GRW-05 · SEC. 10 Marketing & Growth

Decide How to Split a Fixed Budget Across Paid and Organic Channels

Feed channel performance data and a fixed budget in, get a specific dollar allocation out.

FORMAT: goal
DIFFICULTY: advanced
TIME: 25 min
TOOLS: universal
MODELS: any
COPIES: 0 so far

When to use this

You have a fixed budget for next month or quarter, spend history across 2 or more channels (paid search, paid social, content/SEO, affiliate, whatever you run), and need to decide the split — not a vague "invest more in what's working" recommendation, an actual set of numbers that sum to your budget.

The pattern

Pastes as plain text

My total budget, period, constraints, and channel performance rows are filled in below the line at the end of this block. Each channel row is: Channel | Spend | Results (with the unit: leads, signups, customers) | Notes on data reliability.

Your task:
1. Calculate cost-per-result for every channel. If a channel's result count is too low to be statistically meaningful (fewer than ~30 results, or a channel with a known long lag like SEO), flag it as a data-confidence issue rather than treating it as equal-quality data.
2. Rank channels by cost-per-result, cheapest first, noting confidence flags next to any that apply.
3. Propose a specific dollar allocation across all channels that sums exactly to the total budget. Move dollars toward proven-cheap, high-confidence channels and away from expensive or thin-data ones. Do not zero out every channel but one — state your one-line reasoning for keeping any diversification.
4. Call out explicitly: which channel gets the biggest cut, which gets the biggest increase, and the dollar amount of each move.
5. Name the one metric and threshold that would prove this reallocation was right by the end of the period.

Definition of done: a table of Channel -> New Budget -> Dollar Delta From Last Period that sums exactly to the total budget, plus one line of reasoning per channel, plus the one metric to check the decision against.

Total budget and period:
Constraints (minimum spends, commitments, channels that can't be cut; "none" if none):
Channel rows, one per line:

Real example output

Budget: $40,000/month.

| Channel | Spend | Results | Cost/result | Confidence |
|---|---|---|---|---|
| Paid search | $15,000 | 450 signups | $33 | High |
| Paid social | $12,000 | 180 signups | $67 | High |
| Content/SEO | $8,000 | 310 signups | $26 | Low — 6-month lag, this month reflects work from Q1 |
| Affiliate | $5,000 | 40 signups | $125 | Medium — small sample |

Reallocation:

| Channel | New budget | Delta |
|---|---|---|
| Paid search | $19,000 | +$4,000 |
| Content/SEO | $10,000 | +$2,000 |
| Paid social | $8,000 | -$4,000 |
| Affiliate | $3,000 | -$2,000 |

Reasoning: Paid search is the cheapest high-confidence channel, so it gets the biggest increase. Content/SEO's low cost-per-result is real but lagged, so it gets a moderate increase rather than the biggest one, since this month's number reflects work from before the period. Paid social takes the biggest cut — it's the most expensive high-confidence channel, more than double search's cost-per-result. Affiliate keeps a small floor rather than going to zero, since the sample is too thin to confidently call it dead.

Metric to check: blended cost-per-signup across all four channels, target under $40 by end of next month (down from $44 blended this month).

Why it works

The "sums exactly to the budget" requirement forces a real tradeoff instead of the common non-answer of increasing the best channel while leaving everything else untouched. Flagging data confidence separately from raw cost-per-result stops a lagged or thin-sample channel from getting over-rewarded or over-punished on numbers that aren't trustworthy yet.