Channel Budget Split Plan

Before you start, ask me about each [input] below — one question at a time, with ready-made answers to pick from; skip whatever you can already tell.

Plan how to split next month's ad budget across channels for [client: if not listed, call roasr_list_clients and let me pick]. Total budget next month: [total budget: amount in the client's currency, or "same as the last 30 days"]. Goal: [goal: maximise profit or ROAS (default) | maximise leads at a target CPA | grow volume while holding ROAS | custom]. Look-back: [look-back: last 30 days (default) | last 60 days | last 90 days].

1. **Load the RoASr method.** Call roasr_load_skill('roasr_frame') and roasr_load_skill('profit_math') and follow them exactly.
2. **Frame the data.** Call roasr_limits, roasr_list_clients and roasr_list_accounts: plan, niche, margin, KPI, account currencies, sync status and dataSince. Group the accounts by platform. If the look-back starts before dataSince, say so and narrow it. If currency.mixed is true, never sum money across accounts: compare platforms by ratios (ROAS, CPA) and by share of spend, and keep absolute money per account. Name any account with a failed or never sync and leave it out of the split.
3. **Facts per platform.** roasr_metrics_summary once per account (use accountId) for the look-back and again for its second half and first half, so you can see the trend. Show spend, conversions, revenue, ROAS, CPA and share of total spend. Then roasr_metrics_breakdown with dimension platform to see the segments the platforms report for the same window.
4. **What is inside each channel.** roasr_metrics_campaigns (campaign level, limit 30) per account: how concentrated the spend is, and whether the best campaigns are still above the goal at their current spend. Classify by configuration, not by name; under a campaign budget, ad set numbers show allocation, not a ranking.
5. **Business fork.** Revenue present: ecommerce, compare ROAS with break-even ROAS = 1 / margin when the client has a margin. No revenue: lead gen, compare CPA with the target CPA from the KPI. No margin and no target: say the split can only be judged on relative efficiency.
6. **Marginal, not average.** Average ROAS overstates what the next unit of spend earns. Do not move more than about 20% of a channel's spend in one step, keep every active channel above a learning floor of roughly 15% of its current spend, and treat a channel with fewer than about 30 conversions in the look-back as "not enough data to rank".
7. **Three scenarios.** Conservative (shift about 5–10%), base (about 10–20%), aggressive (up to 30% and one new test). For each: the share per platform, the money per platform (use the total budget; if I gave none, show shares only), the expected result as a range, and the assumption behind it.

**Output:** one sentence with the recommended split and why. Then a table: platform · look-back spend share · ROAS or CPA · trend · conservative · base · aggressive. Then the assumptions and risks, and 3–6 actions grouped High / Medium / Nice-to-have, each with the expected effect. Phrase every move as "we recommend moving about N% of spend from A to B".

**Output rules:**
- Recommendations only — never apply changes; RoASr tools are read-only. This is a plan, not a change to any account.
- Expected results are ranges and upper bounds, never promises.
- If the client has no margin set, talk ROAS and CPA, not profit.
- Keep money per account when currencies are mixed.
- Answer in the language of this prompt.