en
Project management, Project Portfolio Management
Project budget overruns: How PMI explains the 9.4% waste - and how to close it with baseline, risk and portfolio
Project budget overruns are an operational symptom, not a planning one. Most PMO leaders running 30 or more projects in parallel know the pattern: variance shows up late, risk registers go stale by the second month, and the portfolio view is a quarterly PowerPoint stitched together from spreadsheets. The interesting question is not why overruns happen, but why some organizations consistently keep them below 9% of total investment while others sit closer to 11%. The next sections walk through the three operational practices that explain the gap, what the PMI data says about their compound effect, and how a Head of PMO can recognize whether the system is working in their own portfolio.
In this article, you will learn:
|
PMI Pulse of the Profession®: what 3,950 project managers really showed
PMI’s Pulse of the Profession® report, Beyond Agility: The Flexibility Imperative, surveyed 3,950 project managers across industries and regions. Its central finding is one of the few project management statistics worth memorizing: wasted investment due to poor project performance declined to 9.4% globally, down from 11.4% the previous year. Across a portfolio of 100 million USD in active projects, that two-percentage-point movement translates to two million USD in recovered value – every year.
The PMI report did not stop at the headline number. It segmented organizations by their operational practices and found a clear pattern: organizations applying standards consistently – what PMI labels “gymnastic enterprises,” using the metaphor of a gymnast who combines the discipline of training with flexibility of movement – wasted 9% of investment versus 10.5% in traditional organizations. The same group was significantly more likely to have high organizational agility (48% versus 27%), use standardized risk management practices (68% versus 64%) and reach high project management maturity (52% versus 45%).
The 9.4% waste figure – what it actually measures
PMI defines wasted investment as the share of project budgets lost to missed deadlines, overshot budgets and scope creep. It is a backward-looking measure, calculated from completed projects in the prior twelve months and self-reported by project managers within their organizations. Two implications matter for PMO leaders. First, the number is a floor, not a ceiling – industries with longer project cycles and tighter regulation, such as construction, pharmaceuticals or large IT transformations, often report higher waste once compounding effects are included. Second, it is a portfolio-level signal, not a project-level one. A single overrun does not appear in this figure; a pattern of overruns across a portfolio does, and that pattern is what separates organizations that retain investment from those that bleed it. The implication for a Head of PMO is that the figure cannot be improved by fixing one bad project – it has to be improved structurally, across the whole portfolio.
Standards drive the gap, not culture
The 1.5 percentage point difference between high-standard organizations and the rest looks small on a single project. At portfolio scale it compounds quickly. A PMO managing 50 projects of average 2 million USD per year carries 100 million USD of active spend; the gap between 9% and 10.5% waste is 1.5 million USD annually, and over three years that is enough to fund a small portfolio of new initiatives without additional capital. PMI’s data shows that the gap is closed by three practices applied consistently, not by methodology choice. Waterfall PMOs that approve baselines, operate risk registers and run portfolio dashboards outperform agile teams that do not. The headline is straightforward: methodology is a delivery decision, governance is an investment decision, and the data points to governance as the lever for reducing budget overruns.
Why budgets overrun: three structural failure modes
Project budget overruns at portfolio scale are rarely caused by one bad estimate or one unexpected risk. They are caused by gaps in three operational layers that should reinforce each other: baseline, risk, and portfolio visibility. When any one layer is missing or stale, the other two cannot compensate. The next sections diagnose each failure mode using data from PMI, Wellingtone and KPMG; the three sections after that prescribe the operational response.
No working baseline means no measurable variance
According to Wellingtone’s State of Project Management Report 2024, only 48% of organizations “usually or always” baseline their schedules. The implication is straightforward: more than half of all project status reports describe the current plan, not the deviation from the original plan. Without a fixed reference point, “we are tracking to plan” becomes “we are tracking to whatever the plan has become.” Overruns disappear because they cannot be measured – and what cannot be measured cannot be corrected. The pattern is reinforced by the same Wellingtone report, which finds that around half of project managers spend at least one day each month on manual status reporting; that effort produces activity reports, not variance reports, because the baseline that would make variance measurable is missing in the first place.
Risk registers that exist on paper, not in operations
KPMG’s Global Construction Survey 2023 found that 37% of respondents traced budget or schedule overruns directly to inadequate risk management. The same KPMG survey reported that only around half of project owners say their projects finish on time, while 87% noted growing scrutiny of project performance from sponsors and boards. The dominant pattern, consistent across industries, is not the absence of a risk register – it is a register populated during project kickoff and never updated. A risk that was rated “medium” in week two stays “medium” in month six, regardless of what has happened to the underlying conditions. A risk register that is not revisited is a document, not an operational tool, and a document cannot trigger a decision before the risk materializes.
Portfolio-level invisibility hides accumulating overruns
Individual project variances are often small enough to look acceptable in isolation. A 6% overrun on one project, an 8% overrun on another, a delayed milestone on a third – each looks like a manageable exception, the kind of variance that any active portfolio will produce in any given quarter. Without a single portfolio view that aggregates these signals in near-real-time, the pattern emerges only at the quarterly review, when the cumulative effect is already visible in the budget actuals. By then, the PMO is reacting to overruns that have already occurred, not preventing the next ones, and the conversation with the sponsor moves from “should we reprioritize” to “how do we explain the gap to the board.” The structural problem is not that the data does not exist – it exists in every project – but that it is not assembled fast enough to be acted on.
Try FlexiProject!
Baseline as the first lever - how to make it operational
A baseline is not a saved version of the Gantt chart. It is an agreement between the project manager and the sponsor about what success looks like, formalized as a record that cannot be silently changed. Treated this way, baseline becomes a governance object, not a project artefact. The downstream effects are significant: variance becomes measurable, scope changes require explicit approval, and status reports become reports of deviation rather than reports of activity.
What a working baseline looks like in practice
An operational baseline contains three artefacts: an approved schedule with locked milestone dates, an agreed cost plan, and a scope reference that any change must be measured against. The approval is recorded in the system – who approved it, when, and against which version of the plan. Without that record, “what was approved” becomes a matter of memory and email archaeology, both of which tend to favor the version that explains the overrun rather than the version that contradicts it. The discipline matters most at the moment of change. When a new requirement appears in month four, the question is not “is this a good idea” – it is “what does this do to the approved baseline, who needs to approve the change, and what is the impact on cost and schedule.” Without a locked reference, that conversation never happens; the change is absorbed silently and the overrun appears two months later as a surprise.
How FlexiProject manages baseline approval and variance tracking
FlexiProject treats baseline approval as a workflow, not a checkbox. The schedule and cost plan move through an approval path before becoming baseline; once approved, the baseline is locked as a reference, and any subsequent modification requires a change request with explicit re-approval from the assigned authority. Variance against baseline appears on the project dashboard automatically – deviations in schedule, cost and scope are tracked continuously, not reported manually. For PMOs running large portfolios, the operational effect is that variance reporting moves from a monthly exercise of stitching together spreadsheets into a continuous signal available on demand. The project manager spends less time producing the status and more time acting on it, the sponsor receives variance data on the day a milestone slips rather than at the next steering committee, and the audit trail of who approved what and when is captured by the system rather than reconstructed after the fact. The full mechanics of baseline approval and variance tracking are described in the FlexiProject schedule and Gantt documentation.
Risk management as the second lever - registers that survive past kickoff
The PMI finding that 68% of high-standard organizations use standardized risk management practices, against 64% in the rest, looks like a small gap. The compounding effect across a portfolio tells a different story. A risk register that is operated – reviewed regularly, owned by named individuals, linked to schedule and budget impact – surfaces issues weeks before they become overruns. A register that exists only as a document does not.
From risk theatre to risk operations
Three concrete shifts distinguish operating a risk register from maintaining one. First, the review cadence: bi-weekly with the project sponsor, not quarterly at the steering committee – by the time a quarterly review happens, several of the risks listed at the previous one have either materialized or become irrelevant, and the review is reacting rather than anticipating. Second, ownership: every risk has a named owner who is not the project manager, with a clear response action and a deadline; when the owner is “the PM” by default, the response is never specific enough to be tracked, and the risk becomes a documentation entry rather than an actionable item. Third, escalation triggers: risk status is tied to milestone progress, so that a slipping milestone automatically elevates the related risks rather than waiting for the next review to surface them manually. Each of these three shifts is small in isolation; together they convert the risk register from a record that is filled in to a system that drives decisions.
How FlexiProject standardizes risk practices across a portfolio
FlexiProject stores risk registers as a structured part of the project record, linked to the schedule and the project charter. Risks have owners, response actions, and impact estimates that roll up to portfolio-level risk reports – a Head of PMO can see, across the entire portfolio, which categories of risk are growing, which projects are exposed to which threats, and which response actions are overdue. The reporting layer is built from the risk register data, not assembled separately for each steering committee meeting; the portfolio risk report on the day of the review is the same report that was available the week before, which means decisions can be made before the meeting rather than during it. The structure also enables comparison across projects: a risk pattern that appears in three projects simultaneously is visible as a portfolio risk, not as three isolated project risks, and the response can be standardized rather than improvised. The detailed mechanics of risk registers and project-level reporting are described in the FlexiProject reports documentation and the project charter documentation.
Projects organized by project portfolios in FlexiPorject PPM system
Portfolio governance as the third lever - making overruns visible early
Baseline and risk practices are necessary but not sufficient. They operate at the project level and answer the question: “is this project on track?” The portfolio layer answers a different question: “given everything we know about all our projects, which decisions should we be making now?” The PMI finding on PM maturity (52% versus 45%) reflects this gap. Organizations that close it are not better at executing individual projects – they are better at making portfolio-level decisions on the basis of project-level data.
What portfolio visibility actually shows a Head of PMO
A working portfolio dashboard does not show project status. It shows variance from baseline, risk exposure by category, and benefit realization against business case – across every active project, in near-real-time. The output is not a status report; it is a basis for reprioritization, which is a different decision with different inputs. Which projects need additional resources, which should be paused, which are exposed to risks that no longer have a containment plan – these questions get answered from the dashboard, not from the next steering committee. The decisive shift is in the relationship between data and meeting. In a status-driven PMO, the steering committee is where the data is presented and decisions are taken on the spot, often with incomplete information; in a portfolio-driven PMO, the data is already known before the meeting, and the meeting is used for decisions that require sponsor authority – reprioritization, resource reallocation, project cancellation. The number of decisions per meeting goes up because the time spent reviewing data goes down.
How FlexiProject’s portfolio layer connects to project baselines and risk registers
FlexiProject’s portfolio view aggregates data that already exists in the projects: baseline, variance, risk register, milestone status, cost. Nothing is entered twice – the project manager updates the project, and the portfolio view updates automatically, which removes the most common failure mode in spreadsheet-based portfolios where the project-level and portfolio-level numbers diverge silently between reviews. The Head of PMO sees a single view that connects project baselines to portfolio-level decisions: a project drifting from baseline appears immediately in the portfolio heatmap; a risk that crosses an impact threshold escalates to a portfolio risk view; a project that misses its benefit realization milestone surfaces in the portfolio benefits review. The cycle – project baseline, variance, portfolio rollup, executive decision – runs continuously, not quarterly, which means decisions can be made when they are still preventive rather than corrective. Detailed configuration of portfolio views and project reviews is covered in the FlexiProject portfolio documentation and project reviews documentation.
Risk tab wif risk register in project portfolio view in FlexiProject
Try FlexiProject!
What changes when all three levers operate together
The PMI data does not show three independent effects. It shows one cumulative effect: organizations that operate baseline, risk and portfolio practices as a coherent system waste 9% of investment, while those that operate them inconsistently – or not at all – waste 10.5% or more. A PMO running 30 to 50 projects can recognize whether the system is working by looking at a few specific signals.
Three signals that the system is working
Variance reports are produced in minutes from system data, not in days from spreadsheet aggregation; the project manager who used to spend the first working day of every month preparing the status pack now uses that day for risk response and stakeholder management. Risk decisions move from monthly steering committees to weekly project reviews, because the data supports faster cycles and the cost of waiting becomes higher than the cost of deciding. Portfolio reprioritization happens on the basis of dashboard signals – a project drifting from baseline, a risk crossing a threshold, a benefit realization slipping – rather than on the basis of who is loudest at the quarterly review or which project has the most senior sponsor. The conversation in the steering committee changes character: less time on “what is the status of project X,” more time on “given the portfolio data, which project should be paused to free resources for project Y.”
Three signals it isn’t yet
Status meetings still produce the data instead of consuming it: the project manager spends a day each month preparing the status pack, and the data is already a week old by the time it is presented. The risk register is stale by month-end; the same risks have been listed at “medium” for three quarters with no review entries, no owner accountability, and no link to the milestones that should be triggering re-evaluation. The portfolio review revisits the same five problematic projects every quarter because the underlying data is not updated between reviews, and decisions taken at one review have to be re-litigated at the next one when the situation has drifted again. These are not personality issues – the project managers and PMO staff are typically working hard within the system they have been given. They are signals that one or more of the three layers is not operating as a system, and the response is structural, not motivational.
Frequently asked questions about project budget overruns and PMO governance
What is considered an acceptable level of project budget overrun?
There is no universal threshold, and the answer depends on the type of project, the industry and the maturity of the governance framework. PMI’s data shows that high-standard organizations lose around 9% of project investment on average, while the rest lose 10.5% or more – both figures include projects that finish on budget and those that overrun, so the per-project overrun on actually-overrun projects is meaningfully higher. At the individual project level, most PMO governance frameworks treat variances above 10% as requiring formal sponsor review and a documented decision, with smaller deviations tracked through standard variance reporting. The more useful question is not “what overrun is acceptable” but “at what variance level does the PMO take a decision,” and that threshold should be defined in the governance framework before the first project starts.
How does baseline approval differ from saving a Gantt chart?
Saving a Gantt chart preserves the current plan. Baseline approval creates a locked reference that the current plan is measured against. The difference is operational: a saved Gantt can be edited silently, with no record of what changed, while an approved baseline requires a change request and explicit re-approval to be modified. Without that record, variance against baseline is not measurable, because the baseline itself moves with the plan – every status report concludes that the project is on track, because “the plan” is whatever the plan is today. The same distinction explains why some organizations report consistent on-time delivery while their actual budget actuals diverge from the original business case: the baseline they are measuring against was reset along the way, often without explicit approval, so the variance was never visible in the first place.
Do you need to be “agile” to reduce project budget overruns?
PMI data suggests not. The factor that correlates with lower waste is the consistent use of standards – baseline, risk management, portfolio governance – not the choice of methodology. A waterfall PMO that approves baselines, operates risk registers and runs a portfolio dashboard will outperform an “agile” team that does not. Methodology is a delivery choice; standards are a governance choice. The data points to governance, not delivery, as the lever for reducing budget overruns, which is also a useful framing for PMOs that operate hybrid environments: the same governance standards apply to waterfall, agile and hybrid projects, while delivery methodology is selected per project based on the type of work.
How often should a PMO review the risk register?
For active projects, bi-weekly between the project manager and the sponsor is a working cadence in most environments. Quarterly is too slow for risks tied to active milestones – by the time the steering committee reviews the register, the milestone has passed, and the risk has either materialized or become irrelevant. At portfolio level, monthly aggregation of risk data is generally sufficient to support reprioritization decisions, provided that project-level reviews are happening on the shorter cadence. The combination matters: a portfolio-level monthly review that aggregates stale project-level data produces decisions based on a snapshot that no longer reflects reality, while a project-level bi-weekly review without a portfolio rollup produces good local decisions that may conflict with each other across the portfolio.
What’s the difference between project-level and portfolio-level governance?
Project-level governance answers “is this project on track against its baseline and risks?” Portfolio-level governance answers “given the state of all our projects, which decisions should the organization be making now?” The two layers require different data views and different decision rhythms – project governance runs on a weekly or bi-weekly cycle and operates on detailed milestone and risk data, while portfolio governance runs on a monthly or quarterly cycle and operates on aggregated variance, risk exposure and benefit realization data. Most PMOs that struggle with budget overruns have one of the two layers operating, not both. The common failure mode is strong project-level governance – good baselines, active risk registers – without a portfolio rollup that converts those signals into portfolio decisions; the projects are well managed individually, but the portfolio as a whole drifts.
Closing the gap between standards and outcomes
The PMI data does not point to agility, methodology or culture as the lever for reducing project budget overruns. It points to three operational practices applied consistently: baseline approval that locks the reference against which variance is measured, risk management that operates between kickoff and closure rather than at kickoff alone, and portfolio governance that aggregates project-level signals into portfolio-level decisions. The 1.5 percentage point gap between high-standard organizations and the rest looks modest until it is multiplied across an active portfolio – at which point it represents real money, year after year.
The strongest PMO initiatives are not the ones that add a fourth tool to fix a gap in the third. They are the ones that operate baseline, risk and portfolio governance as a single system, with shared data, shared cadence and shared ownership. FlexiProject stands out because it combines all three in one platform: baseline approval workflows that produce continuous variance signals, risk registers linked to schedule and rolled up to portfolio level, and a portfolio dashboard built from the same underlying data, not from spreadsheets stitched together for the next review. For a Head of PMO whose portfolio carries tens of millions in active spend, closing even part of the 1.5 percentage point gap repays the operational investment many times over – and the path to closing it runs through governance, not methodology.
AUTHOR
Dominik Wrzosek
General Manager at FlexiProject
![What is a Gantt Chart, and how to create one? [EXAMPLE]](https://flexi-project.com/wp-content/uploads/2024/04/What-is-Gantt-chart-e1714048900401.png)
