NLPES Question of the Month

January 2001

What steps do you take to ensure that your projects finish on time?

From:   Jeanne Jarrett, Missouri

In Missouri's Oversight Division, we establish estimated hours to complete projects, but only as a planning tool.  By statute, we have one year from the date assigned in which to complete evaluations.  If we don't complete them within the year, we must ask our governing committee, the Committee on Legislative Research, for an extension of time to complete the project.  This has happened twice within the past seven years, but because of unusual circumstances.  We do our best to complete evaluations by January of each year in order to present the information to the legislature at the beginning of the legislative session.  We have a series of meetings throughout the course of an evaluation for the purpose of keeping focused and addressing issues quickly.  The first is a brainstorming meeting with the team assigned to do the evaluation.  We usually have two or three progress meetings along the way to address where we're at in the project and whether we are on target to meet deadlines.  The Director, Team Leader and Team Members all attend these meetings. If a project is dragging, we can usually identify the causes early and adjust resources or strategy.

From:   Shan Hays, Arizona

In Arizona, we have a pretty structured approach.  At the beginning of our 2-year cycle, we do some advance work and estimate how many hours we'll need for each project.  We then allocate the projects among our teams, based on the number of hours the team has available, team preferences, expertise, etc.  Each team is responsible for figuring out how to allocate staff and calendar time to get its set of audits done by the deadlines.

Budget hours for each project vary from about 500-4,500, but the typical project is probably around 2,500 hours.  We have targets for allocating the budget between the 3 major phases of the project: preliminary survey (20%), fieldwork (50% for smaller projects, 55% for larger ones), and report writing/processing (30% for smaller projects, 25% for larger ones).  We set these guidelines in 1997 after doing a study that found that most audits that went over 25% in preliminary survey ended up going over budget.

At the beginning of the project, the senior lays out a rough calendar, based on the size of the team and the budget.  These dates are usually treated as targets, not cast in stone.  However, later, near the end of fieldwork, we put together a detailed schedule for all the stops along the report review process for each report.  We do this on a spreadsheet.  This lets us make sure the editor, secretary, etc. don't get a stack of reports all at once.

These are the mechanics of scheduling.  We do lots of other things designed to make our work more efficient, so we can actually complete the work in the hours budgeted.  For example, we have a "lessons learned" process where we assess what we could have done differently; we consider every methods decision in light of how much time and budget we have; we try to staff projects with the right mix of people; we double up on supervision so things can run concurrently; share staff among teams when we need to, and so on.

No matter how well we plan, we still run into things like agency data problems that double the time it takes to complete a project.  Still, having a plan in place for the remaining projects lets us minimize the overall impact of these problems.

From:   Jane Thesing, South Carolina

We establish a budget for each audit when we have completed our survey work and have developed the audit plan.  The budget is stated in total hours and target dates are calculated for each step of the audit process (fieldwork, draft report, management review, exit, etc.).  Target dates are also established for each auditor for each section of the audit plan.  We monitor those dates as we progress.

If it appears that we are not going to meet our target dates, we consider the significance of the situation.  If having the audit published by a certain date is important to the General Assembly or to the members who requested it, we make adjustments to enable us to complete it on time.  One adjustment is to shorten the in-house review time scheduled.  For example, normally management would have a week to review a draft report.  If time were of the essence, the manager could ask the other managers to review the draft in a day or two.  Another adjustment is to drop less material/significant areas of the plan or narrow the scope of the fieldwork areas that have yet to be completed. This involves a tradeoff, but we have to do some hard thinking and consider whether the General Assembly/State would rather have the main essence of the work sooner or the entire product later.  Also, staff commonly put in extra hours (not required) to expedite the work.

If having the audit published by a certain date is not of concern to the General Assembly or to the members who requested the audit, and the audit objectives are considered significant, we would continue and complete the plan, adjusting the target dates (we don't adjust the budget in hours and continue to record our hours, so there is a record of how much time/money the audit took compared to the original plan). Improving our timeliness is a goal we are always working with, so I look forward to reading suggestions from everyone else!

From:   Rachel Cohen, Texas

The key steps we take to help ensure that projects get done on time include:

  • Developing a realistic study scope that is both responsive to the legislature's policy needs and attentive to the time and staff resources available to carry out the study.
  • Scoping a project by focusing on the report due date, then "backtracking" in our estimates of the time and resources needed to accomplish each of the major study phases.
  • Developing an overall schedule as part of the study work plan that includes, as necessary, key milestone dates.
  • Frequent team meetings throughout the study process to check progress on, and/or problems with, meeting the schedule outlined in the study work plan.
  • Requiring individual auditor monthly status reports to management.
  • Use of a "project management" checklist to check-off key project milestones as they're met.  This is a joint responsibility of both the audit team and the project supervisor.
  • Bringing any matters that may cause delays to the attention of team leaders, staff directors, and, if necessary on major issues, to Committee legislators.
  • Use of "storyboard sessions" and/or writing one-page summaries, prior to writing the first draft, in order to help focus the report writing phase.

Continuing to experiment with succinct versions of our studies, audits and reviews to facilitate the completion of our reports within a reasonable timeframe.
If it becomes clear, despite our best efforts, that the original deadline cannot realistically be met, we will extend the deadline if it is possible to do so.  These "extensions" are rare, and are usually only for one month.

From: Ken Levine, Texas Sunset Advisory Commission

  • Good planning of resources and time in the first place (however we all know there is never enough of either...)
  • Track and ensure teams meet major milestones
  • Close, but helpful, supervision.  Supervisors must know if a project is not going to meet a major milestone and be able to step in and help the project manager get the project back on track.
  • Quality, timely editing of drafts, particularly the developmental edit of a first draft where the content is analyzed for thoroughness, other approaches etc.  I guess I'm saying to find the major holes or problem areas as early as possible in the drafting of a report.
  • Make sure that timeliness is one of the significant elements of performance used in personnel evaluations.

From: Gary Brown, Michigan

In Michigan, we have not yet discovered the secret for finishing projects on time.  We do use a detailed project budgeting and project tracking system.  We find it to be of considerable value, both in attempting to finish projects on time and to plan workflow and staff assignments.  However, it certainly does not result in all projects being completed on time!

From: Joel Alter, Minnesota

The large majority of program evaluations done by our office are issued during the early weeks of the legislative session—typically in January or February.  Reports issued later in the session may miss their opportunity to have an impact, and reports issued when the Legislature is not in session may be “old news” by the time the Legislature meets again.

Aside from setting milestone target dates for key points in each study, our project scheduling approach is fairly informal.  The Deputy Legislative Auditor determines how many staff to assign to each project based on a general notion of project size.  Most of our projects start and end at roughly the same time, so a project with four people assigned to it should be able to accomplish a lot more than a project with two staff.  We track the total hours consumed by each project (and by several project phases, such as “research” and “report preparation”)—but project deadlines are the main tool used to track project timeliness, rather than comparisons of projected and actual hours used.

Each project manager sets target dates for the project workplan, outline of findings, internal draft, closing conference, and report release.  Also, each manager prepares biweekly status reports for the Legislative Auditor and Deputy Legislative Auditor.  These reports indicate activities of project staff in the previous two weeks and priorities for the upcoming two weeks—as well as any changes in the key target dates.  Project staff meet monthly with the Legislative Auditor and Deputy Legislative Auditor to discuss the status of the project and its schedule.

Our support staff prepare a timeline for each manager—working backwards from the release date to the date when project staff have completed an internal draft.  This timeline includes interim deadlines for the last few weeks of the project—including things like when sidebars should be inserted in the report and when mailing labels should be prepared for report distribution.

Despite these efforts, we have still had mixed success issuing reports that meet our internal deadlines.