Scenario Planning for Marketing Agencies: A Model You'll Actually Use

Scenario planning for marketing agencies means building more than one version of next year's numbers, on purpose, before you need them. Most agency owners build one budget in January and then spend the rest of the year explaining why actuals don't match it. A real scenario planning process starts from the assumption that at least one big thing will change (a client leaves, a client doubles, a hire doesn't work out) and gets the math ready for all three before any of them happen.

What scenario planning actually is (and what it isn't)

Scenario planning is not a budget with a pessimistic column added to it. A budget is one forecast you commit to operating against. Scenario planning is two or three forecasts built from the same cost structure, each one triggered by a specific, named event, with a decision already attached to it. If you've already built an annual budget, you have the base case. Scenario planning asks what happens to that same P&L if your second-largest client walks, or if the three deals sitting in your pipeline all close in the same month. We cover the base-case build in our guide to the marketing agency annual budget template; this is what you do with it once it exists.

The distinction matters because a budget without scenarios gives you one number to miss. A budget with scenarios gives you a range to operate inside, and a set of decisions you already made in a calm moment instead of a panicked one three weeks into a bad quarter.

Why agencies need it more than most businesses

Agency revenue concentrates and evaporates faster than almost any other service business. A single client at 25 to 35 percent of revenue is common, not rare, and that client can leave with 30 days notice in a month-to-month retainer or 60 to 90 days in a signed contract. On the cost side, payroll typically runs 50 to 60 percent of revenue and most of it is fixed in the short term: you can't un-hire a strategist mid-quarter the way a product business can cut a marketing budget. That combination (revenue that can drop 20 percent in a single notice period, costs that can't drop nearly as fast) is exactly the setup scenario planning exists to handle. We go deeper on the concentration side specifically in client concentration risk for marketing agencies, which is worth reading alongside this if one client is already past 25 percent of your book.

Build the three cases: base, downside, upside

Start from the base case: your trailing 3-month average revenue, run forward at the renewal and growth rates you already know about, against your actual current cost structure. Then build two more cases off the same spreadsheet, changing only the inputs that would actually change.

The downside case assumes your largest or second-largest client leaves on their notice period, with no replacement signed. Model the revenue drop at the client's actual monthly fee, not a round percentage, and apply it starting the month their notice period ends. Keep every cost the same except the ones directly tied to that account, like a contractor retained specifically for that client's media buying.

The upside case assumes your two or three best pipeline deals close in the same quarter, and models the hiring or contractor cost required to actually service that new revenue, not just the revenue itself. Agencies that build upside cases without the delivery cost attached routinely end up less profitable after a growth quarter than before it, because utilization drops while new hires ramp.

Run all three against your breakeven point, not just your profit line. We walk through the breakeven formula in detail in our breakeven analysis for marketing agencies, and it's the single most useful number to carry into a scenario model, because it tells you exactly how far revenue has to fall before cash becomes the problem, not just margin.

A worked example: losing a $30,000 a month retainer

Take an agency at $150,000 in monthly revenue, 55 percent gross margin, and $60,000 in fixed monthly overhead (salaries not tied to delivery, rent, software, insurance). Breakeven revenue, using the formula fixed costs divided by gross margin percentage, is $109,091 a month. Current monthly profit is roughly $22,500 ($150,000 times 55 percent, minus $60,000).

Now model the loss of a single $30,000 a month retainer, 90 days out on a signed contract. Revenue drops to $120,000. Gross profit at the same 55 percent margin is $66,000. Monthly profit falls to $6,000, a 73 percent drop in profit on a 20 percent drop in revenue, because fixed costs didn't move. The agency is still above breakeven, but the margin of safety that was $40,909 a month ($150,000 minus $109,091) has shrunk to $10,909. One more mid-size client leaving in the same window, and this agency is underwater. That's the number a scenario model surfaces that a single-case budget never will, not just whether you survive one loss, but how much room is left for a second one.

Set trigger points, not just projections

A scenario without a trigger is just a spreadsheet. For each case, write down the specific number that moves you from "base case" to "downside case" in practice, and the decision that fires automatically when you hit it. Useful triggers: cash reserve below six weeks of operating expenses (see our guide on how much cash reserve an agency should keep for the math behind that number), a signed client notice of non-renewal, or two consecutive months of utilization below 70 percent. The decision attached to each trigger should already be written, not decided in the moment: freeze hiring, cut one specific discretionary line, or draw down a specific amount of the reserve line before touching payroll.

How often to update the model

Rebuild the base case monthly, using the prior month's actuals as the new starting point. Rebuild the downside and upside cases quarterly, or immediately after any signed contract change (a new client signs, a client gives notice, a renewal comes in smaller than expected). A scenario model that's a year old at your biggest client's renewal date isn't protecting you from anything. This is one of the recurring deliverables we build into a fractional CFO for marketing agencies engagement, because the model only earns its keep if someone is actually updating it before the quarter it was meant to predict.