Landing Page Performance

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.

Review the Google Ads landing pages of [client: if not listed, call roasr_list_clients and let me pick]: which pages drag results down, which carry them, and what to check on the weak ones.

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. Window: [window: last 30 days (default) | last 14 days | last 90 days | custom]. Minimum clicks for a page to get a verdict: [min clicks: 100 (default) | 50 | 300 | custom]. Use only Google accounts. If the window starts before dataSince, say so and narrow it. If an account has sync.state failed or never, name it and draw no conclusions from it. If currency.mixed is true, never sum money across accounts: do the work per account with accountId.
3. **Pull the pages.** roasr_signals with kind google_landing_pages and limit 200, per account. The result is the latest snapshot per account inside the window, capped by the limit, and payloads of different accounts are not aggregated: read each account on its own and say how many rows came back. Use only the fields that are present; if a field such as mobile speed or mobile friendliness is absent, say it is not in the data and do not guess it.
4. **Get the account baseline.** roasr_metrics_campaigns (campaign level, limit 50) for the same window: the account's average CPA, ROAS and conversion rate, so that pages are judged against their own account and not against an invented norm.
5. **Score each page.** For every page above the click threshold: spend, clicks, conversions, conversion rate (conversions / clicks), CPA. Problem page: conversion rate under half the account average, or CPA 50% or more above it, with enough clicks to be sure. Mark pages below the click threshold "no verdict". Group URL variants that differ only by tracking parameters when the data shows they are the same page, and say you did.
6. **Separate the page from the traffic.** A weak page can be a traffic problem: if the keywords or campaigns sending clicks are off-intent, the page is not at fault. If the data links the page to campaigns, name them. If the skill points at landing page experience as the weak Quality Score component, you may also check roasr_signals with kind google_keywords_qs.
7. **Rule out tracking.** A page with many clicks and zero conversions while the others convert may be a tracking gap on its thank-you step, not a bad page. Name that as a hypothesis to verify first.
8. **Say what to check.** For each weak page, a short checklist matching the evidence: mobile load speed and layout, message match with the ad and keyword, form length, a visible call to action, thank-you tracking.

**Output:** open with one sentence: how much spend goes to pages that convert under half the average, in numbers. Then a table: page · spend · clicks · conversions · conversion rate · CPA · vs account · mobile experience (if in data) · verdict. Then 3-8 actions grouped High / Medium / Nice-to-have, each with the expected effect and the assumption.

**Output rules:**
- Recommendations only — never apply changes; RoASr tools are read-only.
- Use only fields present in the data; never invent page speed or quality scores.
- A page below the click threshold gets "no verdict", not a guess.
- If the client has no margin set, talk ROAS and CPA, not profit.
- Answer in the language of this prompt.