A ten-point net margin gap between two same-sized agencies in the same city usually has nothing to do with their rate cards.
The lower-margin shop will almost always trace the problem to pricing pressure—clients pushing back on rates, the new business team discounting to close deals, the competitor down the street undercutting fees. The higher-margin shop, more often than not, has a quieter explanation: they ran their projects tighter.
From the vantage point of watching hundreds of agency projects run through delivery at once, the disconnect between where agencies say their margin goes and where the data shows it actually goes is the most consistent pattern in the industry.
Pricing dominates the leadership conversation because it’s the variable everyone in the room can see clearly.
Data from Promethean Research, Parakeeto, SoDA, the 4A’s, and PMI tells a consistent story when read against real operational patterns: margin loss is a delivery problem dressed up as a pricing problem.
The piece below maps where the margin actually goes, how the pattern shifts by project type, and five questions to ask your own data this week.
Where Agencies Look Vs. Where Margin Actually Goes
That instinct toward the rate card isn’t careless—it’s structural. Pricing lives on a single visible artifact, gets debated with the client in real time, and produces a clean win-or-lose result on every deal.
The leadership team can change agency pricing within a week, which is why it tends to absorb most of the available oxygen even when the underlying issue isn’t sitting on the proposal.
The public benchmarks consistently point somewhere else:
- Promethean Research puts the average digital agency project margin at 35% and net margin at 13% for 2025. The gap between those numbers is where the agency business is won or lost, and it opens up between kickoff and final invoice rather than at the proposal stage.
- Parakeeto’s benchmarks put a healthy delivery margin above 50% at the P&L level and closer to 70% at the project level. Agencies that miss those numbers almost never miss them at pricing; they miss them because delivery cost more than the model assumed.
- PMI’s 2018 Pulse found uncontrolled scope changes on 52% of projects globally, up from 43% five years earlier. Project-based service businesses sit close to the centre of that trend.
The Categories That Quietly Eat Margin
Cross-reference the public benchmarks with what surfaces across a wide enough sample of live agency projects, and the same handful of categories keep recurring. None of them are surprising in isolation; what stands out is how stubbornly they show up across project types, agency sizes, and service mixes:
Scope creep and over-delivery.
The SoDA 2018 KPI Benchmark Study found the largest single source of lost agency profitability was over-delivering on scope without getting paid for it. In delivery data, scope creep looks like a project that hit its committed timeline but consumed roughly 30% more hours than the original estimate assumed.
Rework cycles.
Round three turns into round five, and each loop adds delivery hours without adding billable hours. Revisions per asset is one of the fastest leading indicators when margin starts to move in the wrong direction.
Approval stalls.
A project that sits with the client for a week on approval becomes a project carrying a week of opportunity cost on the team behind it. The downstream effect on utilization compounds quickly across a portfolio.
Discovery underestimation.
When discovery is compressed to fit a tight proposal deadline, every downstream phase inherits the same partial picture, and the compounding usually doesn’t surface until the second invoice cycle.
Unbilled coordination.
Internal handoffs, status meetings, and vendor management rarely appear on a proposal and usually get logged as internal admin time. Across a year, this is often the single largest category of unbilled hours an agency carries.
How The Pattern Shifts By Project Type
The categories above are roughly the same across an agency’s book of work. The weights vary substantially by project type, which is why an agency-wide margin fix rarely translates cleanly when the project mix is varied:
Fixed-fee creative projects.
The leak runs almost entirely through revisions and approval stalls, because the price is locked at the proposal, and every additional round of client feedback becomes a direct cost hit.
SoDA’s benchmarking has shown fixed-fee projects overrun more often than time-and-materials work by roughly five percentage points.
Ongoing retainers.
Scope drift and unbilled coordination do most of the damage here because the relationship is continuous, and small expansion requests feel uncomfortable to push back on. Each accommodation is small, but the cumulative effect over twelve months is usually substantial.
Development builds.
Discovery underestimation and rework dominate. Build projects are unusually sensitive to upstream brief quality and unforgiving when QA cycles compound—a 10% discovery underestimate can easily turn into a 30-40% delivery overrun by the time integrations are wired up.
Paid media engagements.
The margin lives in the labor wrapped around the spend, not the spend itself. Reporting, optimization, ad ops, and weekly check-ins are the hardest pieces to forecast accurately and the easiest to over-deliver against, particularly under the fixed-pricing models, the 4A’s/ANA compensation work shows, still dominates the market.
The Process Gaps Behind The Patterns
The categories above aren’t usually caused by individual sloppiness—they’re caused by process gaps that compound under deadline pressure. The patterns map almost one-to-one with the practices higher-margin agencies handle differently from the rest of the market:
Brief quality.
Briefs that read like creative direction but function as wishlists, with no measurable success criterion or out-of-scope boundaries, turn every ambiguity into a downstream rework cycle.
Time capture.
Manual time entry tends to undercapture actual delivery hours, which means the margin-by-project number reads more optimistic than reality, and the team manages toward a target it can’t accurately see.
Change-order discipline.
When account managers fear the conversation more than they fear absorbing the cost, change requests get absorbed rather than priced. PMI’s 2023 research found organizations with strong change-control practices experience significantly less scope creep.
Mid-project re-scoping.
When a project goes off track, the instinct is to add bodies rather than renegotiate scope or timeline, and the cost ratio usually gets worse rather than better.
Approval routing.
Multi-stakeholder approval extends a project’s cycle exponentially when stakeholders are added, and agencies that don’t lock the routing at kickoff pay for it on every project after.
The top-margin segments in Promethean’s data, running net margins well above the 13% industry average, look very different operationally—not commercially—from the rest of the market.
Five Questions To Ask Your Delivery Data This Week
Before reaching for the rate card or rewriting the proposal template, run these five questions against your last six months of project data.
They take an afternoon to answer if the underlying data is in reasonable shape, and they will tell you with reasonable confidence where your margin actually went:
- Average hours overrun, as a percentage of estimate, by project type. Any category running consistently over 15-20% is where the bleed is concentrated.
- Average revision count per asset, plotted against project margin. If the high-revision projects are also the low-margin projects, the upstream issue is brief quality, not pricing.
- Share of total logged hours sitting in internal or non-billable categories. Anything above 20-25% deserves a closer look at how coordination time is being scoped against client fees.
- Gap between approval requested and approval received, by client. Long gaps typically point toward unresolved scoping decisions at kickoff, not the client’s calendar.
- Delivery margin by client and by project, with the spread between best and worst. A thirty-point spread is a process spread, not a pricing spread, because the same team is delivering both ends of the distribution.
None of these are sophisticated questions, just rarely asked in sequence. The answers usually cluster around the same handful of process gaps regardless of agency size or service mix.
Why Margin Is Not A Pricing Issue
Pricing is the easiest lever to talk about because it’s the most visible, the most directly controllable inside a quarter, and the most natural narrative to take into a board meeting.
Delivery is harder to see, harder to talk about, and considerably harder to change, which is exactly why most of the margin tends to be sitting there.
The agencies running the highest margins are usually the ones running tighter projects underneath better briefs, cleaner change orders, faster approvals, more accurate time capture, and stronger operational discipline at the points where margin most often goes missing.
The same discipline applies when delivery runs through a partner, which is why white-label engagement models that quote change requests separately keep scope additions off the agency’s margin.
If the numbers come back the way they tend to when an agency works through the five questions above, the next leadership conversation probably isn’t about price.
Frequently Asked Questions
FAQs
How Do I Know If My Margin Loss Is From Pricing Or Delivery?
The cleanest test is the margin variance across similar projects. When comparable scopes priced at comparable rates produce wildly different margins, the cause is downstream of pricing.
When margins are consistently weak across every project, regardless of scope or rate, a pricing floor problem becomes more plausible. The first pattern shows up far more often in agency data than the second.
What Is A Realistic Net Margin Target For A Mid-Size Agency?
Promethean Research puts the average digital agency net margin at around 13% for 2025, with focused agencies and disciplined service mixes pushing into the 20-30% range.
A reasonable target for a well-run mid-size shop is 15-20% net, and moving meaningfully above that almost always requires operational tightening on the delivery side rather than rate-card changes alone.
Should We Switch From Fixed-Fee To Time-And-Materials To Protect Margin?
It can help, but it isn’t a fix in isolation. SoDA’s benchmarking shows T&M projects overrun at lower rates than fixed-fee work, but the delivery discipline underneath the contract matters more than the contract type itself.
A disciplined fixed-fee shop will out-margin a sloppy T&M shop most of the time, because pricing structure only protects margin when the operational practice underneath is solid.
How Often Should We Review the Delivery Margin By Project?
Monthly at the agency level and quarterly at the project level is a reasonable cadence for most mid-size shops.
The lag between a project starting to slip and the margin signal appearing on the P&L is typically four to eight weeks, so less-frequent review usually means the projects that needed course-correction are already closed by the time the data surfaces clearly enough to act on.
