Skip to main content
    Back to blog
    27 October 2026 6 min read

    Brand and Performance Are One Budget: The Demand vs Capture Split

    A practical way to split ecommerce media budget between creating demand and capturing it, based on how much demand already exists.

    Founders often ask how to split their budget between brand and performance, and that's the wrong question. The useful split is between creating demand and capturing it, and the ratio you need isn't a style preference, it's a function of how much demand already exists for your brand before you spend a dollar this month.

    What demand creation and demand capture actually do

    Capture channels take people who already want something and route them to you. Branded search, retargeting, shopping ads on your own name, most of your Meta conversion campaigns to warm audiences. They're measurable, they close fast, and they feel efficient because the return on ad spend (ROAS) looks great. It looks great because you're mostly collecting demand someone else (or your past self) already created.

    Creation channels build the want in the first place. Prospecting on paid social, connected TV (CTV), content, sponsorships, anything reaching someone who hasn't searched for you and won't convert this session. The payback is slower and the attribution is messier. That messiness is often used as an excuse to defund it, which is the mistake this whole post is about.

    Here's the part people skip: capture only works on a pool. If the pool is small, you can spend brilliantly on capture and still get a small result, because you're not short on efficiency, you're short on demand to be efficient with.

    How to estimate how much demand already exists

    You don't need a brand study to get a rough number. Look at three things you likely already have in Google Analytics and your ad accounts: branded search volume and share of total search clicks, direct traffic as a percentage of sessions, and repeat purchase rate as a percentage of revenue. If all three are low, you have very little existing demand, no matter how good your performance campaigns look in isolation.

    A brand with strong branded search, meaningful direct traffic, and healthy repeat revenue has a real pool to fish from. A brand with almost no branded search, traffic that's almost entirely paid-driven, and a repeat rate near zero has close to nothing. Many early-stage Australian ecommerce brands tend to sit in the second group without realising it, because the paid dashboards still show decent numbers week to week.

    A good test to run before touching the budget split: pull branded search impressions for the last 90 days from Google Ads and Search Console, check what share of sessions in Analytics are direct or branded organic, and look at the percentage of revenue from customers who've bought more than once. If two of those three are weak, treat your demand pool as thin.

    What ratio fits each maturity stage

    A brand with a thin pool spending 80 to 90% on capture is paying premium cost-per-click to fight every competitor for a tiny number of people already in-market. That's not an efficiency problem, it's a pool problem, and no amount of bid strategy tuning fixes it. The money is being spent in the right channel for the wrong stage.

    As a rough guide, not a rule: a brand with little to no existing demand should be closer to 60 to 70% creation and 30 to 40% capture, accepting that near-term ROAS will look worse while the pool grows. A brand with a decent branded search base and repeat rate can shift to roughly 50/50. An established brand with strong direct and repeat numbers can lean 60 to 70% capture, because the pool is large enough that marginal capture spend still finds real demand.

    These bands move depending on category and price point, so treat them as a starting split to test from, not a target to defend. The caveat: moving too much budget into creation too fast, before you've got measurement in place to see the branded search lift, will look like a performance decline for a few months even when it's working.

    A planning method with gates

    Rather than set the split once a quarter and argue about it, a better approach is to use gates tied to the demand signals above, reviewed monthly.

    • Pull branded search share, direct traffic percentage and repeat revenue percentage for the trailing 90 days.
    • Compare against the previous period: is the pool growing, flat, or shrinking.
    • If the pool is growing, hold or slightly increase capture share, since creation spend is converting into searchable demand.
    • If the pool is flat or shrinking, increase creation share even if blended ROAS dips for the next cycle.
    • Set a floor on creation spend (say, 20% as a minimum) so it never gets zeroed out during a tight month, because that's exactly when it gets cut and exactly when it shouldn't be.
    • Re-run the three-signal check again at the next monthly gate.

    This isn't a model that spits out a precise number. It's a discipline that stops the split from drifting purely toward whatever channel reports the best ROAS this week.

    Key takeaways

    The real split is demand creation vs demand capture, not brand vs performance. The labels matter less than what each dollar is doing.

    Measure your existing demand pool with branded search, direct traffic and repeat rate before setting any ratio. Guessing the split without this is guessing twice.

    A thin demand pool with heavy capture spend buys expensive clicks against a small audience. The dashboard can look fine while the underlying growth stalls.

    Review the split monthly against pool growth, not just blended ROAS. Set a creation spend floor so it survives the months it's needed most.

    If you want a second pair of eyes on where your demand pool actually sits and what split it supports, that's what full-funnel media planning is for.