TL;DR: Capacity planning is how a practice works out how much client work its team can realistically take on, and who should do it. Most practices plan against 7–8 billable hours per person per day. The RIBA’s own benchmarking shows partners and directors spend less than half their time on billable work. Get that number wrong, and projects run over budget before the first drawing is produced. WorkflowMAX’s Capacity Planning module makes the real picture visible.
Capacity planning is one of the least visible margin problems in an architecture practice. Most Directors price commissions against an assumed 7 or 8 chargeable hours per person per day. Before the project has started, that assumption is already wrong.
This isn’t a new problem. It’s a structural one. The gap between how practices think their teams work and how they actually work is wide, consistent, and well-documented. What’s less common is a practice that’s done the maths and adjusted accordingly.
This post runs through what capacity planning actually means in a practice context, why the standard assumptions are so far off, what’s genuinely eating your team’s available time, and what a working capacity model looks like. We also cover WorkflowMAX’s Capacity Planning module for practices that are ready to move from a spreadsheet to a system.
What is capacity planning for an architecture practice?
Capacity planning is the process of working out how much client work your team can realistically deliver over a given period, based on who’s available, what they’re already committed to, and how much non-project time they carry. For a practice, it’s the discipline that sits between winning a commission and actually delivering it without burning margin.
It isn’t the same as project scheduling. Scheduling tells you when tasks happen. Capacity planning tells you whether the team can absorb the work in the first place. Without it, you’re running two disconnected systems: a fee-proposal process that assumes the team can deliver, and a project process that discovers at the halfway point that they can’t.
Across professional services, the target utilisation rate sits between 70% and 80%: high enough to drive revenue, with enough headroom for the overhead every practice carries. Most UK architecture practices don’t track this number at all.
The utilisation rate trap: why 7–8 hours is a fiction
Start with a standard full-time week: 37.5 hours. Planning at 7 or 8 billable hours per person per day implies utilisation of 93–107%. That’s not a planning assumption. It’s a fiction.
Begin with what comes off the top. Statutory and contractual leave in a UK practice typically runs to 28 days plus eight bank holidays: 36 days, around 270 hours, gone before the year starts. RIBA’s mandatory CPD requirement adds a further 35 hours per year. Add an average sick leave of around five days. You’re already more than 340 hours below the theoretical maximum before a single internal meeting has been counted.
Then layer in the weekly overhead. Data from a 450-firm study of architecture and engineering practices puts the average annual billable hours per professional employee at 1,600–1,900, not the 1,950 hours a 37.5-hour week implies. [Benchmarking data from Monograph](https://monograph.com/blog/unlocking-utilization-rates-benchmarks-for-architects-and-architecture-firms) breaks it down: general admin accounts for around 11.5% of working time, business development around 5.3%, and paid leave around 8.7%.
Now look at the people running the practice. The RIBA Business Benchmarking Report is unambiguous: partners, directors, and sole principals spend less than 46% of their time on billable work. That’s under 3.5 chargeable hours per day for the people most often responsible for resourcing decisions.
A 1,000-hour commission priced for three people delivering 7 billable hours a day looks like 48 working days. At realistic utilisation, the same project runs to 70 or more. That’s the trap. Not scope creep. Not overservicing. A wrong assumption, baked in from the start.
What’s actually eating your team’s billable time?
The non-billable categories in an architecture practice are predictable, but practices rarely add them up. They include: annual and bank holiday leave (around 36 days per year), RIBA-mandated CPD (35 hours annually), business development and proposal writing, internal project reviews and meetings, admin and software management, and sick leave. Combined, these account for roughly 25–35% of a working year for fee earners, and over 50% for partners and directors, per RIBA’s 2025 benchmarking data.
The mix shifts by role. An architect delivering a Stage 4 package loses less time to BD than a director managing three live pitches. A Part III-qualified architect with structured learning requirements carries more CPD overhead than a technologist on a fixed programme. Role-level figures matter more than practice-wide averages.
UK accountants who specialise in architecture practices are consistent on this: the practices that manage chargeable time well treat non-billable overhead as a fixed planning input, not as something to deal with after the project runs over.
What does poor capacity planning look like in an architecture practice?
Poor capacity planning shows up in familiar patterns: projects that routinely run over hours with no clear explanation, a persistent “everyone’s at capacity” feeling with no visibility of who’s committed to what, and work landing with whoever has space in their schedule rather than whoever has genuine availability across their whole job list. These aren’t discipline failures. They’re planning failures, and they share a root: not knowing the real numbers.
The most common version is the informal check. A Director looks at the team, makes a judgment call, confirms the commission. Nobody looks at remaining hours on live jobs. Nobody accounts for the leave one person has planned next month. The project goes
The spreadsheet version is just as common. Research across professional services firms puts 44% using no scheduling software at all, with Google Sheets the most common tool at 28%. For a 2-person studio, a spreadsheet can work. For a practice of six or more across concurrent projects, it breaks faster than it helps.
How to build a capacity plan your practice will actually use
A working capacity plan needs three inputs: realistic available hours per person (after planned leave, CPD, and known non-billable commitments), a clear picture of what each person is already committed to, and enough pipeline visibility to plan work before it arrives rather than when it lands. Without all three, you’re making structured guesses.
In practice: set a realistic daily billable hours figure per role, built from actual overhead rather than theoretical hours. Map planned leave and CPD for the next quarter before confirming a commission. Record existing allocations, not just completed work. Then, before the fee proposal goes out, check whether the team can absorb the project at realistic utilisation.
For practices evaluating what tools can hold this together, the critical question is whether the capacity view connects to live job data. If your time tracking, resourcing, and WIP management all live in separate places, you’re maintaining three versions of the same information.
How WorkflowMAX’s Capacity Planning module handles this
WorkflowMAX’s Capacity Planning module gives a colour-coded weekly timeline across the team: yellow for under-utilised, green for the target zone, red for over-allocated. Each person’s capacity is built from their configured working hours, so part-time arrangements, compressed weeks, and variable schedules are reflected accurately from the start.
Leave and non-project time (CPD blocks, BD, internal reviews) go in as capacity-reducing jobs. When someone is on annual leave, their available hours drop automatically. You’re never planning against time that isn’t there.
The module shows remaining hours (estimated time minus planned time) and a capacity percentage per person based on their configured week. Over-allocation is flagged clearly but not blocked: you can see the problem and redistribute before it becomes one. Allocation is drag-and-drop. Moving a task, resizing a block, or spreading hours across a week takes seconds.
This matters for the utilisation trap specifically because the baseline is built from each person’s actual configured hours, not a flat 7.5-hour assumption. You’re planning against realistic capacity from the first allocation. And because it lives in the same system as quoting, time tracking, WIP, and invoicing, a capacity decision in the planner carries through directly to fee recovery and margin tracking.
For a full picture of how it’s implemented and configured for a practice your size, reach out to a member of our team here.
The Utilisation Trap
The utilisation rate trap starts with one wrong number. For fee earners, the real billable day is closer to 5.5–6 hours, not 7–8. For directors and partners, the RIBA’s own data puts it below 3.5. Price a project against the wrong figure, and no amount of project management recovers the margin later.
Getting capacity planning right means knowing the actual available hours per person, knowing what they’re already committed to, and checking both before the commission is confirmed. A good system makes that data easy to hold and easy to read.
If you’d like to understand how WorkflowMAX is configured for a practice like yours, book a discovery call with us. Or if you’d like to assess tool fit before committing to implementation, the App Fit Sprint is the right starting point.
