Poor estimation is a major cause of project failure.
According to the PMI, inaccurate cost estimates is a contributing factor in 28% of projects that fail.
Many project professionals have first-hand experience in how devastating it can be to underestimate a project. Perhaps you have too. When an estimate has been delivered, clients and executives will often see it as binding and will expect the project to track to it. If it later turns out that the project was underestimated, it could be classified as a failure even if the project delivers exactly what the customer wanted.
As a leader of strategic projects, you have to mitigate the risk of cost-overruns as you don’t want to deliver an excellent product or service that ends up being labelled as a failure because it was underestimated, or because it no longer makes good business sense. If the project has been underestimated, it changes the business case and it may be that the project should never have been undertaken.
Producing Realistic Estimates for Strategic Projects
To a large extent, good estimation comes down to how honest and realistic you are in your outlook. You have to be careful not to be overly optimistic, but you also cannot be too pessimistic, as that will overinflate the estimate and undermine the process. Many teams look at the sunny-case scenarios and only provide best-case estimates. They fail to consider the full spectrum of activities that need to be carried out and often leave out contingency to cater for risks and uncertainty.
Years ago I was running a large technology project in the city of London.
I tried very hard to produce a realistic estimate and integrate all the lessons learned from past projects. We first broke down all the work (i.e. analysis, design, build, testing and deployment) in as much detail as possible.
We discussed what each feature entailed so that the team could estimate it accurately. The people who provided the estimates were those who would actually be carrying out the work and had the knowledge to do it.
When we received the estimates, we added a buffer to each item as some team members might have given us optimistic estimates.
At the end, we added further 30% contingency to the overall estimate to account for any unforeseen events or misunderstood requirements. We then converted it into a monetary value, which was relatively easy given that we had ring-fenced team members and knew everyone’s day rate.
With this approach, we honestly believed that we had nailed it and that we had produced a realistic estimate. But what do you think happened?
Most of the work took A LOT more effort to complete than expected. This was a bit of a blow because we had put so much energy into getting it right.
Why Estimations Go Wrong
So, what went wrong?
I would say that it was a combination of factors. What’s important when estimating is that we know what we’re going to build – in quite a bit of detail. How else can we estimate it?
On my project we did have a pretty good idea, but not good enough. Since we were operating in a very complex business domain, not all the requirements were understood. So although we added 30% contingency, we didn’t add anywhere near enough to cater for all the uncertainty.
To a large extent, good estimation comes down to how honest and realistic you are in your outlook
Was there something more we could have done to understand better what we were going to build?
I believe we could have spent more time in workshops fleshing out the requirements and gaining a detailed understanding of what the solution looked like. I also believe that one of the best methods for producing a realistic estimate is to prototype the solution. Prototyping removes a large part of the uncertainly and provides the team with working knowledge of what it needs to build. We didn’t do that.
The Impact of Culture on Estimating
In some organisations, there is a culture of always challenging the estimates and reducing the numbers that the team arrived at.
There may also be a lack of openness, respect and honestly which means that information about project risks and worst-case scenarios is not welcome and therefore suppressed. In these circumstances, the team may be able to estimate accurately, but the information isn’t surfacing. Domineering managers who work to their agenda and think they know all the answers is a big inhibitor to realistic estimation.
On my project, I don’t feel there was an agenda of keeping the estimates down. But I do think something else was going on whereby team members didn’t feel accountable to the estimates they submitted.
If a person doesn’t feel accountable, he or she is going to ask fewer questions and put a lot less effort into the estimation.
Changing this attitude won’t happen by telling team members to be more accountable. It will only change through collaborative working, trusting one another, having a stake in the successful outcome of the project and by caring about each other.
Looking back I’m convinced that more could have been done to build the team, to create a collaborative spirit where everyone was in it together and where everyone felt accountable.
Top Tips for Realistic Project Estimations
To get better at estimating your projects, consider the following tips:
- Spend a reasonable amount of time analysing the users’ needs and ensure that everyone is in agreement about what needs to be delivered and estimated.
- Break the effort into as many constituent parts and detailed tasks as you can.
- Research and experiment with different estimation tools and techniques.
- Draw in as many experienced people as is practical, and take on board their expert advice.
- Have different groups of people estimate the same thing and compare the outcomes.
- Seek to prototype the solution before you provide a binding estimate.
- Examine past projects of similar size and complexity for guidance.
- Provide both best-case and worst-case estimates for each feature.
- Make use of the PERT method (Programme Evaluation and Review Technique). The PERT calculation is (P+4M+O)/6. P is the Pessimistic estimate, M is the Most likely estimate and O is the Optimistic This gives a weighted average toward the Most likely estimate and circumvents overly optimistic estimation.
- Factor in all project phases and activities, including management, documentation, training and handover.
- Account for uncertainty. You can quantify known unknowns by assessing the cost of a risk if it were to occur (i.e., its impact) times its likelihood.
- Add contingency to every part of the estimate, particularly to account for unknown unknowns.
- Step back from the detail and apply your gut instinct to the overall estimate. Does it make sense?
- Be aware that estimating duration and effort are two different things. Estimate your project’s effort in points of labour hours as opposed to calendar time and then apply a separate conversion factor to translate your effort into calendar time. This helps you cater to the fact that your team is never 100 per cent effective and that a person rarely works on a task for eight hours a day.
- Present your estimate as a range instead of just one single number.
- Re-estimate the project at regular intervals and keep the steering committee informed.
- Feed your estimates into the project’s business case to ensure that the justification for undertaking the project remains realistic and compelling.
- Create a culture of honesty, openness, and collaboration. It will help people express their opinions and feel accountable to their estimates.
When an estimate has been produced, don’t just see this as a tick in the box. Instead, take a step back, consult your gut instinct and look at the meaning of the estimate.
Does it make sense?
If something doesn’t feel right, question it and dig into the detail. Also, consider the bigger picture and put the estimate into the context of the business case and the justification for undertaking the project. Ask yourself:
In which ways does the estimate impact the viability of the project? Is there still a strong economic incentive for completing the project? By how much can the estimate overrun before it makes the business case invalid?
As a leader of a strategic project, you must think like an owner and consider what you would do if your own money were on the line. Try that for a minute! Imagine that you are the personal sponsor of the project. How does that make you think and feel differently about the investment and the benefits? Would you still press ahead with the project?
It is in everybody’s interest that you think and act like an owner and that you take joint responsibility for the business case.
If someone else has already written it, ask to see it and query anything you don’t understand. If it has not yet been written, take action and pull it together. If you don’t know how to compose it and how to quantify benefits and calculate the payback period, decide to study it. Don’t let anything hold you back. Behave like a project leader and take action to learn about the aspects that you don’t yet master.