How to Build and Set Better Marketing Budgets

Me: Is your goal to grow sales or decrease costs?

Client: Both.

Me: *audible sigh whilst shaking head disapprovingly*

Of course you want it all. You need to make more revenue AND save money on marketing. Who wouldn’t want to do more with less?

Sorry, but “priorities” is a one-word oxymoron.

It’s priority; singular. As in, out of the millions of things you could do, what’s the one thing you should be doing?

This isn’t just semantics; having more than one priority means you can’t choose between them when one conflicts with the other.

That conflict is inevitable.

Spending more on Facebook ads might increase your sales volume, but it’s guaranteed to hurt your ROI if you’ve grown mostly organically to date.

Writing more blog content will grow your traffic, but usually hurts conversion rate because those visitors are higher up the funnel and not ready to buy just yet.

Increasing conversion rate at one stage of your marketing funnel commonly hurts conversion rate at the next stage, or has an adverse affect on lead quality.

So whether your priority is increasing sales volume, better ROI, growing traffic, improving conversion rate or getting higher quality leads, you will have tradeoffs to make.

You can’t optimize for all of these metrics at once. It’s too complex a problem to solve and you’re very likely to end up failing on all fronts.

So how exactly do you set a better marketing budget?


Download a PDF version of this article!


Solve for X

Solving for one variable is easy.

If I told you to go spend $10,000 on a car, you could do it without a problem.

If instead I specified that it must be a black BMW 3-Series, 2014 model or newer, with leather seats and fewer than 60,000 miles on it, the task becomes a lot harder, if not impossible.

The more variables you add, the more difficult the task. Too many variables, and there might not even be a possible solution that satisfies all your criteria.

Say your goal is to grow revenue 60%.

You have $10,000 to spend on marketing. Let’s say that gets you 50% of the way there.

Now you have choices to make.

  • Do you try to improve conversion rate? If you convert more of your incoming traffic, that might get you closer goal.
  • Maybe retention can be improved, which would mean more of your goal coming from existing customers.
  • Potentially your advertising efficiency can be improved through creative testing, so you squeeze more out of that $10k budget.

An experienced growth marketer can help you figure this out, but the point I’m making is that even a simple goal like ‘grow revenue 60%’ is incredibly difficult and complex to solve.

What’s the Counterfactual?

Maybe it’s impossible to solve.

US GDP growth was 1.6% in 2016. In simplified terms, for every company growing 60%, there are others making losses, balancing that out.

What makes you special? If you aren’t doing anything really above average, then why would you expect to grow faster than average?

If you’re a startup, 9 in 10 of startups fail. They all planned for 60% growth at some point too. Even 88% of the Fortune 500 dropped out over 60 years. That trend is accelerating – the average lifespan of a Fortune 500 firm has been cut from 75 to 15 years, and that trend is continuing downwards.

All of these firms failing entirely… and you expect to grow?

Maybe your product is actually destined to decline in revenues by 60% because a new competitor has arrived on the scene ready to blow your product out of the water.

This is called the counterfactual — i.e. what would have happened had you not taken the actions you took.

This catches everyone out.

Time and time again I see it; a marketer is happy with a series of successful improvements, and their boss disagrees – “If you really did move the needle, why am I not seeing it in the top line?”.

Often it’s because, without those improvements, performance would have been down.

Past results are not a good indicator of future success. Just because you grew 60% last month, doesn’t mean you aren’t destined for a 60% decline next month,

In that scenario, where the default is a decline, you have to ‘grow’ 60% just to stand still. Suddenly your reasonable 60% goal is actually an unreasonable 120% growth goal.

It’s like being given a goal not to buy a single black BMW, but to grow your 10 car fleet by 10%. If a car breaks down, you now need to buy 2 black BMWs just to hit your original target.

Time is a Factor

Of course it gets harder when time is a factor.

If I told you I absolutely needed that black BMW today, you’d have to rush to the nearest dealership and pay any price they asked.

If you had a whole year, you could spend some time casually browsing online auctions until you found an absolute steal.

The shorter the time frame, the more expensive the unit cost.

It works the same in marketing. $10k a year will get you a much better ROI than $10k a day.

Volume is Key

Volume is the other side to the equation.

One Black BMW a year is easy, but if I asked you to buy 10,000 black BMWs this year, you couldn’t do it alone. You couldn’t go dealer to dealer yourself. You’d need an expensive procurement and legal team to negotiate with big-time suppliers.

At scale, risk increases too. I could spend all year negotiating with a supplier and the deal could fall through at the last minute. In fact, the more points I need to negotiate, the higher the likelihood of that happening. That’s yet another reason why you need to keep your goal simple

As marketers, we can’t simply go from $10k a year to $100k a year without building all the infrastructure that goes with it. A lot of those costs are fixed and take a long time to put in place.

Time, Cost, Quantity

This simple relationship between volume, time, and cost catches out many marketers and business people who think that they can double spend and get the same ROI.

In reality, it’s a tradeoff.

time cost quantity

All three are related, but you can only pick two.

  • Big sales target (quantity) in a short space of time? Expect low ROI (cost).
  • Need high ROI but still want to hit that high sales target? Be prepared to take your time doing it.
  • Can’t take it slow and still need to maintain a high ROI? Lower your sales target.

This is a fundamental law in digital marketing.

There are some domains where you see increasing returns to scale, such as building a factory that allows you to mass-produce a core component for a cheap cost per component.

You see this in the supermarket or chains like Costco, where you can buy food in bulk to get a cheaper price per unit.

Some marketing even works this way. For example WPP, a British ad agency, negotiates with big TV networks to get a good price on running ads for their agencies and brands.

However, digital marketing works on the auction model. Unlike calling up BMW and negotiating for a fleet of cars, you’re attending car auction after car auction to buy up whatever cars are available on the market.

More accurately, it’s like an automated eBay, where you’re competing in multiple auctions simultaneously based off a series of rules to get what you need to hit your goal.

As an auction-based marketplace, it is subject to the laws of supply and demand. Higher demand or limited supply make the cost per unit go up.

Diminishing Marginal Returns

This translates to the law of diminishing returns; the more volume you need, the more it costs you per unit.

Therefore it’s easy to guarantee a specific spend level in digital marketing OR a specific performance goal (ROI, CPA), but it’s very difficult to hit both.

We could spend the full budget but we might not get good performance. On the other hand, we could keep to the ROI target, but only at the expense of volume by turning off or bidding down the worst campaigns.

Here’s what that looks like in practice:

spend vs volume chart

That chart is from real campaign data and we use this method ourselves to predict and budget performance. You can replicate it yourself and you’ll almost always see this pattern.

If you have a budget in mind, you can choose a point on the chart and it’ll tell you your volume. If you have a volume target, you can divine your cost per conversion.

It’s not a perfect system, and there’s plenty of noise, but it’s the closest we can get.

Escape the Law

There is a way to escape the law of diminishing marginal returns. If you test a new way of doing things and it performs fundamentally better, the curve lifts upwards.

It could be new creative or messaging, a different way of targeting your audience, or a change to campaign structure or user journey.

But the only true way to improve performance for a well-optimized campaign is through testing.

It’s important to run marketing tests because we don’t know ahead of time what will move this baseline. Yes, we have a good idea of what’s best practice (or not), but typically 9/10 marketing tests fail.

So if our target requires us to hit a certain quantity in a specified time and for a certain cost (all three sides of the triangle), we can’t do it without finding a successful test that fundamentally shifts the baseline.

Certainly that can be done, but if only 1 in 10 tests succeed, we can’t really predict ahead of time what combination of channel, audience, creative, and copy will actually work to hit that goal.

Realistic Goals

So in practical terms, how do we budget for our goals effectively given what we know?

After doing this for 120+ startups at Ladder, here’s my process:

1. Research Benchmarks

Figure out what’s “reasonable”. Are your competitors growing fast? What’s the normal conversion rate for your industry? How fast have you been growing in the past? Any assumption in your budget should be backed by research with attribution links so others can challenge it.

2. Establish the Counterfactual

Given the above research, your planned product development, and competitive movements, what is likely to happen with no improvement? Be relatively pessimistic; no sandbagging. Remember, this isn’t your goal. You’re just trying to establish a worst-case scenario.

3. Draw from Existential Goals

What needs to happen to avoid massive repercussions for your business? If you need to hit a target to get your next funding round, or you’re unprofitable and need to stave off bankruptcy, be honest with your team. A meaningful goal beats an arbitrary goal every time.

4. Carve out Testing Budget

I recommend writing off 20% of your budget for testing. Like the military black ops budget, you don’t know what it’s being spent on, but you trust that it’s needed. Don’t worry about being efficient here; the goal is to explore the unknown and find ways to improve your baseline.

5. Estimate Test Success

You might not know what you’re going to test yet (if you work on an agile process, that’ll get worked out on the fly). But you can estimate a rough number of experiments and an average success and impact rate. That gives you an aggregated idea of performance from testing.


Take these steps and you’ll already be better at budgeting and goal setting than 99% of the marketers I know.

It’s a tough job, but it’s honestly the easy part.

The tough part is motivating your team to hit the goal.

BONUS: Creativity and Motivation

Make sure you know what you’re asking. Simply changing the 10% figure to 60% in Excel isn’t difficult. And yeah, of course it would be great to grow faster.

But it has repercussions.

It’s not leadership to pin a number on the wall. You have to give your team the resources and motivation to hit it. This type of work is fundamentally difficult to do. It’s what Dave Pollard calls a ‘Wicked Problem’, or a ‘Complex’ problem in Snowden’s Four Ontologies.

snowden four ontologies

It requires a lot of research, creativity, and the ability to handle ambiguity. This isn’t something you can churn out of a factory. It’s more like an art.

A good growth marketer is a Renaissance woman or man; an artist and a scientist, making fundamental scientific breakthroughs to deliver art the world hasn’t seen before.

If you tell Michelangelo exactly what to paint in the Sistine Chapel, or that it’s just to cover up the cracks in the ceiling, you won’t get a good result.

But tell him to paint the ceiling “for the greater glory of God and as an inspiration and lesson to his people” and you end up with this:

sistine chapel michelangelo

And that’s the mistake people make. They think that because digital marketing is more measurable and scientific, it isn’t a creative endeavor.

On the contrary; the constraints make it more creative than ever. Because everything is so optimizable, the only way to move the needle is to come up with something fundamentally unique versus your competition.

Now isn’t the time for optimization. Save that for when you’ve already struck gold and need to scale an efficient mining operation.

Finding new growth tactics that work is like panning for gold. You can only improve the efficiency if you knew where the gold is. If you knew where the gold was, then you wouldn’t be panning in the first place. If it was obvious to find out, then others would have beaten you to it.

As a creative process, testing is messy. There are times when you make no progress. Other times, you leap light years ahead seemingly overnight. It’s unpredictable. It doesn’t fit easily into your budget spreadsheet.

A big goal backed by a big increase in resources and the promise of sharing in the spoils is invigorating. Built from sound reasoning or existential goals (hit this target or we go bankrupt), a big goal can be hugely motivating.

But I’ve never seen a good marketer stop growing just because they hit the goal – usually, they’re more ambitious than their bosses.

I have seen the opposite, however; when given an arbitrarily set goal with no real logic and not based in reality, the best marketers just turn off.

Or worse, they try to hit the goal anyway because they trust you and want to do well… but they don’t. Instead, they burn out trying to do the impossible, and either leave or start phoning it in.

Asking someone to do more with less (without a good reason) is a recipe for failure. It’s all about the framing. There’s a big difference between “let’s see if this is possible” and “you have to do this”.

So be careful with the goals and budgets you set. Understand what you’re asking. Be reverent towards your team, don’t insult them with arbitrary goals, and make sure you’re doing everything in your power to motivate them.

“If you want to build a ship,

don’t drum up the people

to gather wood, divide the

work, and give orders.

Instead, teach them to yearn

for the vast and endless sea.”

Antoine de Saint-Exupéry, the author of The Little Prince. Found on Netflix’s culture page.