
I started DJing in college, and kept at it when I moved to New York, through my early agency career: college parties, New York bars, even weddings. My marketing was a stack of mix CDs, left for social chairs and student organizations and handed to bar owners around the city. They brought in a lot of work.
Running that business taught me three things I now ask every agency planning its year. Nobody paid me to make those CDs, but they were where the bookings came from. When formal season hit, I turned work away because I had no bench. And at about $500 a night, I priced the night, not the full cost of the gear, the help, the gas and the setup.
A one-person DJ business and a 100-person agency run into the same problems. The agency’s are just bigger, and they cost a lot more.
Most agency plans for 2027 start with a revenue target. The better ones start with the scopes of work. A revenue target tells you what you hope to make. The scopes tell you who’s paid for, who isn’t, and which months you’ll run short of people. That’s where the margin is won or lost.
Here are the ten questions I start with when I help an agency plan its year. They’re written for mid-size and larger agencies. If you’re a startup or a small shop, some will apply now and others will matter as you grow. They’re not the whole picture, but the ones that are hardest to answer usually show where the plan has gaps.
Everything else in the plan follows from this.
Your scopes set your projected revenue and hold the flighting for every project. The monthly forecast shows your busy months, your quiet ones, and when cash and staffing get tight.
Put every role into one place, a master spreadsheet or a staffing tool, January through December. That includes everyone in the agency, plus the ghost roles you plan to hire but haven’t yet. Then project the work by week, with every project and every staffing allocation against each person. That’s your capacity plan. It shows where people are overbooked, where they’re underused, and where you’ll need help, before it happens. Plan on about 1,800 hours a person, not 2,080. Nobody actually has 2,080 hours to bill, and a plan that assumes they do is wrong from day one.
Map the scopes to the people. Some are fully covered by client work, some partly, some not at all. Do the same by client. Your biggest client may not be your most profitable one.
The same capacity view shows where you’re short of people, and when. You won’t book a freelancer four or five months out, but you can spot the gaps early, know which skills you’ll need, and line up the people you’d call. That way you’re not scrambling at the last minute.
It’s casting. Based on how people are paid for and the new capabilities you’re adding, put each person on the work they’re best at, not just whoever’s free.
New services need people with the right skills, and time to build them, before a client pays for the work.
Investment hires ahead of revenue. Non-billable roles that are still essential, like finance, HR, operations and new business. Training. Capital expenses like equipment, studio or office space. Software and technology: what you pay for, what people actually use, and what you can cut. AI: which tools, who’s trained on them, and what they change in how you staff and price. Decide what’s worth it up front, and put it in the P&L.
A website overhaul, if you’re doing one. Case studies and case study videos: how many, and who’s making them. Award submissions: which shows, what the entries cost, and who’s writing them. Your content strategy: who’s developing it, and how much you’re producing. Your sales and new business pipeline, and a new business budget for pitches. PR, and your overall corporate marketing strategy. Plan these like client projects. They’re often more work than one, and they’re rarely in anyone’s plan.
Build a projected P&L with all targets: revenue from your scopes, against the cost of everyone in your plan, freelancers, tools and overhead. Then check the levers: how many non-billable roles your revenue can support, who should be fractional, and what utilization targets your department heads have. Are they leaning in and doing the work?
Client concentration. What share of revenue comes from your top one or two clients?
Signed vs. upside revenue. How much of next year is signed, and how much is upside you haven’t won yet? For each piece of upside, what’s the realistic chance it converts, and what’s it worth once you weight it by that chance?
Contract renewal dates. When does each contract end or come up for renewal?
Payment terms and cash flow. Revenue isn’t cash. On net 60 terms, January’s work is paid in March. Depending on when you’re paid, terms and pass-through costs can put real stress on the P&L.
People costs that rise without new hires. Raises, promotions, benefits and replacing people who leave.
Time to hire. If a role takes three months to fill, start before the work arrives.
Pricing. When did you last raise your rates? And if AI makes your team faster, does hourly pricing still make sense?
Scenario planning. A plan for the expected year, one for losing a big client, and one for winning one.
Plan vs. actual. Who checks the plan against reality each month, and what triggers a change?
Annual planning comes down to trade-offs. The P&L tells you which ones you can afford.
Building that master view is usually the first thing I do with an agency. I’ll go deeper on some of these in future posts.
How should an agency start its annual plan? Start with the business strategy and the scopes of work, not a revenue target. The scopes set projected revenue month by month and show when cash and staffing get tight.
How many billable hours should an agency plan per person? Plan on about 1,800 hours a person, not 2,080. Holidays, time off, training and internal work mean nobody has 2,080 hours to bill.
What is a ghost role in agency planning? A role you plan to hire but haven’t filled yet. Putting ghost roles in the staffing plan shows the cost and timing of hires before the work arrives.
What should an agency’s P&L plan include? Revenue from scopes, the cost of everyone in the plan, freelancers, tools, overhead and investments no client pays for, with an EBITDA target to check the plan against.