Project Cost Overruns!!


Posted on 8th December, by JimYoung in Blog. Comments Off on Project Cost Overruns!!

Projects cost overruns can affect our organisation’s performance, inhibit the achievement of strategic goals, and even jeopardise our credibility as project managers.

Some organisations don’t have data on project completion costs perhaps because their accounting system doesn’t allocate expenditure to projects, the invoices from suppliers often roll across several projects, or because going back over past performance to collect the data is a bother.

If a project costs more than is budgeted for and then needs more funding, other projects may be put on hold through lack of funds. On the other hand, if we have budgeted far more than the actual cost of the project, funds are tied up that could have been used elsewhere. It is also possible that a number of proposed projects may have been rejected because the costs were greatly over-estimated and thus the opportunities they presented were lost.

In deriving a budget, we typically build up a spreadsheet listing the individual cost items at some suitable level of detail, and assigning each item a best guess cost. These best guesses are added up to create a budget figure. However, this can be a very optimistic figure that has not allowed for anything to go wrong, and if it’s a big, novel and complex project then there will be plenty of things that might go wrong, and almost certainly something bad will go bump-in-the-night and add to the cost. In risk analysis, we can describe uncertainty with a probability distribution, typically like the one shown below:

Untitled

Consider for a minute an investment some of us may make – the renovation of a house. Let’s say that we’ve just bought our dream home, and had an architect come round to assess the renovation work. The architect estimates the work will cost $120,000 – if all goes well. We should not hope to pay much less than that figure because the costs are itemised and the estimates are built up from standard prices. But if the house is old, it may be hiding many secrets (uncertainties) – leaks, dry rot, poor ventilation, unstable foundations, broken beams, and so on. Thus the price we end up paying could be far greater than that original $120,000.

Generally, we design a project to be as cost-efficient as possible, so there is not a great deal we can do to achieve a smaller cost, so the minimum cost (‘Lowest’) will be quite close to our ‘Best Guess’. In contrast there will be many things that could drive the cost up, and the maximum cost (‘Highest’) can be far more distant from the ‘Best Guess’. In these circumstances there is considerable chance that the cost will exceed the ‘Best Guess. ’ It is little surprise, therefore, that budgets built this way – without an allowance for risk – often get exceeded. To more accurately estimate cost and to allow for risk, the following calculations may help:

First apply PERT (Programme Evaluation and Review Technique), which is a weighted average formula that takes the lowest estimate (Optimistic or O), the best guess (Most Likely or L), and highest (Pessimistic or P) estimates into account:

PERT Estimate = (O + 4L + P)/6

Probability theory tells us that L (one standard deviation either side of the norm) has a 68% likelihood of occurring, and P and O each have a 16% likelihood of occurring, which is why L has four times the weight of O and P.  For example, consider the weekly grocery shop – worst case or P might be $300 cost, best case or O might be $100, but experience tells us that the most likely cost will be $240, thus:

PERT Estimate = {$120 + (4 x $240) + $300} / 6 = $230

The resultant PERT Estimate is an accurate estimate insofar as actual cost is equally likely to be more or less – it provides us with a 50:50 chance of on-budget completion. However, if we wish to be more than 50% confident of completing our project within budget, a contingency sum needs to be added to allow for this risk. While there are a variety of ways to determine the size of this budget contingency, one method is explained here:

Untitledd

Determine costs to the nearest whole number that will provide the following levels of confidence for achieving on-budget project completion:

Untitledd - Version 2

The same methods of calculation can be applied to time estimates. The Standard Factor is fixed for all projects, but if you are interested in its derivation please check out my free book “Managing Murphy” at http://www.skillpower.co.nz/wp-content/uploads/2014/03/Managing-Murphy-PDF.pdf

Another factor that can blow the project estimate is company culture. There are many different manifestations of this. For example:

  • A sponsor underestimates the costs because they really want the project approved and believe in it, even if the numbers would say otherwise.
  • Project costs are sometimes deliberately under-estimated to get them started.
  • A tenderer submits a low bid in order to win the contract, but then makes their profit by charging huge amounts for any exceptions or variations (that it often knows will be needed at the time of bidding).
  • Management removes a percentage from the cost estimate because the estimators 
‘always exaggerate’. Thus, estimators build in fat in anticipation of inevitable budget cuts by management.
  • A department overestimates because it could use the extra cash to prop up other projects that are over budget or likely to be so.
  • A department overestimates because it looks good or gets a bonus for coming in under budget.
  • A department over or under-estimates because it doesn’t have a lot of experience, detailed knowledge about the project or is given insufficient time to determine costs, but is still required to provide a budget.

The following checklist of ideas may help improve our cost estimating ability:

  1. Before I undertake an estimating job I always find out the level of accuracy required for the estimate.
  1. Before I prepare an estimate I always clarify the project outcomes and scope of work.
  1. I always include contingency in my estimate to allow for predicted risks, but never to compensate for my lack of estimating aptitude.
  1. I always compare my estimates with actuals in order to validate or improve my estimating technique, assumptions and database knowledge, and I don’t react defensively.
  1. When I have insufficient time to complete a comprehensive assessment, I focus on the bigger chunks of work, recognising the wisdom of the Pareto Principle or 80:20 Rule.
  1. My completed estimates are always accompanied by documented assumptions on how they were derived. I don’t make assumptions when the facts are available, and my assumptions are always reasonable (ie, low-level risks).
  1. Should my client wish to alter the project scope, I always advise them about the impact of such changes on the project estimates of time and cost, and don’t undertake the change until my client formally approves the additional estimated expenditure.
  1. I base my estimates on normal numbers of people completing each task, and don’t overload the job or plan that my team work overtime, weekends or statutory holidays.
  1. My estimates are always accompanied by an indication of their accuracy (range or percentage), and a date until when the estimate remains valid.
  1. I proactively seek feedback about the accuracy of my estimates, avoid contributing poor results to ‘uncontrollable situational factors’ and good results to my own brilliance, and attempt to uncover the true cause of my estimating errors in order that I might continuously improve.
  1. Before I undertake an estimating job I always find out the level of accuracy required for the estimate.
  1. Before I prepare an estimate I always clarify the project outcomes and scope of work.
  1. I always include contingency in my estimate to allow for predicted risks, but never to compensate for my lack of estimating aptitude.
  1. I always compare my estimates with actuals in order to validate or improve my estimating technique, assumptions and database knowledge, and I don’t react defensively.
  1. When I have insufficient time to complete a comprehensive assessment, I focus on the bigger chunks of work, recognising the wisdom of the Pareto Principle or 80:20 Rule.
  1. My completed estimates are always accompanied by documented assumptions on how they were derived. I don’t make assumptions when the facts are available, and my assumptions are always reasonable (ie, low-level risks).
  1. Should my client wish to alter the project scope, I always advise them about the impact of such changes on the project estimates of time and cost, and don’t undertake the change until my client formally approves the additional estimated expenditure.
  1. I base my estimates on normal numbers of people completing each task, and don’t overload the job or plan that my team work overtime, weekends or statutory holidays.
  1. My estimates are always accompanied by an indication of their accuracy (range or percentage), and a date until when the estimate remains valid.
  1. I proactively seek feedback about the accuracy of my estimates, avoid contributing poor results to ‘uncontrollable situational factors’ and good results to my own brilliance, and attempt to uncover the true cause of my estimating errors in order that I might continuously improve.
  1. Before I undertake an estimating job I always find out the level of accuracy required for the estimate.
  1. Before I prepare an estimate I always clarify the project outcomes and scope of work.
  1. I always include contingency in my estimate to allow for predicted risks, but never to compensate for my lack of estimating aptitude.
  1. I always compare my estimates with actuals in order to validate or improve my estimating technique, assumptions and database knowledge, and I don’t react defensively.
  1. When I have insufficient time to complete a comprehensive assessment, I focus on the bigger chunks of work, recognising the wisdom of the Pareto Principle or 80:20 Rule.
  1. My completed estimates are always accompanied by documented assumptions on how they were derived. I don’t make assumptions when the facts are available, and my assumptions are always reasonable (ie, low-level risks).
  1. Should my client wish to alter the project scope, I always advise them about the impact of such changes on the project estimates of time and cost, and don’t undertake the change until my client formally approves the additional estimated expenditure.
  1. I base my estimates on normal numbers of people completing each task, and don’t overload the job or plan that my team work overtime, weekends or statutory holidays.
  1. My estimates are always accompanied by an indication of their accuracy (range or percentage), and a date until when the estimate remains valid.
  1. I proactively seek feedback about the accuracy of my estimates, avoid contributing poor results to ‘uncontrollable situational factors’ and good results to my own brilliance, and attempt to uncover the true cause of my estimating errors in order that I might continuously improve.
  2. Before I undertake an estimating job I always find out the level of accuracy required for the estimate.
  1. Before I prepare an estimate I always clarify the project outcomes and scope of work.
  1. I always include contingency in my estimate to allow for predicted risks, but never to compensate for my lack of estimating aptitude.
  1. I always compare my estimates with actuals in order to validate or improve my estimating technique, assumptions and database knowledge, and I don’t react defensively.
  1. When I have insufficient time to complete a comprehensive assessment, I focus on the bigger chunks of work, recognising the wisdom of the Pareto Principle or 80:20 Rule.
  1. My completed estimates are always accompanied by documented assumptions on how they were derived. I don’t make assumptions when the facts are available, and my assumptions are always reasonable (ie, low-level risks).
  1. Should my client wish to alter the project scope, I always advise them about the impact of such changes on the project estimates of time and cost, and don’t undertake the change until my client formally approves the additional estimated expenditure.
  1. I base my estimates on normal numbers of people completing each task, and don’t overload the job or plan that my team work overtime, weekends or statutory holidays.
  1. My estimates are always accompanied by an indication of their accuracy (range or percentage), and a date until when the estimate remains valid.
  1. I proactively seek feedback about the accuracy of my estimates, avoid contributing poor results to ‘uncontrollable situational factors’ and good results to my own brilliance, and attempt to uncover the true cause of my estimating errors in order that I might continuously improve.
  1. I consistently apply a standard and valid estimating process, which I update with experience.
  1. I’m sensitive to my estimating biases and common mistakes, and always seek colleagues’ opinions, appreciating several viewpoints provide some safeguard against my excessive optimism or pessimism.
  1. I’m familiar with the use of the following tools and techniques which can help me with estimating:
  • – work breakdown structure (WBS)
  • – trade-off analysis
  • – network diagramming
  • – sensitivity analysis
  • – expected monetary value (EMV)
  • – learning curve
  • – parametric modelling
  • – phased estimating
  • – PERT formula
  • – Delphi technique
  • – wide-band Delphi technique
  • – estimating spreadsheet
  • – bottom-up and top-down techniques.
  1. I realise that a person doesn’t work at 100 percent productivity and 70-80% is a more realistic expectation.
  1. I understand the difference between elapsed time, duration and work effort.
  1. I can determine the true cost of employees’ labour and express this as an hourly rate for accurate work effort costing purposes.
  1. I realise that resources who work multiple projects take longer to complete tasks due to time lost switching between projects.
  1. I realise that adding resources will not necessarily reduce duration. There is a diminishing return.
  1. I realise that only by reducing the duration of critical tasks will I complete my project earlier.
  1. Wherever practicable I always involve in the estimating process the people who will actually do the work.
  1. I always revise my estimate whenever the following occur:
  • – scope changes (variations)
  • – risks events happen
  • – assumptions prove to be wrong
  • – new completion dates advised
  • – deliverable specification changes
  • – labour or material costs escalate
  • – exchange rates alter.
  1. I always allow myself sufficient time to undertake proper estimating.
  1. I always undertake estimating with a colleague or invite them to check my figures.

Beware too of project scope creep (also called requirement creep, function creep and feature creep), which refers to uncontrolled changes or continuous growth in a project’s scope. This is more likely to occur when the scope of a project is not properly defined, documented or controlled. The consequence is that the additional requirements can overwhelm the capacity of the resources allocated to the project resulting in the project missing deadlines, budgets or complete failure. The best way to avoid scope creep is a very clearly defined contract that identifies exactly what is to be done (and perhaps not done) and to what standard by when at what cost, including labour and material costs for possible variations. Invariably the scope of a project will change and such variations need to carefully costed and extra budget (and a later completion time) approved where appropriate before implementing the change. Some unscrupulous contractors, who rely on variations to make a profit, will welcome our poorly defined project work, although such extra work can be separately contracted.

To solve over-expenditure more easily, detect such variance early. If costs are exceeding our project budget one or some of the following ideas might help:

  • – Check schedule progress. Ahead of schedule work incurs early unexpected expenditure.
  • – Renegotiate project budget, work scope, specifications, pay rates and material costs.
  • – Minimise order quantities and/or purchase economical order quantities (EOQ).
  • – Sell off excess inventory for immediate cash injection. Avoid stockpiling.    Practice JIT.
  • – Substitute cheaper processes, labour, materials and equipment.
  • – Eliminate advances, deposits, rework, wastage, theft, spoilage and scope creep.
  • – Have more progress payments from client.       Delay own payments (ie, improve cash flow).
  • – Optimise schedule/budget.   An ambitious schedule is usually an expensive schedule.
  • – Shun perfection and extra audits. Curb ultra-perfectionists. Settle for ‘good enough’.
  • – Ensure all charges against project are accurate, legitimate and properly authorised.
  • – Review delegated financial authorities.  Centralise financial approvals.  Vet all purchases.
  • – Ensure payments are only made for satisfactory completed work. Apply retentions.

 

 





Comments are closed.



Blog

My latest thoughts on Project Management and life.

Business Analyst (BA) and Analysis

To survive or preferably thrive, organisations cannot stand still. They must evolve and progress to remain alive. Such evolution and progress requires projects. And...

Free Book: “Managing Smaller and Medium-Sized Projects” by Dr Jim Young PMP

I acknowledge with sincere appreciation the contributions to this book from numerous clients, student, colleagues, friends and project management practitioners. Their sometimes unwitting input...

The “Lean” Project

Lean is an often-used adjective in business these days, but there’s some confusion over its exact definition. In essence, the goal of Lean is...