It usually starts with a Tuesday morning. Your project manager opens the sprint board, stares at it for a moment too long, then walks over to your desk. “We said we’d deliver the new homepage by the end of the month. That’s not happening.” You already knew, somewhere in the back of your mind. You just hadn’t said it out loud yet.
Overpromising is the agency industry’s original sin. It’s not born from dishonesty — it comes from an absence of real data. When someone asks whether your team can take on a new project, most agency owners are making an educated guess based on gut feel. Sometimes the gut is right. Often it isn’t, and that’s when late nights, rushed work, and awkward client calls follow.
The fix isn’t to promise less. It’s to plan better — with actual numbers, actual visibility, and actual processes. Here’s how to build that.
Why Agencies Overpromise (It’s Not What You Think)
The instinct is to blame the sales team for committing to unrealistic timelines, or leadership for chasing revenue at the expense of delivery quality. But in a small agency of five to fifteen people, the root cause is almost always structural: you have no single, reliable view of where your team’s time is actually going.
Think about the typical pipeline. A prospect says yes on a Friday. The MD calculates that between three staff members, there are “probably enough hours” to start next week. Nobody checks whether those same three people are already running a website rebuild, handling a support queue, and chasing a client for sign-off on something from two months ago. The new project gets added to the pile, and the pile quietly gets heavier.
The second failure mode is underestimating non-billable time. In agencies that track hours at all, the tracking tends to cover project work. It rarely captures the account management overhead — the status calls, the revision loops, the inbox time. For a typical agency relationship, that invisible overhead can consume 20–30% of your real capacity without appearing anywhere in your scheduling assumptions.
The rule of thumb: If you think your team has 40 billable hours per person per week available, the realistic number — accounting for meetings, internal admin, and non-project tasks — is closer to 28–32. Plan from that figure, not the theoretical ceiling.
The Utilisation Number You Should Actually Target
Utilisation is the ratio of billable hours to total available hours. At 100% utilisation, everyone is billing every hour they work. That sounds ideal. It’s actually a warning sign — it means there’s no slack for training, business development, or the unexpected revision that lands on a Thursday afternoon.
The target range for a healthy digital agency is 65–75% billable utilisation. Below 60%, you have a revenue problem. Above 80%, you have a sustainability problem — the team is burning, quality degrades, and good people start polishing their CVs. The 65–75% window is the zone where work gets delivered properly, people aren’t running on empty, and there’s room to absorb the normal turbulence of agency life.
Where this gets complicated is that utilisation isn’t uniform across roles. A senior developer at 70% is healthy. A project manager at 70% often means client-facing work is being deprioritised. A new hire at 70% in their first month might mean they’re not getting enough support. You need to look at utilisation by role and by individual, not just as a blended team average that hides the outliers.
Tracking this properly requires time data. Not just invoiced hours — all hours. If you’re not already running a time tracking system that captures both billable and non-billable time across every project, that’s the first thing to fix. Everything else in capacity planning rests on this foundation.
Building a Simple Capacity Model
You don’t need specialist resource management software to start. A functional capacity model can be built in a spreadsheet — though you’ll quickly find a proper tool pays for itself in avoided overruns. Here’s the logic:
- List every person on the team with their total available hours per week (standard working hours, minus approved leave).
- Apply a realistic utilisation ceiling — typically 70%. A developer with 37.5 hours per week has roughly 26 billable hours available.
- Map every active project against those people, with a weekly hour estimate for each. Include support retainers, recurring tasks, and internal projects — not just new delivery work.
- Calculate the gap: total committed hours minus total available hours, per person, per week. This is your capacity view.
- Extend it four to six weeks forward so you can see where you’re approaching the ceiling before a new brief lands on the desk.
The critical step most agencies skip is step 3 — they model new projects against gross capacity rather than subtracting existing commitments first. If your developers are already allocated 24 of their 26 available hours across current work, a new project that “only needs 10 hours per week” creates an immediate problem. Doing this calculation in advance is the difference between catching that problem before you commit and discovering it after kick-off.
The Signals That Tell You You’re Already in Trouble
Even with a model in place, capacity problems have a habit of emerging between reviews. These are the early signals worth watching:
- Milestone slippage on more than one project simultaneously. A single overrun can have a dozen causes. Multiple concurrent overruns almost always mean the team is too thin.
- Tickets and support requests going unanswered for more than 24 hours. Support work is often the first casualty when project pressure rises. This tends to damage exactly the relationships you can least afford to lose.
- Overtime becoming normalised. A late push once in a while is part of agency life. If more than two people are regularly logging evenings or weekends, something structural needs to change.
- A growing backlog of internal tasks (invoicing, reporting, QA). When admin starts slipping, it’s a sign the team doesn’t have capacity for anything that isn’t on fire.
- The project manager is doing delivery work. This is often invisible in the numbers but obvious on the ground. If the PM is writing copy or building pages to fill gaps, your capacity model is already broken.
The challenge with these signals is that they’re lagging indicators — by the time they’re visible, you’re already in the problem. The goal of proper capacity planning is to make them leading indicators instead, surfaced before commitments are made rather than after.
How to Have the Capacity Conversation with a New Client
One of the most valuable habits you can build is treating every new project enquiry as a capacity check before it becomes a proposal. This sounds obvious; in practice, most agencies don’t do it because they don’t have the information to hand.
When a prospect says they need a new site by October, the first question should not be “what do they need?” It should be “do we have the hours?” Then: “do we have the right hours?” (A ten-week development project when your only available developer is on two weeks’ holiday in September is not a deliverable project for October, regardless of how many total hours exist on paper.)
Being honest about capacity is also a quality signal to clients. An agency that says “our next available start for a project of that scope is in four weeks — here’s what we can do in the interim” comes across as organised and in-demand, not turning work away. Compare that to an agency that says yes immediately, then spends the first two weeks scrambling to reassign work. Clients notice.
If you’re regularly having to turn down work or extend timelines, that’s also useful data. It means you’re operating close to full utilisation — which is often the right time to think about whether you need to hire, raise rates, or restructure your retainer mix to smooth out the demand curve.
Resource Scheduling: From Bird’s-Eye to Week Level
Capacity planning at the strategic level (can we take on X?) needs to connect to scheduling at the operational level (who is doing what on Tuesday?). These two views often live in separate places — a Gantt chart for project timelines, a spreadsheet for staffing, a task board for day-to-day work — and the gaps between them are where overruns hide.
The practical solution is to maintain one scheduling view that shows both project commitments and individual allocation, updated at least weekly. In that view, you want to see:
- Each team member’s committed hours for the current week and the next two weeks
- Which projects those hours are assigned to
- Any flagged risks (a project running behind that may need more hours to recover)
- Leave and public holidays affecting availability
The frequency of review matters as much as the structure. A monthly capacity review is almost useless in a fast-moving agency — things change too quickly. A weekly check-in, even a brief one, keeps the view accurate and surfaces problems while there’s still room to respond. Pair this with a fortnightly pipeline review where you assess likely new project starts against projected availability, and you have a planning cadence that actually works.
Using Your CRM to Close the Capacity Loop
Capacity planning fails when it lives in a different system from the rest of your agency data. If your time tracking is in one tool, your project milestones are in another, your pipeline is in a spreadsheet, and your retainer commitments exist only in email threads, no single person can get a complete view without manually assembling it. That’s exactly the situation that produces confident over-commitment.
The solution is to bring these data streams together. An agency management platform that combines time tracking, project management, retainer management, and pipeline visibility in one place gives you the information to make capacity decisions in real time, not in retrospect.
Marque CRM’s resource scheduling and utilisation reporting modules are built precisely for this. You can see billable versus non-billable hours by team member, track utilisation across the team, manage retainer hours, and link project milestones to the same system where work is tracked. When a new opportunity comes in through the pipeline, you’re making the capacity call with real data — not a guess.
The Gantt chart view gives you the bird’s-eye project timeline. The time tracker captures where hours are actually going. Retainer management shows you your recurring commitments. And because it all sits in the same platform as your CRM, project, and invoicing data, there’s no gap between what’s been sold and what’s been scheduled. That’s the loop you need to close to stop overpromising delivery.
Practical Starting Points for This Week
You don’t need to overhaul your entire operation to start making better capacity decisions. Pick one of these this week and build from there:
- Calculate your team’s real available hours, applying a 70% utilisation ceiling, and compare that against your current project commitments. The gap (or lack of one) will tell you a lot.
- Audit your last three project overruns. Were they caused by scope creep, poor estimation, or genuine capacity shortage? The root cause changes what you fix.
- Book a fortnightly capacity review in the calendar — 30 minutes, the whole team, looking at the next four weeks. Consistency matters more than sophistication here.
- Start tracking non-billable time if you aren’t already. You can’t manage what you can’t see, and the invisible overhead is usually larger than anyone expects.
The agencies that scale sustainably — that maintain quality, keep good people, and convert clients into long-term relationships — are not the ones that say yes to everything. They’re the ones that know what they can actually deliver, communicate that honestly, and have the systems to back it up. Capacity planning is not a constraint on growth. Done properly, it’s the thing that makes growth possible.