Why is this required?
Dr John McManus and Dr Trevor Wood-Harper write about a study conducted from 1998 to 2005, covering 214 information system projects, across 10 sectors from Manufacturing to Agriculture, of value ranging from less than a million euros up to 80 million euros. The intention of the study was to find out at which stage does a project has greater chances of failing, and the reasons for abandoning projects.
Out of 214 projects, 51 (23.8%) were cancelled.
Their findings state that “Results from the analysis of cases indicates that almost one in four of the projects examined were abandoned after the feasibility stage of those projects completed approximately one in three were schedule and budget overruns.”
It pays rich dividends to the project manager to spend good amount of time and effort on cost management.
Cost management is practiced throughout project duration. Why? Because some of the costs are seasonal, while some are inflation-related. Sometimes what you calculated during planning phase may not hold good when it is time to make use of these resources on the project. Also as project progresses you get to know more information which may help you prepare concrete estimates.
At the beginning of project, confidence of accuracy of estimates is less. Hence you will use a rough order of magnitude (ROM) of about -25% to +75% on your estimate numbers. As you get more information and feel confident of better accuracy of estimation you narrow the range to -5% to +10%. Always use higher range on the positive side because the probability of under-estimation is more than over-estimation.
You need to consider lot of ‘plans’ for estimating project costs
Cost management plan, of course. This is the document that says how costs are calculated and managed.
Human resources management plan – people are central to most of the projects. A team needs good mix of skillset for optimal costing – for instance, a software development project may have a workforce pyramid consisting of 5 junior engineers, 2 senior engineers and a technical lead. It is crucial to refer to human resource plan and estimate costs. You would also need to consider additional costs such as training cost, rewards and recognition costs, back-filling cost to manage attrition.
Project scope baseline – helps you to figure out the sources of direct costs. The indirect costs, if you decide to consider, will be common expenses such as administration, facilities, and operational expenses of running the organization, broken down to team’s level.
Scope baseline usually contains the following components –
- Scope document – contains product description, acceptance criteria, assumptions and identified risks
- Work breakdown structure (WBS) – hierarchical structure of scope broken down to lowest level, and related information
Project schedule – gives you schedule activity resources and durations to calculate costs. As noted in Estimate Activity Resources process, cost estimating is to be closely coordinated with it – some of the resource costs may change and need to be constantly looked at to assess impact on budget.s
Risk register – this is a big aspect of costing. Project management is as much of anticipation of ‘unknowns’, risks and mitigating them as managing the ‘knowns’.
There can be negative risks as well as positive risks on a project. Negative risks are threats, and positive risks are opportunities. As an example, a takeover bid of performing organization by another company may be perceived as a threat, but can be an opportunity as well. It may become easy to get skilled resources from the resource pool of the new entity.
And of course, all governed policies, templates, and lessons from earlier projects are considered.
How do you do it?
Expert judgment combined with historical information is very valuable in assessing costs taking into account cost fluctuations around resources. When different methods used to estimate costs return different numbers, you would use expert judgment to analyze and decide which one to consider.
Types of estimates
Analogous estimating – helps you consider previous similar projects, look at their costs and prepare estimates. This is a quick tool, less accurate and useful only when you are comparing projects of pretty similar size and type.
Parametric estimating – you use one of formulae driven spreadsheets, software, or tool to put in parameters specific to your project and let it derive cost numbers.
Bottom-up estimating – is like using Divide and Rule method. Helps you in estimating tasks that are big and complex. You can break them down into simpler, smaller pieces and put cost estimates for these pieces. Then you roll up the cost numbers and get the final cost for the original task.
Three-point estimating – This simple PERT tool is a weighted average of worst case, most-likely and best case estimates to have realistic numbers. Recall that we used this in Estimate Activity Durations process. Again, there are two formulae based on which distribution of estimates are assumed.
If Beta distribution is assumed –
Expected Cost = [Optimistic Cost + (4 x Most likely Cost) + Pessimistic Cost] / 6
cE = [cO + 4cM + cP] / 6
If Triangular distribution is assumed –
Expected Cost = [Optimistic Cost + Most likely Cost + Pessimistic Cost] / 3
cE = [cO + cM + cP] / 3
Reserve analysis – we saw this in Estimate Activity Durations process. There are two types – contingency reserve and management reserve. Contingency reserve is also called contingency allowance, where you add some amount of cost to address uncertainty on known risks. For unknown risks such as discovered tasks that are very much part of project scope you have management reserve. Reserve analysis is a very useful tool that helps you manage uncertainty. This could be applied at activity level, WBS component level, or even milestone level.
In short, contingency reserve is for “known unknowns” and management reserve is for “unknown unknowns”.
Cost of Quality (CoQ) – is the cost incurred on the effort to ensure that the output produced matches against the requirements.
This includes cost of activities performed to ensure that good quality product is produced (training, testing, equipment) AND the cost of activities performed as a result of producing bad quality product (rework, scrap).
You can see more details in Plan Quality process.
Vendor bid analysis – when you decide to contract out certain amount of work, you would talk to vendors, float request-for-proposal (RFP), look at vendor responses, award contract, measure their output, pay them – all this would cost that need to be estimated for. The details of these activities are part of Project Procurement Management Knowledge Area.
What do you get?
Cost estimates of tasks (or activities) – list of probable cost estimates for all activities on the project. We say ‘probable cost estimates’ because estimating costs is not one time activity. It has to be done throughout the length of project. And each time it is done the activity cost is refined. All supporting assumptions, reasoning and documentation is also included.
These will be used to create project budget. We shall see this in the next lesson.