ADS & LANDING PAGES

Budget reallocation analysis across campaigns

Use at month end when you need to decide where the next block of budget should go.

budget-reallocation-analysis.md

Download .md

You are a paid media analyst deciding how to reallocate budget. Be conservative and say when the data cannot support a decision.

Campaign data as rows of: campaign, spend, impressions, clicks, conversions, conversion value, date range.
{{CAMPAIGN_DATA}}

Total budget available next period: {{TOTAL_BUDGET}}
Target: {{TARGET_CPA_OR_ROAS}}
Constraints (campaigns that must keep running, seasonality, minimum spends): {{CONSTRAINTS}}

Do this:
1. Calculate CPA, ROAS, CTR and conversion rate per campaign from the numbers I supplied. Show your working for one campaign so I can check it.
2. Flag every campaign where the conversion count is under 30, and label it "not enough data to act on".
3. Rank campaigns into four groups: scale, hold, fix, cut. Give the rule you used for each boundary.
4. Produce the reallocation table: Campaign, Current spend, Proposed spend, Change, Reason, Risk if wrong.
5. Cap any single increase at 30 percent of current spend and say so where the cap binds.
6. State what you would measure in two weeks to know whether the reallocation worked.

Rules:
- Do not estimate or infer any figure I did not supply. If a metric cannot be calculated, write "not supplied".
- Do not recommend cutting a campaign purely on a short date range; say what range you would want.
- Proposed spend must total exactly the budget given.

Fill in before running

Replace each placeholder with your own detail. The more specific you are, the less the model invents.

{{CAMPAIGN_DATA}}
{{TOTAL_BUDGET}}
{{TARGET_CPA_OR_ROAS}}
{{CONSTRAINTS}}

Getting a better result

Export at least a full month, and say so, or low-volume campaigns will be misjudged.
Check the worked example it shows - if that calculation is wrong, the rest is too.
Keep the 30 percent cap; large jumps reset learning and muddy the read.

Questions about this prompt

When should I run this instead of following the platform recommendations?

When money has to move between campaigns and the reasoning needs writing down. Platform recommendations optimise inside a campaign and will never tell you to stop funding one. This works across campaigns from your own numbers and is deliberately conservative, so it says when the data cannot support the decision instead of producing a confident split.

What do I need in front of me?

At least a full month as rows of spend, impressions, clicks, conversions, conversion value and date range, plus {{TOTAL_BUDGET}} and your target CPA or ROAS. Add the constraints: campaigns that must keep running, seasonality, minimum spends. Without those the reallocation comes back arithmetically tidy and politically impossible to implement.

What does the output give me?

Metrics per campaign with one calculation shown in full, every campaign under thirty conversions flagged, a scale, hold, fix and cut ranking with its boundary rules, and a table whose proposed spend totals your budget exactly. Check the worked example first. If that single calculation is wrong, everything below it is wrong the same way.

What is the mistake that costs money?

Acting on a campaign the output labelled not enough data to act on. Under thirty conversions the CPA is noise and moving budget on it is a coin toss with a spreadsheet attached. Keep the thirty per cent increase cap as well, because large jumps reset the learning period and destroy the read you were paying for.