The Agency Ops Playbook for 5-10 Person Teams

The jump from 5 to 10 people breaks most small agencies. You were managing everyone before. Now you have project managers managing teams.

You have multiple projects running in parallel. Revenue might be $300K+ monthly.

And you have no idea what's actually happening on half your projects until the client complains.

This is where operations matter. Not process documentation or six-week planning cycles. Just operational visibility that prevents disasters.

You Need a Project Manager (or Three)

At 5-10 people, you can't manage client relationships and delivery yourself. You need a project manager who owns each major client account.

A good project manager is worth $15-20K monthly in revenue they save by preventing scope creep and timeline slippage. If your project manager costs $50K annually, they pay for themselves in three months.

What does your PM do?

Owns the client relationship (primary contact)
Tracks timelines and dependencies
Surfaces risks early (project is going over budget, timeline is tight)
Manages revisions and scope changes
Communicates with delivery teams weekly

The PM isn't a project coordinator - they're part strategist, part manager, part therapist. They unblock issues and keep projects moving.

Implement Weekly Project Status Reviews

Running 10-15 concurrent projects means projects fall through cracks constantly. Someone isn't communicating about blockers. A deliverable slips two weeks without anyone noticing.

Have your PMs (or you) run 15-minute weekly standup meetings per project. In the standup:

Green/yellow/red status (is this project on track?)
What got done this week
What's blocked or at risk
What's needed from the client

Use one tool for this. Pull all projects into one view (Huddle does this well, pulling from Asana, ClickUp, Linear, etc.).

Look at status weekly. Address red projects immediately.

This one habit prevents 80% of project disasters.

Define Clear Decision Rights

At 5-10 people, someone's always asking "can we do this?" or "should we say yes to that?"

Create a simple decision matrix:

Client requests under $1K and within scope: PM approves
Client requests $1-5K: You and PM approve together
Client requests over $5K: You decide
Changes to timeline: PM can push up to one week, longer changes need your approval
Service requests outside core offerings: You decide

Write this down and share it with your team. Now people stop asking permission for every small decision.

Track Revenue by Project and Client

With 10+ projects, some are profitable and some are disasters. You probably don't know which is which.

Track monthly: revenue per project and cost per project (billable labor). That number tells you profitability.

If a project is $40K revenue with $20K labor cost (50% cost), you're healthy. If it's $40K revenue with $35K labor cost (88% cost), you're losing money on that client.

Kill projects that are bleeding margin. You can't grow while subsidizing bad client relationships.

Build a Bench for Your Busy Season

Most agencies run feast or famine - either slammed for three months or slow for three months. This creates constant hiring and firing.

At 5-10 people, plan for your busiest month. If you're usually slammed September through November, build a bench of contractors for those months.

Contact them in July: "Can you commit to 40 hours weekly September-November?" Get commitments before you need them. They'll be happy for the steady work. You'll have capacity without hiring permanently.

This costs 5-10% more than hiring temps on the spot, but you get better people and no scrambling.

Stop Saying Yes to Everything

This is the hardest part. Most agencies grow by saying yes to everything. Now that you're 5-10 people, you need to start saying no.

Create a filtering system for new projects:

Is it in your service area? (If no, decline)
Is your project manager available? (If no, decline)
Does it fit your pricing model?

(If below your minimums, decline)

Is the client healthy? (If they demand, complain, or don't pay on time, decline)

You'll probably decline 30-40% of projects that come in. Your margins will improve. Your team will be happier because they're not on terrible projects.

Create a Standard Project Kickoff Template

When you're managing five projects yourself, kickoffs are quick. With 10+, they become a bottleneck.

Create a kickoff meeting agenda:

Introductions (client, team members, roles)
Goals and success metrics (what does done look like)
Deliverables and timeline (dates, milestones)
Communication process (how often do we talk)
Revision process (how many rounds, what happens after)
Escalation path (who do they contact if something's urgent)

Run the same kickoff for every project. Saves two hours per project in scope clarification.

Establish a Monthly Financial Review

Pull your metrics monthly:

Total revenue
Revenue per project
Revenue per employee
Average project margin
Top three most profitable clients
Bottom three least profitable clients

Spend 30 minutes reviewing. If any client is under 40% margin, make a plan to fix it (raise prices, reduce scope, or fire them).

This discipline prevents small margin problems from becoming existential crises by year three.

FAQ

When should I hire a project manager?

When you have 5+ concurrent projects and you're spending more than 20 hours weekly on project management. At that point, hire a dedicated PM. You'll either improve project delivery or realize you need ops help.

What if my project manager isn't working out?

Six months in, you'll know. Good PMs are communicators who can manage up (to you) and down (to the team).

If they're not, replace them. A bad PM kills more value than a good one creates.

How do I handle projects that are clearly going over budget?

Talk to the client immediately, not at the end. "We're tracking toward 15 extra hours because of scope expansion.

Do you want to (a) extend the timeline, (b) reduce scope, or (c) pay for the extra hours?" Most clients choose (b) or (c). You never lose money by talking early.

Should I specialize or stay generalist at this stage?

You can be successful either way. But specialization is easier to manage operationally.

If you do "brand and web" really well, PM onboarding is simpler. If you do everything, you need more experienced people.