Before You Set Next Year’s Budget, Ask What Growth Requires

Every fall, marketing teams sit down to build next year’s budget, and I’ve sat in a lot of those rooms. Last year’s spending is right there, already formatted, already approved once. Next year’s growth is a blank column with an attitude.

The process usually sounds something like this. Pull last year’s numbers. Check what the ad platforms and the CRM said worked. Then go around the table.

“What do you think we can get from Meta?”

“I wouldn’t go above $225K a month. That seems to be the sweet spot, and I think we can hold a $60 cost per lead.”

That gut call isn’t the problem. The person closest to a channel often knows things no dashboard shows. I once watched a marketing mix model recommend more than tripling spend in a single channel. The people who had bought that channel for years said flatly it wouldn’t work. The model was reaching far beyond any spend level it had ever seen, and their experience was dismissed as bias. A test would have settled it. Instead, the model won the argument. It did not win the year.

Statistics without business judgment is just confidence with decimals. But the opposite trap is just as common: last year’s number and a few gut calls quietly become the whole plan.

I love a good spreadsheet. I just don’t like betting next year’s growth on “last year, plus a little.”

Why marketing budget planning starts in the wrong place

There’s a name for it: anchoring bias. Last year’s number becomes the starting point, and the whole meeting goes to debating how much to change it. Nobody stops to ask whether the plan behind the number still makes sense.

It isn’t just the rooms I’ve sat in. When The CMO Survey asked 2,791 senior marketers in 2021 how they set their budgets, the most common answer, at 41%, was last year’s budget, adjusted as needed. Only 28% built it from scratch around new objectives. And the squeeze is real: in Gartner’s 2026 CMO Spend Survey of 401 marketing leaders, budgets averaged 7.8% of revenue, barely up from 7.7%, and 56% said they lack the budget to deliver their strategy. That average comes mostly from companies above $1 billion in revenue. Direct-to-consumer brands play a different game: in one review of 10 public DTC and consumer brands’ 10-K filings, selling and marketing ran a median of about 13% of revenue, with a few above 20%, and DTC health can run far higher.

The evidence those budgets anchor to has its own blind spots. The ad platforms report performance through their own lens. Last-touch attribution in the CRM credits the final click and ignores everything that made the customer ready to click. And gut can carry assumptions nobody has questioned in years. Each one is useful. None of them is a plan.

Some years, a few adjustments are exactly right. Other years, growth needs a different customer, offer or approach. The question that rarely gets asked is which kind of year this is.

There’s a second anchor, and it’s even harder to argue with: the number someone needs. I’ve seen it more than once. A growth target arrives from above, sized for the board, with no plan behind how to reach it and retention nowhere in the conversation. The team says the target needs to come down. The target stays.

What happens next goes one of two ways. Sometimes marketing walks into the same meeting every month to explain why they missed it again and what they’ll try next. That wears good people down fast. Other times, the team finds a way to hit it, usually by leaning hard on promotions to old leads. The number gets made. The margin and next quarter’s pipeline pay for it.

To be fair, sometimes the miss belongs to the strategy, not the target. That’s always worth checking. But when a forecast is really the answer to “what do we want this to say?”, nobody wins, and the budget turns into a monthly apology.

Cheap leads can get expensive in a hurry

The line in my budget that cost the most once looked like the cheapest.

On one acquisition program I worked on, the marketing dashboard looked fantastic. Plenty of leads. Low cost per lead. High fives all around.

Then I walked over to sales. They were spending their days on people who were never going to buy. Not a fun way to spend a workday, and not a great way to grow.

That was the tension built into how we planned. Sales forecasted sales. I forecasted the leads that were supposed to produce them. Somewhere in the middle, the two numbers had to meet, and the handoff turned into the same standoff every time. It didn’t help that we weren’t even counting the same thing. Marketing’s “qualified lead” and sales’ “qualified lead” were two different animals, so before anyone could argue about performance, we had to argue about definitions. “Give me better leads.” “Then convert the ones you’ve got.” We were both right, and we were both measuring the wrong thing.

The budget had done exactly what we’d asked. We’d just asked the wrong question. So we rebuilt acquisition around lead quality, including a predictive lead score that flagged the most valuable leads before sales ever picked up the phone. Margin per lead rose by double digits.

That one stuck with me, because the original numbers were so easy to celebrate. Optimize to one number, and there are consequences above and below it.

The questions a marketing budget can’t answer on its own

The stakes get bigger the more a business runs on marketing. Direct-to-consumer health is a good example. Hims & Hers reported $919 million in marketing expense on $2.35 billion in revenue in 2025, about 39 cents of every revenue dollar. When marketing is that big, the marketing budget isn’t a line item. It’s the business plan.

Media is usually the biggest expense in models like that. But whether media pays out depends on everything around it: price, offer, access, customer fit, follow-up and the capacity to serve. So these days, before I touch the media plan, I ask a different set of questions.

Should we raise prices, even if we sell fewer? And if so, what’s the right price? A price increase can lower volume and still raise margin. It can also quietly push customers to a competitor. Finding the right number takes real pricing trials, not a hunch, and either way the answer changes the forecast. It belongs in the same conversation as the media.

Is it time to open a new location, and where? Sometimes the demand is already there, and we’re already paying to create it. I’ve watched one part-time local consultant double conversion in a market that had plenty of leads and no office. When interest outruns presence, a new location converts demand that was slipping away for lack of convenience. It has to earn back its build-out and operating costs, and it may need some local marketing of its own. But it turns demand we’re already paying for into customers, so the cost per new customer can fall before a single new media dollar goes out. The question is where it adds the most new business without just moving existing customers from one location to another.

Are our cheapest customers costing us the most? Low-cost channels can look great on cost per acquisition or return on ad spend and still bring in customers who churn faster or spend less. The media buyer isn’t doing anything wrong. They’re doing exactly the right thing against the wrong metric. The savings just disappear later, on someone else’s report.

Are we only talking to people who are ready to buy today? John Dawes of the Ehrenberg-Bass Institute estimates that in B2B, only about 5% of buyers are in the market at any given time. Consumer categories turn over faster, but the principle holds: the buyers we can track this quarter are a small slice of the buyers we’ll need next year.

Have we let retention slip while chasing growth? In Gartner’s 2026 survey, loyalty and retention fell 29% since 2024, to under 15% of media spend. The only way I know to find the right balance is to study churn. Sometimes the analysis shows that slowing churn lets the acquisition target come down substantially. When keeping a customer costs less than replacing one, which it often does, part of those savings can fund retention. The point is to test that trade-off, not assume it.

What happens if we go dark to make this year’s numbers? In one long-running Ehrenberg-Bass study of a single Australian consumer goods category, brands that stopped advertising for a year saw sales fall 16% on average, and 25% after two years. Restarting didn’t reliably bring back the old trend. Cutting waste is smart. Going silent has a bill that arrives later.

Can we actually serve the growth we’re buying? If the phones are ringing off the hook, every appointment slot is booked or the warehouse shelf is empty, more media just buys more frustration.

There’s an old line about this, usually credited to Einstein but actually written by sociologist William Bruce Cameron: “Not everything that can be counted counts, and not everything that counts can be counted.” Even the quote has an attribution problem. I find that oddly comforting.

How to set a marketing budget by working backward from growth

I learned the better order on a high-priced health product that insurance didn’t cover. People bought it for the big moments: a cruise, a wedding, a trip to see the grandkids. Makes sense on paper. But those moments already come with a big bill, and sitting next to it, a few thousand dollars for the product looked even bigger. Meanwhile, many of the people who needed it barely left the house, afraid of being caught without it.

So we stopped selling the big moment and started selling everyday freedom. The testimonials moved out of head-and-shoulders interviews and into real life: the golf course, the softball field, a 5K, the grocery store, even axe throwing. The goal was to turn a one-time splurge into a daily investment in quality of life.

The data had one more surprise. One smaller, lighter model was bought mostly by older women. So we stopped marketing the category and started matching each model to the people actually choosing it. Marketing 101, sure. It’s amazing how often Marketing 101 is the thing nobody had time to check.

Only after the message and the audience were right did it make sense to ask how much to spend, and where.

I built that plan by hand, over weeks of spreadsheets. The thinking was the fun part. The spreadsheet gymnastics, less so. But it gave me the order I still plan in today.

1. Start with the growth target and the economics

“Grow revenue 20%” is a wish. A plan explains where the 20% comes from: how much from existing customers, how much from new ones, what margin it has to produce and what it costs to serve. I’ve seen plans hit their revenue target and still disappoint on profit, so I get margin into the room early.

2. Translate the target into customers

Here’s a simple example with illustrative numbers. Say the goal is $1 million in new revenue from new customers, and each one is worth $5,000 in the planning period. That’s 200 customers. If one in five qualified opportunities closes, that’s 1,000 opportunities. At $300 per qualified opportunity, that’s $300,000 in acquisition spend, before sales capacity, timing or margin.

Now the plan has something to argue with. If the close rate is really 10%, it takes twice the opportunities. If sales can’t handle the volume, more media won’t fix it.

3. Find what has to change before more spend will help

Sometimes the constraint is reach. More often, in my experience, it’s the offer, the message, the follow-up or the capacity to serve. I’m all for working hard; I just want the work to have a fighting chance. I fix the constraint before pouring money into the channel that’s easiest to buy.

4. Fund demand for today and for tomorrow

Some customers are ready now. Others need time to notice the problem and trust the brand. In Cultivate and Harvest, I wrote about why funding only the harvest shrinks next year’s crop. Each investment gets a clear job, and a measure that fits that job.

5. Put the unknowns and the tests in the budget

I write down every assumption. Which conversion rate has evidence behind it, and which customer segment is still a hunch? Then I fund the significant few tests, the ones that could actually change the plan, and agree up front on the result that would change it.

Why I’m building Elevate Signal around these questions

I use marketing mix modeling, and it’s a valuable input. But a model only answers the question it’s given. Before it can optimize anything, someone has to decide what we’re trying to improve. More sales? More profitable customers? Better retention? A new location that converts demand we’re already paying to create? Experience is what frames those questions, spots what the model might be missing and decides which assumptions deserve a test.

Think of it like a road trip. If I’m driving from Los Angeles to New York, staring at New York doesn’t tell me much. A week in, I’m in Chicago. Progress, right? Until I check the route and see I went by way of Seattle. I’m closer, but I’ve burned a lot of gas and time for those miles. Looking back is how you check whether the route is working. It just can’t be the only thing you look at.

And honestly? I love this work. Figuring out a business, what its customers care about, where growth is hiding and what’s in the way, is the most fun I have at a desk. What I don’t love is losing weeks assembling data before the real conversation can even start.

So I built Elevate Signal. It brings customers, brand, competitors, marketing and finance into one growth plan. It shows what a price change, a new location or a shift toward retention would mean for the forecast, alongside the media. Where the evidence is thin, it ranks the tests worth running first, and the results update the plan.

What Signal won’t do is take the wheel. Remember that model that wanted to triple a channel? A tool that overrules the people who know the business isn’t decisive. It’s just confidently wrong. Signal is built so the strategist stays in the driver’s seat, and so a recommendation that reaches beyond what the data has seen gets a test before it gets the money. It’s a highly capable tool, and like any tool, it’s only as good as the craftsman holding it.

Signal is built on proven modeling, wrapped in the questions I’ve spent twenty years asking. I’m opening it to a small group of first clients now. Plenty of tools can tell you where you’ve been. I can’t wait to help them map where they’re going next.

Marketing budget planning FAQ

How much should a company spend on marketing? There’s no universal percentage. Gartner’s 2026 average was 7.8% of revenue, mostly among companies above $1 billion in revenue, while public DTC and consumer brands in one 10-K review ran a median of about 13%, and Hims & Hers reached about 39% in 2025. The right number depends on the strategy. A land grab, buying share now because the real money is in keeping customers for years, can justify a much higher percentage for a while, as long as those customers actually stick around. A stable business, a slower-growth plan or one fighting churn usually runs lower. That’s why the question matters more than the answer. Benchmarks are a sanity check, not a plan.

Should a marketing budget be set as a percent of revenue? A percentage is a fine guardrail for what the business can afford. It doesn’t say whether that spend can produce the growth needed. I work backward from customers and economics, then compare the answer to the guardrail.

How should a budget split between brand and performance? It depends on how long customers take to decide and how much growth has to come from people who aren’t shopping yet. Both building demand and converting it need funding, each judged by measures that fit its job.

What if the growth plan needs more money than the budget allows? Make the gap visible. Revisit the target, timing, customer mix, offer and conversion assumptions, and look for changes that improve the economics before adding spend. Then test the assumptions that matter most.

What it’s like now

I still love a good spreadsheet. Mine just start in a different place now: with what growth requires, not with what we spent.

If any of this sounds familiar, the anchored budget, the dashboard that looked great while sales struggled, the questions nobody had time to model, you’re in good company. I’ve sat in that room too. And if you’d like a second set of eyes on what next year’s growth really requires, let’s talk.

---

Erin Barr is the founder of Elevate Marketing and the creator of Elevate Signal. Erin helps growth-stage and midmarket companies turn marketing investment into profitable growth, with experience spanning healthcare, elective medical, insurance, senior markets, ecommerce and consumer goods.