Implement successfully

In companies and organisations alike, a lot of decisions are being made daily. Many of these are simple and implemented using the Nike statement of “Just do it”. However, some decisions require more diligence and consideration/planning to be implemented successfully and hence meet targets.

Often implementation or update of company IT systems, capacity expansion, market entry, change management and other projects as well as strategic initiatives take more effort to be implemented successfully. For the sake of simplicity, let me refer to such decision implementation as “projects” as most of these will be temporary and well-defined tasks to be solved.

I assume, perhaps naively, that no-one decides to start implementing with failure in mind, and I trust leaders to want to drive successful implementation. As Benjamin Franklin stated more than two centuries ago “by failing to prepare, you are preparing to fail”.

As I see it, there are a number of elements, that need to be in place to drive successful projects. Some may be able to succeed without one or more of these – but that would be hoping for luck, and as Peter Drucker stated decades ago “hope is not a strategy”.

1. Valid targets
2. Good governance
3. Experienced team
4. Uncertainty-based planning

Now this may seem logical and a complete no-brainer, yet I have seen multiple projects, which suffer on several of these parameters.

1 – Valid targets

Prior to planning and even contemplate implementation of a decision, it is paramount we know what success looks like. “We want to establish IT system XX” is rarely a valid target. If you are not the IT vendor, one may ask why do you want this system. How do you measure the value of a successful implementation?

Yes, there may be more than one target for any one project. Often these will be based on benefit, cost and timing.

By and large, success criteria should match company purpose, strategic aspiration or some key performance indicator derived from this. If it does not – how do you envision the implementation should strengthen company performance?

Target definition must be SMART. The term is so well known, I will not even discuss this further.

Valid especially means realistic. Many targets are based on optimism bias, where we as fallible human beings tend to overestimate the benefits, we will get, underestimate the issues and problems we may be facing, and overestimate our ability to handle these effectively. This may be just normal bias, but in an organisation, it may, even worse, be accelerated in a “fight” for resources where project teams face that “if we do not deliver this much with that little, this fast, we will never get managerial approval”. In such cases, deciding (executive) teams will find themselves prioritising between a series of equally unrealistic prospects – which cannot be a desired situation.

Finally, we know the future is unpredictable, and hence it will be highly valuable if/when management defines both what is success as well as what is failure. Here, failure is unacceptable performance and NOT just the missing a target. Red/amber/green charts are very popular, and can be used here as well:

• What does it take to deliver a green/good performance?
• What does it take to deliver a red/failed performance?
• What is the “not good, but acceptable” performance in between?

In an example case, aimed at enhancing productivity above the current 100 we have today. The target is to enhance by 25% using a new approach:

• Resulting productivity beyond 125 is green
• Resulting performance below the current 100 is red
• Resulting performance between 100 and 125 is recognised as an improvement, but it not meeting the defined target and hence amber

This acknowledgement of ranges will help project managers make the needed decisions and trade-offs to execute the project.

2 – Good governance

Governance or leadership is important. There is a reason, orchestras have conductors and sports teams have coaches/trainers. Good governance adds value and has a range of elements to it.

• Decision clarity. It must be clear when the project team/manager can make a decision and just keep the governing body, often a project steering committee (PSC) informed. The project manager also needs to know when decisions have to be escalated, and what decision material the PSC will require to make the needed decision
• Management commitment. The PSC members should regard themselves as being on the board of directors of “Project XX Inc.” and especially the chairman of the PSC should adopt the role of “chairman of the board” and be able and willing to devote as much attention, time and focus to the project as would be the case of a “real” board chair position. Former IMD professor Xavier Gilbert et al writes about this in their book Smarter Execution. A further, and equally important element of good governance is “stability”. If/when the PSC changes resource allocation, timing, priorities and/or scope during the implementation they severely hamper the likelihood of meeting what is now a moving target. In essence, every such decision should be treated as the base for a new planning, target setting etc.
• Reporting structure. What and when reporting takes place should be defined and adhered to. This may include issues, where extraordinary reporting/meeting is required to ensure successful implementation. It would be natural that the PSC would request regular updates on:

1. Progress vis a vis plan
2. Likelihood of meeting targets
3. Key issues being addressed

• Steering committees should be provided with reporting showing the likelihood of meeting targets. The uncertainty-based planning approach discussed later will enable reports to include charts like the below red/amber/green charts which gives insight as to project performance.

Failing to apply good governance is essentially depriving the project team any decent chance of being successful, yet probably the most important/common reason many end up failing. That is bad and unfair leadership of those involved.

3 – Experienced team

The more important a project is defined to be, the more important it is, that those involved have the needed competencies both in terms of professional competence and in terms of experience. Now, no-one starts with having experience and experience is a hands-on skill, not something that can be studied. New, untested project people can be involved as supporting members and/or in smaller projects where the stakes are smaller.

This need for competency and experience goes beyond the project manager and project team members. It relates to the PSC (project steering committee) as well. If Vice President Jones has never been on the PSC of a double-digit multi-million-dollar project before – do not let Jones be the chairman of your most important strategic project irrespective of what other competencies Jones may have.

Too often, I have seen project managers being “on their own” even on large and complicated projects. This is inefficient at best. The leadership task of a project manager makes this a full-time job, and should be supported by some sort of “project coordinator” role who, as direct support to the project manager:

• Monitors whatever measures are to be monitored, and drives the project manager’s attention to where it may be needed
• Operates as the practical liaison with insights of the many issues that emerge throughout the project
• Prepares reporting to project management as well as the PSC and potentially, do the project report calculations

Being a project coordinator is a great learning position (known to Star Wars? fans as the Padawan learner) for someone aspiring to become project manager.

4 – Uncertainty-based planning

My guess is, that 90% or more project plans in the world today are based on single point estimates of task duration, schedules costs, risks etc. Some of the best have done some separated sensitivity analysis where a single parameter is altered, and looking at how this affected overall performance.

However, we all intellectually know that:

• Most assumptions will materialize within a range of plausible outcomes
• Some risks will materialise and others will not
• Those that do materialize will affect performance within some range of plausible outcomes
• There will be levers (good risks) that help performance

Still, I cannot help but wonder why even admirable companies and project specialists insist on ignoring this when doing planning. I recognize this will be rebutted by statements like “we do take this into due consideration, this is what we have decided on buffers for, and that is why estimates are cautious”. Honestly, you need to be VERY experienced to get these buffers defined correctly when holding on to single point estimates.

Instead, define all relevant parameters as ranges – most often like a minimum, a most likely (what you originally planned for) and maximum value. Allow your minimum and maximum to be the 90% best/worst case rather than total absolutes.

Define risk impact the same way, and define the likelihood based on 1 – “the likelihood it does not happen at all” rather than the likelihood you think it will materialise as you will tend to use the likelihood which will have the most likely effect.

When you define the risk impact range, avoid guessing as much as you can. Nobel laureates Kahneman and Thaler as well as experts such as Douglas Hubbard, Bent Flyvbjerg and many others neuro-scientists have demonstrated human beings are horrible at guessing. So – do make the effort of finding supporting data by which to define outcome ranges. Douglas Hubbard has demonstrated data are always available, and Professor Bent Flyvbjerg refers to this as “reference class forecasting”, an approach which cannot be recommended enough.

Define levers/opportunities/windfalls/positive risks the same way as negative – not all news are bad news, and you will be too negative, if you do not consider good fortune as well.

Modelling your project scheduling, cost and benefit to the extent this can be Monte Carlo simulated will enable you to calculate your likelihood of meeting targets. Often, the first approach will lead to a plan which has an unacceptable outcome (too low likelihoods of success). Not to worry (yet). You now leverage the analytical capabilities of Monte Carlo simulation to show which uncertainties, risks and levers can most efficiently be addressed to improve projected performance.

Your plan evolves through a number of iterations as shown here.

There is an old saying that goes something like “A stitch in time saves nine”. This goes for planning as well. A well adjusted and realistic plan is more likely to lead to a successful outcome than a single-point estimated one-off plan.

Conclusion

Look at the performance in your project portfolio of the past 2, 5, 10 years. Compare the targets and plans from when the “go” decision was made with what actually materialized. If this shows you tend to meet deadlines, keep within budgets and deliver on scopes – good for you, and I apologise for having wasted your time reading this far. My guess is, actual performance lacks some distance to be desired.

I guess not adhering to one or more of the four elements of this paper can be seen as reasons for the lack of success. Now I work with risk management and uncertainties, so I do NOT guarantee your outcome will be better if you apply these four – but I challenge you to try, and see if it does not. The added effort spent will in many cases be miniscule compared to the value gained.

Should you wish to learn more and also get a reference project Monte Carlo model, I have put this in my book "Decide to Succeed", which is available on Amazon as well as through my website. I also train people in this as the core of my business.

Good luck – knowing that “luck favours the well prepared”

Hans L?ss?e