Most DTC founders I speak to ask some version of the same question. How much should go to Meta, how much to Google, and is TikTok worth it yet? The honest answer is that it depends on your margins, your creative output and how much demand already exists for what you sell. But "it depends" is no use when you have a budget to spend on Monday. So here is how I approach DTC paid media budget allocation, with starting numbers you can use and the rules I use to move money once real data comes in.
Give each channel a job before you give it a budget
The biggest mistake I see is treating Meta, Google and TikTok as three interchangeable places to buy sales. They aren't. Each does a different job, and if you don't know which job you're paying for, you'll judge it against the wrong target.
Meta creates demand. For most DTC brands, people weren't looking for your product before they saw it. Good creative in front of the right person makes them want it. That's prospecting, and it's where the bulk of new customers usually come from.
Google captures demand. Brand search, Shopping and category searches catch people who already know what they want. It tends to convert well, but its ceiling is set by search volume. You can't spend your way past the number of people actually searching.
TikTok buys attention. It can be cheap reach, especially with younger audiences, but it eats creative and the attribution is messy. Until it proves itself, treat it as a testing channel.
At Money Clothing the mix was PPC, paid social and affiliate, and each channel had a different role. Affiliate is worth a mention here because it often gets left out of the budget conversation entirely, even though it competes for the same customers and the same margin.
A sensible starting split
If you're spending somewhere between £10k and £50k a month and you don't have strong historical data to go on, this is a reasonable place to start:
- Meta: 50-60%
- Google: 25-35%
- TikTok: 10-15%
Treat that as a starting point, not a target. Then adjust for your situation.
If you sell into a category people actively search for, think trainers, mattresses or skincare for a specific concern, push Google towards the top of its range. If your product is new, visual or an impulse buy, Meta and TikTok deserve more, because nobody is searching for something they haven't heard of yet.
If your customers are mostly under 30, TikTok's share can go up. If you can't produce at least a handful of new short videos every week, TikTok goes to zero for now. Running stale creative on TikTok is just a slow way of burning money.
One more thing. Carve brand search out as its own line. It will look brilliant in reports because those people were already looking for you. Fund it so competitors can't take your name, but don't let its numbers flatter Google's overall performance.
Your budget size decides how many channels you can run
The ad platforms learn from conversions. Spread a small budget across too many channels and none of them gets enough data to optimise properly. Everything stays stuck in learning and looks worse than it should.
My rough rules of thumb:
Under about £5k a month. One main channel, plus brand search. For most DTC brands that main channel is Meta. Get one thing working before adding another.
£5k to £20k a month. Meta and Google run properly. TikTok gets a small, ringfenced test, if you have the creative to feed it.
£20k a month and up. All three can earn their place, alongside a test budget of around 10% that exists purely to try new audiences, formats and channels.
At Money Clothing, paid media normally ran at £10-20k a month, and ROAS stayed consistently above 4x. At that level of spend, focus matters more than coverage. You're far better off doing two channels well than five channels badly.
Stop judging channels on average ROAS
In-platform ROAS tells you the average return on everything you've spent. What you actually need to know is the return on the next pound. Those are very different numbers.
Brand search might show a huge ROAS, but many of those sales would have happened anyway. Meta prospecting might look modest in its own dashboard while quietly driving the brand searches that Google takes credit for. If you cut Meta because its ROAS looks weaker, you can watch Google's numbers fall a few weeks later.
So set one number as the source of truth: blended return across all paid spend. That's total revenue divided by total ad spend, regardless of what each platform claims.
Work out your target from your margins. If your gross margin after product cost, fulfilment and returns is 50%, you break even on the first order at a 2x blended return. Anything above that is contribution. If you have strong repeat purchase, you can afford to sit closer to break-even on acquisition and make your money on the second and third order.
Then ask a simple question each month: if I added another £1k to this channel, what would I get back? You find that out by increasing spend in steps and watching what happens to blended revenue, not by reading the platform's own report.
Peak months need a different plan
Peak periods change the maths. At Money Clothing, spend went into six figures in peak sales months, with ROAS in the 5-6x range during sales periods, and the brand hit record Black Friday numbers four years running. Peak is where a lot of the year gets made, so the plan for it deserves more thought than "spend more in November".
My advice for peak:
Start prospecting earlier, in October rather than the week before. Ad costs climb as everyone piles in, so building audiences and growing your email list before the rush is cheaper than buying attention at the busiest moment.
Shift weight towards capturing existing demand once the sale starts. People search with intent during sales, and warm audiences who've already visited, browsed or signed up are the cheapest conversions you'll get.
Make sure your email flows are ready to do their share. At Money Clothing I built every Klaviyo flow from scratch, because there was no email marketing before. That matters for budget allocation. When retention is handled by owned channels, paid media can concentrate on bringing in new customers instead of paying again to reach people you already have.
A weekly routine for moving money
Budget allocation isn't a decision you make once a quarter. It's a habit. Here's a routine you can start this week.
- Monday: check blended return against target. If you're above target, you have room to scale. If you're below, find out which channel moved before you cut anything.
- Look at what last week's changes did. If you added 20% to Meta, did blended revenue rise in line with it? If not, the extra spend hit diminishing returns.
- Move money in steps of 10-20%. Doubling a budget overnight resets learning and makes it impossible to tell what happened.
- Protect the test budget. Around 10% stays reserved for experiments, even in a bad week. Otherwise you never find the next thing that works.
- Monthly: review creative output. Your ability to produce new creative often caps Meta and TikTok spend more than budget does.
- Quarterly: run a holdout. Pause a channel or campaign type for a couple of weeks, or in one region, and see what happens to total sales. It's the cheapest honest answer you'll get about what's actually working.
FAQ
What percentage of revenue should a DTC brand spend on paid media?
Work it out from your margins rather than copying another brand. Start with your break-even return on the first order, then decide how much you're willing to spend to acquire a customer based on what they're worth over time. Growing brands often run paid at a higher share of revenue than mature ones, because they're investing in a customer base that pays back later.
Should a new DTC brand start on TikTok or Meta?
For most brands, Meta first. It's more predictable, the targeting does a lot of the heavy lifting, and the reporting is easier to trust. Add TikTok once you have a steady supply of native-feeling video and enough budget to give it a fair test.
How often should I change my budget split?
Review weekly, adjust in small steps, and rethink the overall split quarterly or before peak periods. Reacting to a single bad day usually does more harm than good. Give changes at least a week to settle before you judge them.
There's no perfect split for Meta, Google and TikTok, only a sensible starting point and a disciplined way of moving money once the data arrives. Give each channel a clear job, judge them on blended return rather than their own dashboards, and let the numbers tell you where the next pound should go. If you're working through this for your own brand, I'm open to Head of Marketing and Head of Growth roles, with DTC and consumer brands first, and I take on consulting work too. You can get in touch via the contact link on tomosborne.me.