Time for timesheets.. or wait until tomorrow?

If you want to get control of your projects, you should use timesheets. Or should you? Is the accepted wisdom really wise?

Timesheets allow you to measure effort; i.e. man-days of work. All project management methods suggest you should control time and cost. Human effort one dimension of cost. It is normally expressed in man-hours or man-days of work (which usually can be  converted into a financial cost). To control effort, you need to estimate it when you plan and then track the actual effort. That’s where timesheets come in, to track the actual effort.

The Prince2 project management method follows this approach. The Prince2 planning process has an estimating step. For Prince2, effort is the primary resource estimation for each task. (Secondary resource needs include equipment, travel or money). By saying that you should control effort, Prince2 ask you to plan and then to track effort. Prince2 therefore assumes that you will use timesheets.

That’s what they are doing at one major bank using Prince2. Effort is indeed the main measure of project resource estimation. A small banking project may have a budget of 50 man-days of effort, whereas a large project may have a budget of 3000 man-days. To monitor progress, the bank uses timesheets to record actual effort by the project teams. So, for example, a 10-week project estimated at 50 days may finish with 45 (or 55 or whatever) actual days of effort. This is calculated from the timesheets of all the project team members submitted during the 10 weeks.

So timesheet solutions aggregate a lot of data from a lot of people. You need timesheet software, which lots of people need to use correctly and regularly. That’s where the problem starts. The software vendors will tell you that the software is easy to use (maybe true, maybe not!), but they won’t always tell you how much work is needed to achieve a sustained, widely-used solution which truly delivers benefits to your wider organization.

In some organisations, there can be hundreds of people working on various projects. A large hospital tried to implement a timesheet solution to control project effort.  Inside the IT department of about 50 people, this worked OK.  But most projects go beyond IT. The hospital has about 3000 staff, very many of whom could work on a project. You can’t ask 3000 people to fill in timesheets, just because some of them may work on a project from time to time.

Even for the small IT department, the hospital hit another problem, of scope creep. Once they had estimates of project effort for their teams, they had a partial forecast everyone’s future work plans. So they decided to go for a 100% solution, and complete the picture with plans on non-working time (holidays, etc.) and non-project work (support, maintenance, staff meetings, etc.). Once you try to manage the full working week for your entire team, you add extra complexity for little benefit. The hospital could not sustain the effort, and the timesheet solution failed.

There are a lot of failures with timesheet implementations. It’s hard work and there can be a lot of resistance. People don’t like filling in timesheets. It takes a lot of work to get everyone to submit a time sheet each week (and to sustain this 52 weeks per year). Furthermore, if you succeed, the data quality may be low, and consequently the benefits for project control may be slim.

Timesheets can work well in certain environments, for example where hours are billed to the customer. Or in closed environments, such as software development teams, where all projects are done by a limited number of project team members. In these cases, there is a clear benefit to the organization, so you can more easily succeed with timesheets.

If you choose not to implement timesheets, what can you do? One major chemicals company has decided not to use timesheets today for their projects. They know it’s premature – timesheets wouldn’t work today. Their solution is to track cost and time, but not effort.  Costs are tracked (expenditures); time is tracked (activity start and finish dates); but effort is neither estimated nor tracked.

So should you use timesheets? If your goal is to improve your team’s or your company’s project management maturity, yes, it should be on your long term agenda. Should it be a priority? You need to analyse timesheet deployment as a long term project. It’s not a quick-win software implementation project – you need sustained long-term commitment to get any benefits. Work out the business case, study the risks of failure.

You’ll possibly find that deploying timesheets is not a priority project. Your return on investment is perhaps too uncertain, your risk too high. Focus instead today on easier project management improvements with more certain benefits. Wait a while until you tackle timesheets. Mañana, demain, morgen, tomorrow.

Can you measure your project performance?

The Dutch say “Meten is Weten” – to measure is to know. Measurements have become fashionable everywhere. In companies, in schools, in hospitals … everything is driven by targets and measurements. Everyone is trying to deliver their targets and key performance indicators (KPI’s).

The fashion has come to the world of Project Management. However, in Project Management, targets are used slightly differently. Rather than set an arbitrary target, you create a plan. To measure, you compare your plan against reality (your “actuals”). Project performance measures how well you are delivering against your plan.

The traditional performance measurements for projects have always been cost and time. There’s even an acronym OTOB which neatly sums up the Project Manager’s drive to deliver on time, on budget. When people speak about project performance, they typically are talking about OTOB: are you delivering on time, on budget, relative to your original agreed plan?

If OTOB is the traditional measurement of Project Performance, modern Project methods have followed the fashion, and added more measurements. Some people have added a third dimension, Quality, and transformed OTOB into OTOBOQ.

The latest Prince2 manual (2009) follows the fashion. The subject of measurements arrives in Chapter 1, which introduces 6 key aspects of performance (whereas in the previous version it was buried in a later chapter). Prince2 suggests that the best practice is to control not just Time and Cost, but also Scope, Quality, Risk and Benefits.

In reality, you may need to go a little further: you may need to break Cost into Financial cost and Effort, as these must be controlled very differently. Financial cost is based on expenditure of money (purchase orders, invoices…), while Effort is man-days of work. So you have up to 7 aspects of performance to control. For each, you need to compare plan versus actuals

❑ Time
Time is the easiest measurement. Your plan is probably a schedule of work (typically a Gantt diagram), including milestones (and for Prince2, stage boundaries). So this plan gives your planned time. As the work advances, you track actual time using the dates that the work was actually finished.

❑ Financial Cost
Financial cost data is also quite easy to measure. You can estimate your planned costs in various ways. Some people estimate from the Gantt diagram; others collate the information in a Business Case. You should measure actual costs as soon as you commit to expenditure (e.g. when you raise a purchase order). Be proactive about recording actual costs: don’t wait until the invoices are recorded in your accounting system, and certainly don’t wait until they are paid. 

❑ Effort
Effort is mandays of work. Effort can be easy to estimate, but is often difficult to measure. Using a Gantt diagram, you can estimate effort in man-hours or man-days. This gives your planned effort. Effort is difficult to measure, because most organisations don’t have time-sheet systems (or if they have them, they are not used by all project contributors). And, as it’s surprisingly difficult to deploy time sheets solutions (i.e. to get enough people to use them week by week), measuring actual effort is also surprisingly difficult. (For many projects, it’s effectively impossible).

❑ Scope
Scope used to be hard to control, but with modern methods, it’s a lot easier. If you use Prince2’s product based planning, you can predict your planned scope based on the product decomposition. It’s easy to detect when the products are delivered, so you can easily track actual scope. (It’s not just Prince2 that helps. Some Agile methods, such as Agile Atern, also are making control of scope easier: Atern uses a Prioritised Requirements List).

❑ Quality
Prince2 also makes it easier to control quality. Using Prince2’s product based planning technique, you can do simple-but-effective quality planning, giving your planned quality. As the work is done, each planned quality test is done, which gives your actual quality.

❑ Risk
Risk is typically controlled though regular estimates throughout the project, rather than by measurements. Each time you plan (or replan), it’s wise to perform a risk analysis. This gives your forecast of risk – your planned risk. When you execute your risk counter-measures, you mitigate the risk, and lower your forecast risk level. To measure your actual risk is difficult, and probably impossible. The best you can do is to revise your estimates, using regular risk analyses.

❑ Benefits
Benefits are also typically controlled through regular estimates, rather than measurements. In many projects, you cannot measure the benefits, as they are realised after the project has closed. Your Business Case should give the planned benefits. If you can’t measure actual benefits, you should review estimates regularly: using Prince2, you will review your Business Case at the end of each stage (and the associated Benefits Review Plan).

So for Project Managers, “Meten is Weten” (to measure is to know) is a little simplistic. You need to control project performance, but you can’t measure everything.

Certainly Prince2 helps you to "know" your project -  it gives you a framework for controlling your performance – and that helps with boosting your  performance. Give it a try!