Project predictability is about accurately predicting the outcome of a project early enough that it’s possible to identify trouble spots, take corrective actions and keep the project on course.
For a portfolio of projects, the idea of predictability can be expanded to the entire suite of projects. In this context, project predictability is about increasing consistency and minimizing surprises across a large number of projects.
Traditionally, when industry leaders look to determine project success across the project portfolio, they focus on outcome variance. However, variance is not a reliable predictor, as it only shows the difference between actual and planned costs and schedule. (In fact, variance is even acceptable at smaller percentages.)
Instead, project predictability focuses on timeliness of information in a project cycle—when likely changes to the outcome are identified.
In the engineering and construction industry, for example, it’s estimated that 98% of megaprojects exceed budget by more than 30%. How much of this could be avoided by predicting overruns earlier and resolving them quickly?
Why is Project Predictability Important?
Building processes and systems that improve project predictability is crucial to an organization’s success in many ways.
Timely forecasts provide sufficient time to course-correct, increasing the likelihood of achieving targets and reducing outcome variance. The response to a worrying forecast could be to descope some requirements, modify existing business cases, or even to scrap the project. However, the earlier senior leaders are given an opportunity to make these tough decisions, the better off the organization—this can lead to significant cost savings, as well as higher return on capital employed (ROCE) and saved opportunity costs.
Another reason project predictability is important is organizational credibility. Chronic delays and cost overruns can have a negative impact on the perception of the company. For contractors in particular, the worse the reputation is damaged, the harder it will be to win future projects. If the problems continue for some time, it can even reduce shareholder value and lead to an organization’s downfall.
Predictability ensures organizations are known for their excellence in execution—not for putting out urgent fires.
The Pillars of Project Predictability
Achieving project predictability requires a solid foundation of best practices. It’s an ongoing process that requires long-term commitment, from top executives to all employees down the chain. It also requires incentivization and a culture of data-driven decision-making.
Here are the five building blocks of project predictability:
1. Portfolio Management
The objective of project portfolio management (PPM) is to identify the optimal combination of projects that can deliver maximum strategic value to the organization.
To determine the right portfolio, projects are scrutinized thoroughly, and ranked and scored to measure their potential contribution to the organization. It also helps to benchmark them against past indicators to arrive at early metrics of project performance.
Analyzing resource availability is a crucial part of this pillar, as it is necessary to understand capacity of resources and compare that with the demand of the portfolio. This way, it becomes possible to estimate whether there are impending roadblocks during execution. In case resources are not available as expected, organizations can plan accordingly. For example, you can hire new people, move resources from other projects as a workaround, account for more budget to purchase new equipment, etc.
2. Project and Contract Management
Drilling down further from portfolios, we move to project and contract management. In the context of predictability, the focus here is to integrate change and risk functions with the rest of the project platform. This pillar is an acknowledgement of changing dynamics in projects and a shift in assumptions that existed at the beginning when deciding on the portfolio. These changes must be addressed quickly in order to cause minimal impact to ongoing projects.
By managing changes, risks and issues as part of a holistic technology platform, you can automatically assess the impact on project performance and forecasts, and therefore, respond to them at the earliest point. It also provides an opportunity to connect live project data with predictability metrics.
3. Project and Contract Controls
The goal of this pillar is to enforce tighter controls by integrating end-to-end project functions, including accounting, scheduling, resource management, procurement, and budgeting. Rather than using fragmented siloed solutions for each of these functions, having a consolidated solution (one platform, one login, one database approach) ensures that data is shared between different project departments and increasing overall visibility.
Using native, automated integration ensures that project managers receive near real-time alerts and data analytics that can drive corrective actions in a timely manner. Moreover, it provides transparency and promotes a culture that supports what-if scenario analysis and drilling down to the root cause of trends.
4. Performance Management
To derive a reliable predictability model, you must avoid monolithic, just-in-time forecasting methods in favor of methods that constantly adjust forecasts regularly during project execution.
To achieve this, it’s important to collect trustworthy progress data from internal and external sources in real-time, collate them, and map them to the actual impact on cost and schedule overruns. This multi-method progress measurement process allows for the right source of progress to be assigned to the project deliverable.
The other aspect to performance management is to deliver metrics at the right time: before it’s too late to react. Building visual tools, such as dashboards, can help to share information with relevant stakeholders on an ongoing basis.
5. Predictability Measurement
To accurately measure predictability, projects need to employ metrics that provide visibility into forecast timelines. The Construction Industry Institute (CII) recommends two predictability indices: Normalized Cost Timeliness (NCT) and Cost Predictability (CP = NCT * % Cost Deviation). While NCT measures the timeliness of the forecast, CP is an indication of the accuracy of the forecasts by factoring in the size of variances.
If you create a bubble graph plotting NCT against CP, you want to be at the lower end of the quadrant with a low NCT (pointing to less time taken to predict) and a low CP (less deviation from forecast). These graphs can give a definite sense of how past and present projects compare against each other.
Predictability measurement is also about integrating forecasting with incentives. This may include direct incentives (such as project bonuses being tied to early, accurate reporting) and indirect incentives (such as correlating a person’s competence and integrity with high predictability). As controversial as these may be, project predictability incentives< tend to serve as a key regulator of human behavior and biases, among the largest causes of project overruns.
Best Practices in Project Predictability
Although predictability is an intuitive concept, implementing it on the ground can be a huge task when organizations run hundreds or thousands of projects simultaneously. Additionally, multiple factors, such as resource dependencies, customer commitment, requirements and budget, are in play here. Any of these factors could change, making the entire predictability process even more complex.
Here are a few best practices to help tackle the predictability challenge:
Establish the Five Predictability Pillars
A significant part of this process is to get all five pillars—portfolio management, project and contract management, project and contract controls, performance management, and predictability measurement—to function effectively. As these are the foundational blocks that enable predictability, it’s important to get them right and integrate data from these pillars with the complete project lifecycle.
Promote a Culture of Transparency
Achieving predictability is closely tied to the human factor, which includes not just the people on the project, but also the organizational culture and incentive structure built around projects.
For example, most projects’ reward is for finishing tasks on time and on budget. So, if a team member identifies an issue at 20% stage of the project, he has no incentive to report it at that point. The optimism bias kicks in, and he believes he can somehow finish the project on time. In reality, the situation worsens, and the issue gets reported when the project is at 70%, when it’s too late to take corrective actions. Late reporting also has potential negative ramifications, such as blame, micromanagement, and perhaps even eventually a toxic team culture.
It is therefore important to make sure incentives line up with your approach to project predictability. If the goal is to surface potential problems earlier, incentivize early disclosures and reinforce this expectation in meetings and one-on-one conversations throughout the life of the project.
Leverage Predictability Indices
Outcome variance analysis alone is not enough to improve project predictability. As an alternative, CII introduced the predictability index. This index measures a project team’s ability to predict cost and schedule performance by assessing three core competencies:
- Timeliness of forecasts
- Accuracy of forecasts
- Deviations at completion
Computing these metrics helps organizations get a sense of how well their projects are doing on the predictability scale, which indirectly tells how well projects are being executed without overruns.
Use Technology as an Enabler
Technology can function as a transformative medium for project predictability. However, it has to blend together with other factors, such as people and processes.
According to KPMG, the missing link in transforming the performance of project-driven organizations is integrating people, governance and technology. This integration can make a huge difference to projects’ success.
Using an enterprise project performance platform can help support predictability across project portfolios and the organization overall. For example, consider a platform that contains all the data from the five pillars and can automatically compute predictability indices behind-the-scenes, apply multiple forecasting methods, and flag problem areas in need of further investigation.
Organizations can benefit greatly from a solution that functions as a central project data hub, collects data from thousands of internal and industry-wide projects over years, and uses this data for improving project predictability.
Benchmark Predictability Data
Benchmarking project information against a wider pool of data is very important to recognize issues and set off alarm bells sooner. It’s also a useful method to measure and compare against other projects, regions, business units, etc.
However, benchmarking is not easy to implement because it’s not always an apples-to-apples comparison when it comes to different projects. Most organizations have neither sufficient data nor expertise to do it, and teams are often reluctant to share data.
This is where enterprise project performance platforms can be useful. By offering a streamlined method to collate predictability-based metrics across organizations, they are able to consolidate and analyze huge volumes of data. They also have the functionality to normalize metrics data for external factors, such as regulatory changes and unforeseeable risk events.
Common Challenges for Project Predictability
There are several reasons why projects have low predictability. To start, we can group these into two buckets: human behavior and systemic design.
In its report on improving project outcomes and predictability, CII identified human behavior as the biggest factor contributing to project overruns, trumping the usual suspects: planning, scope development and project management issues.
Optimism bias, the mindset of people to believe that project overruns can somehow be recovered in due time, is an example of how human behavior can negatively impact projects.
Another dimension of human behavior is insufficient attention to detail. This could translate in the form of not assigning enough resources, poor change and risk management, or a laid-back focus on process and project controls. As projects grow in scale, errors like these could become impossible to fix or manage.
The unwillingness to deliver bad news is tied to low transparency and accountability in organizations. It raises the question of whether fundamental reporting systems are in place, as well as if organizations have the right processes and systems to capture a single version of truth amidst chaos.
Another common challenge is low organizational maturity. Low maturity could be the use of a basic and straightforward approach to project planning, estimation and risk management, without factoring in the complexities of the environment. Or, the use of siloed systems and processes in various projects that makes it very difficult for people to compare metrics across the board, which then results in insufficient data to make sensible decisions.
Many systemic issues can be attributed to the lack of an enterprise projects performance platform with the tools to consolidate data and leverage past performance. Adopting such a software platform can provide the mechanism to adjust human behavior by driving teams to disclose issues earlier, highlighting risk areas ahead of the curve and binding project metrics to the rewards system.
How to Achieve World-Class Project Predictability
Most organizations have some systems in place to capture bad news early on. CII reports that, on average, variance reporting only starts at 65% into project completion. This unfortunate statistic gives a sense of why projects constantly overrun their schedule and cost.
In order to push the boundaries of predictability and set your project portfolios up for success, here are a few tips on how to get started:
- Adopt all pillars of predictability.
- Enable these pillars with an enterprise technology platform.
- Implement the best practices discussed above.
- Combine out-of-the-box best practices with your organization’s own data and business processes.
- Automate integration, reporting, and predictability analytics.
- Promote transparent, proactive behaviors.
- Setup incentivization schemes that encourage teams to disclose bad news early.
Predictability as a Progressive Metric
For all the emphasis on predictability, it should be acknowledged that it currently acts as a “hindsight” metric. The future lies in challenging the status quo and in turning these metrics into predictive indicators.
It is certainly possible to achieve this when enterprises use project performance platforms to their greatest ability. By aggregating data (e.g., unstructured status information, information on risk and issues, team competence assessments) across hundreds of projects and applying a combination of machine learning and big data analysis, these past predictability metrics can serve as predictive analytic tools.
See how EcoSys can help you build more predictability into your projects and your organization.