdown Structures, estimates, and schedules.
13. Established quality gateways the project had to pass for future milestones.
14. Scheduled ongoing reviews and mentoring by experienced project management.
Project 1: Lessons Learned
1. Always maintain schedule and cost risk budgets separate from the baseline estimates. If you include them in the schedule and budget they will become part of the actual expense and schedule consumed and you will lose your contingency.
2. Manage moral from the beginning of the project and every day throughout the project. Projects are difficult and high moral is priceless.
3. Try to have an integrated team with a collective identity, which includes customers, your own employees and subcontractors.
4. Be wary of data conversion efforts, keep the customer involved and make the customer retain responsibility for clean up of “dirty” data.
5. Plan for and manage communications and expectations throughout the life of the project.
6. Establish quality, review, and control standards for your project. Train your team leaders and enforce the standards.
7. Obtain key staff with appropriate and sufficient expertise. Availability is not a skill set.
8. Develop Work Breakdown Structures, risk management plans, estimates and schedules and manage to them as living documents throughout the project.
9. Develop a mentoring program to grow your more junior project managers into senior project managers.
Project 2: Geographic Information System (GIS) for an Asian Customer
This project was a large-scale custom software development with a large data conversion and data load. Compaq was only providing management (thin prime contractor). There was a base product used to build upon to create the application. This product required extensive customization plus significant additional software components. This third party had an excellent reputation and references in the United States. The actual company contracted with, although utilizing the same name, was actually the Asian distributor and had no real software development or project experience. The third party was an existing contractor to the customer and had better relationships directly with the customer than Compaq.
Project 2: Findings
1. The subcontractor had no real software development or project management experience.
2. The subcontractor was used to doing product sales and didn't understand fixed price contracts. This resulted in significant cash flow problems for the subcontractor. As a result, they were trying to keep their cash flow to a manageable level. This resulted in inadequate staffing for the development effort.
3. Even if the subcontractor wanted to increase staffing levels, they didn't have the resources readily available.
4. The third party was utilizing the development personnel to also do the system testing.
5. Multiple problem reporting systems were being used for different test levels, resulting in no visibility across all of testing and also resulting in duplicate problem reports and inaccurate problem report closure.
6. The subcontractor was using their preexisting relationship with the customer to transfer blame for their performance problems to Compaq.
7. The system had high political visibility and corresponding high cost of failure to Compaq as the prime contractor.
8. Compaq had significant project management capability and software development capability, but had no real industry application experience in this market. They were depending on the subcontractor to provide that expertise.
Project 2: Corrective Actions
1. Identified a qualified and cost effective software development company to assist the subcontractor in completing the software development and integration.
2. Met with the U.S. subcontractor and escalated the issue. This resulted in incremental funding for the Asian distributor to complete the work.
3. Hired a respected Industry expert to assist with customer relationship management and system testing.
4. Set up a single problem tracking system. The schedule was adjusted to accommodate more testing time.
5. Assigned a second project manager to handle customer coordination during testing and to oversee the subcontractor.
Project 2: Lessons Learned
1. Evaluate subcontractors thoroughly. Just because a company has a good reputation doesn't mean that every organization, location, or division is of the same quality. Research the capability of the specific organization you are contracting with. Contract with organizations that have the exact type of expertise you need and not just industry expertise.
2. Staff at the correct level and correct expertise to enable you to track and manage subcontractors. If they have more expertise and better customer contacts than you do, consider bringing in outside consultants to help you manage them and communicate with your customer.
3. When you have to fill the role of “thin prime,” conduct regular reviews and audits.
Project 3: Custom Point-of-Sale System for a European Government
This contract was to provide a custom point-of-sale system for a European government organization. The deployment was to hundreds of offices across the country. The organization had limited technical support capability and a relatively low computer literacy among the users. The project was originally planned as a single-system deployment with one highly tested production release. This development was slipping schedule as a result of modified requirements and underestimation. The customer was insisting that not only the start and end of rollout dates be met, but that intermediate milestone dates be met, even though there was no operational or business impact as a result of those milestones. In order to meet these dates in some fashion, the project was being broken into multiple, phased releases. This met delivery dates, but deferred functionality. This significantly weakened the test phase; with only the base functionality being rigorously tested. This was to be followed by significant software releases after installation with much less thorough testing.
Project 3: Findings
1. The customer had given responsibility for the oversight and control of the project to their own information technology organization, which had no real involvement with the business aspects of their operations.
2. Dividing the project into releases allowed intermediate milestones to be met, but by doing thorough testing only on the first release. This added significant risk to the deployment. It also added significant overhead for operational release and testing to each site for each of the added releases. This would have, in reality, made the risk of significantly slipping the end date much greater. By allowing intermediate dates to move forward or backward as required and still doing a single well tested release, the end date would be late only a trivial amount of time. This amount of slippage had no meaningful impact on business operations.
Project 3: Corrective Actions
1. Conducted an intervention with the knowledge of the customer. Convinced the customer of the mutual benefit. This resulted in one of the customer's regional directors participating in the review and intervention. This resulted in buy in, on their part, for the recommendations.
2. Restructured the schedule back to the original plan with a single delivery. Established commitments only for very high-level milestones instead of all milestones.
3. Had the customer regional director participate in the review creating a high level of knowledge and awareness on the part of the regional director. The regional director then took control from their information systems organization and drove the project recovery from the customer's side.
Project 3: Lessons Learned
1. Ensure that you have the correct level of customer executive sponsorship to facilitate the size and importance of your project.
2. Insist that an appropriate organization within the customer oversees your project and that the individual has appropriate expertise.
3. Don't commit to intermediate milestone dates as hard dates. Portray milestones as management and control gateways that are planned at points in time. Only allow a project to pass such gateways when these milestones are really complete with sufficient quality.
Project 4: Network Design and Rollout for an Asian Rail Station System
This project was to provide design, configuration, testing, installation, and deployment of a network infrastructure across an entire country in Asia. The project was bid and accepted by personnel in the country who were very inexperienced in projects. This project was identified as being in recovery from the start of contract due to strict delivery penalties, one-sided (against Compaq) terms and conditions, vague requirements, a weak delivery organization, and an unsophisticated and difficult customer. The funding for the project was provided by the World Bank. In this country, a foreign company cannot directly contract with the government, so a tendering company held the actual contract. A local tendering company is always required. This tendering company has the actual contract with the government and they, in turn, subcontract to the real prime contractor. In this case the tendering company was actually employed by the World Bank.
Project 4: Findings
1. Tendering companies normally pay any fees to “lobbyists,” which may be required in order to win and successfully deliver the business. In this case, since they were employed by the World Bank and could not fulfill that roll. As a result, Compaq had to deal with the issue.
2. At the time of the bid, there were no agreed subcontracts and bids for the subcontracted work to be done.
3. Compaq's organization, within this country, was undergoing significant attrition. This resulted in a lack of knowledge about the bid and a shortage of expertise to get the work done.
4. No Work Breakdown Structure, detailed schedule, risk analysis, or firm definition of deliverables was done.
5. Costs and revenue were not associated with the deliverables.
6. The roles and responsibilities and project organizational structure were vague.
7. The customer had a low level of technical ability and little motivation to see the project completed.
8. There was a significant amount of hardware and software products to be imported into the country. Clearing customs would have created significant delays.
Project 4: Corrective Actions
1. Developed and implemented a list of required short-term project management activities.
2. Developed project scope and a completed a list of project deliverables.
3. Identified and prioritized all known risks and used this to develop new risk estimates.
4. Developed a complete Work Breakdown Structure.
5. Using the project phases and deliverables, a new basis of estimate was developed providing a new estimate of the project cost.
6. Developed a project milestone schedule.
7. Developed and agreed to subcontracts with firm quotations.
8. Identified key positions and assigned priorities to staffing them.
9. Assigned a subcontracts manager to the contract due to the large number of subcontractors.
10. Conducted extensive external reviews on a quarterly basis to identify problems, track risks, and implement corrective actions.
11. Subcontracted low-level cabling and installation back to the customer's information technology organization. Allowed them to participate in the project and obtain additional funds. As a result, they worked the political and scheduling problems at the installation sites, smoothing problems and accelerating schedule.
12. Enlisted the customer's information technology organization to use their local contacts to work customs issues.
13. Increased the largest fixed price local subcontractor's scope and budget allowing them to deal with local “lobbyists.”
Project 4: Lessons Learned
1. Establish review and approval authorities for projects and ensure they are used consistently.
2. Establish specific approval criteria for bids and enforce them.
3. Delegate management approval authority not just on position, but also on management experience, management expertise, and organizational maturity.
4. Be aware of local business practices and laws. Have the contractual structure of your contract be reviewed by someone with local expertise.
5. Insist on sufficient planning, estimating, and scope definition before prices are established and bids released.
6. Obtain firm terms and conditions and quotations from subcontractors preferably before bidding, but certainly before signing contracts with customers.
7. Make your customer part of the project, get buy in early and let them handle organizational change management and internal political issues.
Proceedings of the Project Management Institute Annual Seminars & Symposium
October 3–10, 2002 • San Antonio, Texas, USA