To measure project profitability, subtract direct project costs from revenue, using loaded labor cost instead of billed rate, then divide by revenue and multiply by 100. To forecast it, multiply hours logged by loaded cost, add the projected cost of the remaining hours, and compare the total to the approved budget at least every two weeks.
Project profitability is the percentage of a project’s revenue left after what it actually cost to deliver. Many agencies can calculate that percentage. Few can see it while there’s still time to do anything about it.
That gap appears constantly in conversations with agency leaders. The complaint is rarely “we don’t know our margins.” It sounds more like: the project looked fine at kickoff, looked fine at the midpoint check-in, and then the numbers came in at close. Finance and operations leaders describe this more than almost anything else when the topic turns to margin: not seeing a project go underwater until it already has.
This post covers the formula, how to measure it project by project, and the part most guides skip: forecasting it while the work is still in progress.
What “project profitability” actually measures
Project profitability is a delivery-level number, not a firm-level one. It answers one question: on this specific engagement, how much of what the client paid turned into margin after paying for the people and expenses that delivered it.
Agency profitability works differently. It rolls up every project plus overhead, sales cost, and everything else it takes to run the firm. You can run a string of profitable projects and still have a stressed agency if overhead is heavy or utilization is low. You can also have a healthy firm carrying one or two projects that are bleeding money. Both numbers matter. Project profitability is the one a delivery team can actually manage week to week.
The project profitability formula
It isn’t complicated:
Project margin = (Project revenue − Direct project costs) ÷ Project revenue × 100
Direct project costs mean labor at loaded cost, subcontractor fees, and any direct expenses tied to delivering the work. That excludes overhead and sales and marketing. Just what it took to do this specific job.
Here’s a worked example, using made-up round numbers. These figures are an illustration only, not a target or a benchmark for your firm.
Say a project sells for $100,000. Two people work on it: a strategist billed at $150/hour who costs the firm $90/hour loaded (salary, benefits, and payroll taxes divided into an hourly rate), and a designer billed at $125/hour who costs $70/hour loaded. The strategist logs 200 hours. The designer logs 350 hours. There’s also a $5,000 contractor invoice for a specialty animation.
Direct cost: (200 × $90) + (350 × $70) + $5,000 = $18,000 + $24,500 + $5,000 = $47,500.
Margin: ($100,000 − $47,500) ÷ $100,000 × 100 = 52.5%.
That’s the whole calculation. What takes work is getting an accurate loaded cost rate (many agencies underestimate it by using base salary instead of the fully burdened number) and catching it early enough to matter.
How to measure it per project
Measuring project profitability accurately comes down to three inputs.
Revenue is usually straightforward. It’s whatever the project sold for, adjusted for any approved change orders.
Direct cost is where the error usually creeps in. Loaded labor cost, not billed rate, is what belongs in the calculation. A $100/hour billed developer earning an $85,000 salary with benefits typically has a loaded cost closer to $60 to $70/hour, depending on how your firm allocates payroll taxes, benefits, and paid time off into that rate. Using billed rate instead of loaded cost overstates margin and makes a project look healthier than it is.
Timing is the third input. It’s the one most billing systems get wrong by design. A monthly billing report tells you what happened last month. It doesn’t tell you whether the project currently in flight is on track. That’s a different question, and it needs a different cadence: a check at least every two weeks against the current budget and current scope, not just at month-end close.
We’ve written more about why this data ends up scattered across systems that were never built to answer this question together. If you want the structural version, see why agency project profitability is so hard to predict.
Forecasting it before the project ends
Measuring tells you what already happened. Forecasting tells you where the project is headed while there’s still runway to change course. That distinction matters more than most profitability guides admit, because by the time a monthly report or a project-close review shows a margin problem, the hours are already spent.
An in-flight forecast is simpler than it sounds. Take hours logged to date and multiply by loaded cost. Add the projected cost of hours remaining in the current scope, then compare that total against the approved budget. If the projection is trending past budget while the project still has six weeks left, that’s six weeks to have a conversation about scope, staffing, or a change order. If the same trend only appears at the final invoice, there’s no conversation left to have. It’s just a write-off.
This is where estimation accuracy and margin protection connect. A forecast is only as good as the scope and hours it’s projecting against, which is why the estimate you started with matters just as much as the tracking you do afterward.
Estimation methods: the input your forecast depends on
Every profitability forecast starts with an estimate, and the method behind that estimate shapes how much you can trust the result later. A few approaches come up most often at agencies, and each breaks down in a specific way.
Bottom-up estimation builds the estimate from the smallest deliverables up: scope each task, estimate hours for each, then total it. It tends to be the most accurate method for well-defined work, forcing the estimator to list every piece instead of eyeballing the whole. It breaks down fast when the scope itself is unclear. A bottom-up estimate built on an undefined scope is just as uncertain as any other method. It just looks more precise.
Analogous estimation compares the new project to a similar one already delivered and scales the hours to match. It’s fast. That makes it useful early in a sales cycle, before a detailed scope exists. It breaks down when “similar” is doing more work than it should. A rebrand for a 50-person client and a rebrand for a 500-person client aren’t the same project just because the deliverable list matches.
Parametric estimation uses a rate per unit, hours per page or screen, and multiplies by the number of units. It’s consistent and easy to defend to a client, right up until custom work enters the mix that doesn’t fit the standard unit. Many projects eventually have some.
Three-point estimation asks for an optimistic, pessimistic, and most-likely hour estimate for each major piece, then blends them into a range. It doesn’t remove uncertainty so much as surface it, which is smarter when scope is genuinely unclear. This is usually the right method for discovery-phase work or anything with a client who hasn’t fully defined what they want yet.
None of these methods is universally correct. Bottom-up for defined scope, analogous for fast early-stage sizing, three-point when the scope is still forming. What matters more than the method is what happens after the project closes: comparing the estimate to the actual hours, by category, and tracking where you were consistently high or low. Ten projects of that comparison teach an agency more about its own estimating blind spots than any framework on its own.
Contingency isn’t padding
One estimating habit deserves its own mention: naming a contingency instead of hiding it. A “client feedback and revision” line of 10 to 15% on a design phase acknowledges something specific: feedback cycles are hours that happen on almost every project, whether or not they’re written into the scope. The difference between a contingency and padding is whether you can say what it covers and when it gets used. Teams that do this tend to price more accurately than teams that bury it as invisible margin and hope it’s enough.
If you’re evaluating a platform to help with this
Some agencies get to this point and start looking at software. Full professional services automation platforms, Kantata among them, promise a connected view across sales, delivery, and financials once you’re fully implemented. Kantata’s own site frames it as “instant margin and risk clarity,” which is the right goal. The tradeoff is what it takes to get there: a full platform typically means replacing or layering your CRM, project management, and time tracking under one system. Moving all of it takes time.
A lighter option connects the pipeline, capacity, and cost data already in your existing tools, without asking a firm to migrate off systems that are already working. Weigh that distinction before signing anything. For a closer look at Kantata specifically, see Kantata alternatives for agencies.
Utilization is the other lever tied up in all of this. It sits upstream of margin in a different way than estimation does, and it deserves its own explanation. See agency utilization rate: forecasting it, not just measuring it for that piece, and agency profit margin benchmarks if you’re trying to figure out what range is realistic for your agency type.
We’d be glad to walk through what an in-flight margin check could look like for your projects.
None of this requires new software to start. A two-week check-in cadence and an accurate loaded cost rate will catch most of what currently turns into a surprise at project close. Getting ahead of this has less to do with a sophisticated formula and more to do with not waiting for the invoice to say what’s already true.
Frequently Asked Questions
How do you calculate project profitability?
Take the project’s revenue, subtract direct costs (labor at loaded rate, not billed rate, plus subcontractors and direct expenses), divide by revenue, and multiply by 100. The formula is simple. The part that’s easy to get wrong is the loaded rate: use base salary alone and the margin looks better than it is. Consistent methodology across projects matters more than chasing a perfectly precise number.
What’s the difference between measuring project profitability and forecasting it?
Measuring tells you what a project’s margin was, usually after it’s closed or at a scheduled billing cycle. Forecasting tells you where margin is headed while the project is still running, based on hours logged, what’s left in scope, and the cost of getting there. Measuring is a look in the rearview mirror. Forecasting is the only version of the margin that arrives early enough to act on.
What is the difference between project profitability and agency profitability?
Project profitability measures margin on one engagement. Agency profitability measures the firm’s overall financial health once every project, plus overhead and firm-level costs, is accounted for. A firm can carry profitable projects and still struggle if overhead is high or utilization is low. It can also carry one unprofitable project inside an otherwise healthy year. Project profitability is what a delivery team manages directly. Agency profitability needs the firm-wide view.
Which project cost estimation method should agencies use?
It depends on how defined the scope is. Bottom-up estimation is the most accurate for well-scoped work, because it forces every deliverable to be named and sized individually. Analogous estimation, sizing off a similar past project, works for fast early-stage sizing but breaks if “similar” is doing too much work. Three-point estimation, blending an optimistic, pessimistic, and most-likely case, is the better choice when scope is still forming. Many agencies use more than one method depending on where a project sits in the sales cycle.
What is a contingency in project cost estimation?
A contingency is an explicit allocation for work that’s likely but not yet defined, like a client feedback and revision cycle on a design phase. It stays clear of padding as long as you can say what it covers and when it gets used. Agencies that name their contingencies tend to price and scope more accurately than agencies that bury the same hours as invisible margin.
What causes agency projects to become unprofitable?
The most common causes, roughly in order of frequency: scope creep that never gets a change order, an initial estimate that was too optimistic for the actual work, unplanned rework, senior staff absorbing work that was scoped for a mid-level rate, and a late-project hour spike when deferred work catches up in the final weeks. Many unprofitable projects have at least two of these happening at once. Estimation accuracy and change order discipline fix more of this than anything else.
How often should agencies check project profitability while work is in progress?
Monthly billing cycles are the minimum many agencies already have. A two-week check-in against current budget and scope is a meaningful upgrade, because it catches drift early enough to act on it. A project trending over budget in week four still has runway to correct if someone’s looking at week four. The same project discovered off-track at month-end billing usually doesn’t.