Construction Project Dashboard: What a Project Manager Actually Needs
Most dashboards handed to project managers were designed for executives. They show portfolio margin, division roll-ups and quarter-over-quarter trends, none of which help a PM decide what to do on Tuesday. A construction project dashboard in USA is a different object. One job, one person, one week. It answers what changed, what is at risk, and what needs a decision. This covers the seven panels that earn their place, the numbers that predict trouble before the accounting shows it, and what to deliberately leave off.
Design for the Tuesday morning question
Every panel on a construction project dashboard in USA should survive one test: does it change what the project manager does this week?
That test eliminates most of what typically gets built. Quarter-over-quarter margin trend does not change a PM’s week. Committed cost against budget by cost code, with the three codes that moved, does. It also sets the refresh cadence. A PM operates on a weekly rhythm with daily exceptions, which means overnight refresh is right and real-time is an expensive answer to a question nobody asked.
The seven panels
1. Cost to complete, by cost code
Budget, committed, actual, forecast, variance. Sorted by variance rather than by size, so the problems appear at the top rather than the largest line items. This is the core of any construction project dashboard in USA and it is where PMs spend most of their time. It needs to be sortable and it needs drill-through to the transactions, because the first question after any variance is which invoice caused it.
2. Committed cost not yet invoiced
Subcontracts and purchase orders issued but not yet billed. This is the panel most often missing and it is the difference between a forward-looking view and a rear-view mirror. A job can look healthy on actual cost while carrying commitments that will land next month. A PM who can see that wave coming can act. One who only sees posted cost finds out at close.
3. Change order status and aging
Approved, submitted, priced but not submitted, and drafted. In dollars, with days outstanding on each. Unapproved change orders are work being performed at your own risk, and the aging is the number that gets attention. We covered the wider financial exposure in change order management.
4. Labor productivity, this week
Installed units per labor hour against the estimated rate, by cost code, for the current week and the trailing four. For self-perform work this is the earliest reliable signal that a job is drifting, and it moves weeks before job cost reflects it. It requires field quantities and payroll hours joined to the same cost codes, which is the main reason it is missing from most dashboards rather than any disagreement about its value.
5. Schedule status against the baseline
Milestones, activities behind, float consumed. Schedule data usually lives outside the ERP, which makes this the hardest panel to build. Where full integration is not yet feasible, a manually maintained milestone table beats omitting schedule entirely.
6. Open items requiring the PM
RFIs awaiting response with ball-in-court, submittals pending, and approvals sitting with this project manager. A short list, aged. This is the panel PMs actually open first, because it is the only one that tells them what to do rather than what happened. Put it high on the page even though it is the least analytical thing on it.
7. Billing and cash position
Billed to date against earned, retainage held, and the current pay application status. PMs are frequently held accountable for cash outcomes while being shown none of the data, and this panel closes that gap.
The numbers that predict trouble
Four measures give warning before the financial reports do. A construction project dashboard in USA that includes them is genuinely predictive rather than descriptive.
Signal
What it means
Typical lead time
Productivity below estimate for 3 straight weeks
The estimate was wrong or execution has changed
4 to 8 weeks before job cost shows it
Committed cost rising faster than percent complete
Scope creeping or buyout coming in over budget
2 to 6 weeks
Unapproved change order value climbing
Risk accumulating and cash being deferred
Immediate, and it compounds
Overtime percentage rising
Schedule pressure being absorbed by labor cost
2 to 4 weeks before it hits margin
Read together, these are the front end of profit fade. We wrote about building a systematic version of this in profit fade early warning system, and the underlying point is that most profit fade is visible in operational data well before it becomes an accounting event.
What to leave off
Restraint is what makes a dashboard get used.
Portfolio and division roll-ups
Not the PM’s job and not their decision.
Quarter-over-quarter comparisons
The job has a duration, not a fiscal calendar.
Anything requiring interpretation before it means something
If a number needs explaining every week, either explain it once on the page or remove it.
Vanity visuals
Gauges, speedometers and maps look impressive in a demo and communicate less than a sorted table.
More than one screen
If a PM has to scroll past the fold to find the thing they need, they will go back to their spreadsheet.
That last point is worth taking seriously. Project managers abandon dashboards for spreadsheets because the spreadsheet answers their specific question faster, not because they dislike the technology. We unpacked the pattern in why project managers fail with spreadsheets.
Where the data comes from
A single-job dashboard draws on fewer systems than a portfolio view, which is one reason it is a sensible first build.
Panel
Source
Difficulty
Cost to complete
Construction ERP such as Acumatica, Sage 300 CRE, Foundation, CMiC or Jonas
Low
Committed not invoiced
ERP commitments
Low
Change order status
ERP or project platform
Low to medium
Labor productivity
Payroll plus field quantities
High. Two systems, one cost code structure.
Schedule
P6 or Microsoft Project
High
Open items
Procore, Autodesk Construction Cloud or equivalent
Medium
Billing and cash
ERP accounts receivable
Low
Five of the seven panels come from the ERP alone, which means a genuinely useful construction project dashboard in USA is achievable in weeks rather than months. Productivity and schedule are the two that require integration work, and they are worth deferring to a second phase rather than allowing them to delay the first. Where several sources are involved, data warehousing and management is what keeps definitions consistent.
Build it in two phases
Phase one, two to four weeks: the five ERP-sourced panels. Release to three project managers, not the whole company.
Watch how they use it for a month. Note which panel they open first and which they never touch.
Fix what the first three PMs complained about before widening the rollout. Adoption is decided here.
Phase two, four to eight weeks: labor productivity, then schedule.
Add drill-through to transactions at every level. This is what stops people opening the ERP separately, which is the behavior that kills dashboard adoption.
The sequencing matters more than the tooling. A construction project dashboard in USA released to three PMs and improved from their feedback gets adopted. The same dashboard released to forty people at once gets ignored by thirty-seven of them. We covered the common failure modes in why Power BI dashboards fail in construction.
What good looks like
A project manager opens it before their weekly job review rather than rebuilding a spreadshee.
Variance conversations start from the same numbers finance is using.
Change order aging is visible to the PM and the project executive at the same time.
Productivity drift surfaces weeks before it reaches the income statement.
The last refresh time is on the page and nobody asks whether the data is current.
If you want a view on what your current job data can already support, the answer is usually more than expected, because five of the seven panels need only the ERP. talk to our team, or read more about our Power BI dashboards for construction and construction reporting software work. For the PM-specific view in more depth, see Power BI dashboards for project managers.
Frequently Asked Questions
What should a construction project dashboard include?
Seven panels: cost to complete by cost code with drill-through, committed cost not yet invoiced, change order status with aging in dollars, weekly labor productivity against estimate, schedule status against baseline, open items requiring the project manager, and billing and cash position. Five of the seven come from the ERP alone.
How is a project dashboard different from an executive dashboard?
Scope and audience. A construction project dashboard covers one job for one project manager on a weekly rhythm, and answers what changed and what needs a decision. An executive dashboard covers the portfolio, compares jobs and divisions, and answers where to focus attention. Giving a PM the executive view is the most common design error.
How long does it take to build a construction project dashboard?
Two to four weeks for the five panels sourced from the ERP, which is enough to be genuinely useful. Labor productivity and schedule take a further four to eight weeks because they require joining payroll to field quantities and integrating a scheduling tool. Release phase one before starting phase two.
What early warning signals should a project dashboard show?
Four predict trouble before accounting does: productivity below estimate for three consecutive weeks, committed cost rising faster than percent complete, unapproved change order value climbing, and overtime percentage rising. Together they typically give four to eight weeks of warning ahead of the financial reports.
Why do project managers stop using dashboards?
Almost always because the spreadsheet answers their specific question faster. The usual causes are no drill-through to transactions, more than one screen of content, and panels designed for executives rather than for a PM’s weekly decisions. Releasing to three project managers first and fixing their complaints before wider rollout resolves most of it.
