Campaign Budget Allocator

The budget-allocation problem

Deciding how to split a fixed budget across Meta, Google, LinkedIn, TikTok, and Reddit is one of the highest-leverage choices in performance marketing, and one of the least systematic. Most teams anchor on last year’s split or on gut feel about which channel "works," when the data usually argues for a different mix.

A reasonable first principle is to fund efficiency: put more budget where each dollar currently returns the most. This calculator allocates your total in proportion to each channel’s current ROAS, so a channel returning 4× receives more than one returning 2×, and shows the projected revenue and blended ROAS that split implies. It is a transparent starting point you can reason about, not a black box.

Why proportional-to-ROAS is a start, not an answer

The simple allocation assumes ROAS stays constant as you move money, and it does not. Every channel has a diminishing-returns curve: shift a large sum onto your best channel and its ROAS will fall as you push past the cheap, high-intent audience into more expensive impressions. A more sophisticated allocation solves for the point where the marginal return of the next dollar is equal across channels — that is where a fixed budget produces the most total revenue.

The simple split also ignores structural constraints. Channels have minimum viable budgets below which they never exit the learning phase, caps above which they cannot spend efficiently, and roles in the funnel that a single ROAS number cannot capture. Use the proportional split to see the shape of the opportunity, then adjust for these realities before you commit.

Channel roles matter as much as channel ROAS

A ROAS-weighted split treats every channel as interchangeable revenue, but channels play different roles. Google search and retargeting largely harvest existing demand and post high ROAS as a result; prospecting on Meta, TikTok, or Reddit creates the demand those channels later harvest, and looks less efficient on a last-click basis even when it is driving the whole system.

If you optimize purely to reported ROAS, budget drains toward demand-harvesting and starves demand-creation, and a few months later the harvesting channels have less to harvest. The fix is to judge upper-funnel channels on incremental new customers rather than headline ROAS, and to protect a prospecting budget even when its ROAS looks lower than retargeting’s.

A worked example

Suppose you have $30,000 a month and your current ROAS is 4.2× on Google, 3.5× on Meta, 2.8× on TikTok, 2.1× on LinkedIn, and 1.6× on Reddit. Proportional-to-ROAS puts roughly $8,800 on Google, $7,300 on Meta, $5,900 on TikTok, $4,400 on LinkedIn, and $3,400 on Reddit, for a blended ROAS around 3.2× and projected revenue near $96,000.

That is a sensible first pass. Now apply judgment: Reddit ads are often under-priced and under-contested, so a channel showing 1.6× today may have more efficient headroom than a saturated one showing 4×; LinkedIn may deserve more if your ICP is strictly B2B; and if $4,400 is below LinkedIn’s minimum viable budget, you may consolidate it into the channels that can use it well. The calculator gives you the baseline; you supply the strategy.

From a static split to a live optimizer

Any allocation is stale the moment performance shifts. ROAS moves week to week with auction dynamics, creative fatigue, seasonality, and audience saturation, so a budget that was optimal in January is merely adequate by March. The teams that compound reallocate continuously, in small evidence-led steps, rather than resetting once a quarter.

That is the case for automation. A live cross-channel optimizer re-reads performance across every platform and rebalances budget toward what is working — which is what Sprites does under your approval across Meta, Google, LinkedIn, TikTok, and Reddit, turning a one-time calculation like this into a standing process.

Questions, answered

How should I split my ad budget across channels?

A defensible starting point is to weight budget toward the channels returning the most per dollar — allocate in proportion to each channel’s current ROAS. This calculator does exactly that. It is a first pass, not a final answer: it deliberately ignores diminishing returns, minimum viable budgets per channel, and the role each channel plays in the funnel, all of which you layer on next.

Why not just put everything into my highest-ROAS channel?

Because ROAS falls as you scale a channel. The first dollars reach the cheapest, highest-intent audience; each additional dollar reaches a harder, more expensive one, so ROAS decays. Concentrating the whole budget in one channel drives it up that decay curve fast, while starving channels that would have delivered efficient volume. Diversification is how blended efficiency stays high as spend grows.

What is a minimum viable budget per channel?

Each channel needs enough spend to exit the learning phase and generate statistically meaningful data — often a few hundred to a few thousand dollars a week depending on your CPA. Splitting a small budget across five channels can leave every one of them under that threshold, so none optimizes well. With a limited budget, concentrate on two or three channels rather than spreading thin.

How often should I reallocate budget?

Frequently enough to follow performance, rarely enough to let campaigns exit learning and stabilize. Weekly reviews with smaller, incremental shifts usually beat large monthly swings, which reset the platforms’ learning and add volatility. The point is continuous, evidence-led rebalancing rather than a set-and-forget annual plan.

Should prospecting and retargeting share the same budget logic?

No. Retargeting almost always shows a higher ROAS because it harvests demand that prospecting created, so a pure ROAS-weighted split over-funds it and starves the top of the funnel that feeds it. Budget prospecting and retargeting separately: size retargeting to the volume of demand prospecting generates, and judge prospecting on incremental new customers, not headline ROAS.

Does this account for seasonality?

Not directly — it allocates on the ROAS numbers you enter, so it reflects seasonality only insofar as your current ROAS already does. In practice you raise budgets into strong seasons and trim in weak ones, and you re-enter updated ROAS as it moves. A live optimizer does this continuously by re-reading performance and reallocating automatically.

Want these numbers checked against your live accounts?