Quality Score Fix List

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.

Build a Quality Score fix list for [client: if not listed, call roasr_list_clients and let me pick]: which keywords are worth fixing first, what exactly is wrong with each, and what to change.

1. **Load the RoASr method.** Call roasr_load_skill('roasr_frame') and roasr_load_skill('google_search') 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. Use [window: last 30 days (default) | last 14 days | last 90 days | custom]. If the window starts before dataSince, say so and narrow it. If currency.mixed is true, never sum money across accounts; work account by account. Name any account whose sync is failed or never and draw no conclusions from it.
3. **Pull keywords.** roasr_signals with kind google_keywords_qs, limit 200: Quality Score and its three components (expected CTR, ad relevance, landing page experience) per keyword. Each account is its own snapshot; do not merge payloads. Quality Score is a diagnostic, not a KPI: do not judge the account by its average.
4. **Select the problem keywords.** Keep keywords with Quality Score [QS threshold: below 6 (default) | below 5 | below 7 | custom]. Rank them by volume that matters: spend if the payload carries it, otherwise clicks, otherwise impressions, and say which one you used. A weak keyword with no volume is not a priority. A weak Quality Score with CPA already on target is low priority too.
5. **Name the cause per keyword.** Read the three components separately. Below-average ad relevance: ad copy and ad group structure (keyword missing from headlines, too many themes in one group). Below-average expected CTR with relevance fine: ad copy and position. Below-average landing page experience: the page.
6. **Check the pages.** roasr_signals with kind google_landing_pages, limit 200: match pages to the weak keywords' campaigns where the payload allows it, and look at the pages' own conversion and traffic numbers. You cannot open the page, so present speed, message match and mobile layout as things to check, not as findings.

**Output:** open with one sentence — how many problem keywords carry what share of spend and which cause dominates (copy, structure or page). Then a table: keyword · ad group · Quality Score · weakest component · volume · cause · fix type. Then the pages to review, with the keywords they serve. Then 3–8 actions grouped High / Medium / Nice-to-have, each with the expected effect (cheaper clicks, better position at the same bid). Show the top 20 keywords by volume and summarise the tail as one row.

**Output rules:**
- Recommendations only — never apply changes; RoASr tools are read-only.
- Quality Score changes slowly after edits: tell me to re-check in two to four weeks, not the next day.
- Do not promise a specific score or CPC drop; name which way it should move and why.
- If the client has no margin set, talk CPA and ROAS, not profit.
- Answer in the language of this prompt.