Effort estimation in project management
These are the most common methods
Effort estimation is probably one of the most common tasks of a project manager, but not one of the easiest. Thanks to many uncertainties, it is difficult to say how much effort will be required for a task, and it often turns out differently than expected. Well-known examples of faulty effort estimation are the Elbe Philharmonic Hall in Hamburg or Berlin’s BER airport. While the estimation can still be made relatively accurately for similar projects, the situation is completely different for new kinds of projects. However, miscalculations are not only dangerous for large projects, but can also cause small projects to fail.
What is an effort estimate?
An effort estimate is used to determine the likely effort of a project. It also helps to assess the feasibility of a project and thus make a decision for or against a project. Important parts of this are capacity, resource, time and cost planning A procedure for estimating effort should be simple to perform, user-friendly and as accurate as possible.
Why are effort estimates needed?
Effort estimations are usually imprecise and can be very flawed. Nevertheless, at least a rough effort estimate is important for a project, for example to set and apply for the project budget as well as to define an approximate time frame for the project.
>>Our experience shows that in practice it makes sense to carry out the effort estimation only up to a certain level of detail, but to review and adjust it regularly.<<
Walter Epple, Senior Consultant
In order to obtain the most accurate estimate possible, different estimation methods can be used. For example, estimation methods differ in agile and classic project management. It is important that the effort estimation does not only take place at the beginning of a project, but is continuously updated, as this makes the estimation more and more accurate for the further course of the project.
In the top-down approach, the client makes a rough estimate at the beginning of a project, which the project team is supposed to follow. This method is therefore based on the client’s experience and is very easy to carry out. However, the estimations are usually very inaccurate because the client cannot oversee every necessary step of the work.
The bottom-up approach is more complex and meaningful than the top-down method. Here, the individual tasks are evaluated and their expected duration as well as the required resources are determined. This makes this method more accurate. However, it is possible that those who are responsible for the project plan the duration and costs a little more generously in advance and that the project is thus estimated incorrectly.
3. Expert estimation
As a project manager, you usually do not know every project-relevant topic equally well. Therefore, you can get an estimate from someone who is an expert in the respective field and can therefore estimate the effort of a task well.
Our tip: In expert estimation, it makes sense to ask several experts, from whose estimates an average value is then taken. To improve the quality of the estimate, one of the experts should work on the project if possible.
Other variants of expert estimation are:
- PERT Three-point estimation
In this method, three values are estimated based on experience: an optimistic, a pessimistic and a probable value. The expected value is then determined from these values. Since this is a very simple and quick method, it is particularly useful in early project phases.
- Delphi method
The Delphi method is based on the bottom-up approach. However, several experts are consulted for the estimation. They do not discuss the results in advance and give their estimates in secret. If the results differ greatly, they are discussed in order to find a consensus. Afterwards, they estimate and discuss the results again. In this way, the estimation is relatively accurate, but requires several experts over a longer period of time.
- Planning Poker
Planning Poker is similar to the Delphi method. Here, too, experts give hidden estimates. If they differ greatly, however, there is no discussion afterwards. Instead, the experts with the highest and lowest estimates explain their reasons for their respective decisions. Afterwards, all experts give a new estimate until a consensus is reached among the experts.
4. Analogy method
In the analogy method, the effort is derived from similar projects that have already been completed.
A variant of this is the relation method:
The relation method is also based on empirical values of similar projects. In addition, however, fixed guidelines and factors are defined that also flow into the calculation. These are, for example, specific requirements for a product or the experience of employees.
5. Multiplication method
In the multiplication method, task packages are formed that are carried out several times. The total effort is then determined by multiplying the number of packages by their estimated effort. This method is therefore particularly suitable if individual tasks of a project are frequently repeated.
6. Weighting method
This method determines which factors have which influence on a project. The total effort is then determined with the different weightings.
7. Percentage method
With the percentage method, the current actual effort for a project is taken as the basis for calculating the effort for subsequent phases. Alternatively, the effort for a project phase estimated with another method can be used as the basis for the calculation. This makes this method applicable at an early stage.
8. AAlgorithmic estimation methods
Algorithmic estimation methods are often used in later phases of projects, as detailed knowledge of the project must be available for their application. These methods include:
- Function Point
- Data Point
- Object Point
- Cocomo (Constructive Cost Model)
Selecting the appropriate estimation method
Which estimation method is used depends on the one hand on the current progress of the project, but on the other hand also on the time available as well as on available experts. Therefore, the choice of a suitable estimation method should always be made individually. For early, rough estimates, the analogy, relation and percentage methods are suitable. The weighting and multiplication methods, on the other hand, are more suitable if the influencing factors of a project are precisely known. If appropriate experts are available, an expert estimate can be made at any time. Depending on the method used, however, it is more time-consuming to carry out. Algorithmic methods are particularly suitable in later project phases, as they require detailed knowledge.
Effort estimates are time-consuming, often imprecise and mostly dependent on experience. Nevertheless, they are needed to be able to estimate at least roughly how complex a project will be and how much time its implementation will take. In this way, it is also possible to check during the implementation of the project whether the tasks will be completed according to plan or whether the project will exceed the planned effort.
A project management software keeps track of all tasks of a project – regardless of whether you plan classically with a Gantt chart or use a Kanban board. You can assign a planned effort to these tasks and record the efforts already made during the implementation of the project. With a plan-actual comparison or an earned value analysis,you can quickly determine whether the planned effort has been exceeded and in this way make a forecast about the further course of the project. This makes your estimates more accurate during the implementation of the project.
More about the project management software myPARM:
Would you like to get to know myPARM in a demo presentation? Then make an appointment with us right away!