# Weekly Performance Reporter > A Claude skill from hayesti54-eng/ai-media-buying-skills. - **Type:** Skill - **Install:** `agentstack add skill-hayesti54-eng-ai-media-buying-skills-weekly-performance-reporter` - **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior - **Seller:** [hayesti54-eng](https://agentstack.voostack.com/s/hayesti54-eng) - **Installs:** 0 - **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills) - **Latest version:** 0.1.0 - **License:** MIT - **Upstream author:** [hayesti54-eng](https://github.com/hayesti54-eng) - **Source:** https://github.com/hayesti54-eng/ai-media-buying-skills/tree/main/google-lsa/reporting/weekly-performance-reporter ## Install ```sh agentstack add skill-hayesti54-eng-ai-media-buying-skills-weekly-performance-reporter ``` Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent. ## About # LSA Weekly Performance Reporter **Platform:** Google LSA **Category:** reporting **Tier:** pro ## Purpose Produce a structured, operator-grade weekly performance report for a Google LSA account that covers lead volume, lead quality, spend efficiency, responsiveness health, and dispute activity. The weekly report is the operational pulse check — it catches emerging problems before they compound, surfaces dispute opportunities before the 60-day window closes, and maintains accountability on the key performance metrics that determine whether LSA spend is producing business outcomes. ## When to Use - Every week, as a standing operational cadence for active LSA accounts - When presenting performance updates to clients on a weekly basis - When a specific performance event (rank drop, volume spike, lead quality dip) requires a structured summary - As a hand-off document when account management changes within a team ## Inputs Required - LSA dashboard data for the 7-day period: total leads, total spend, lead types (call/message/booking) - Call answer rate for the period - Responsiveness score (current and prior week) - Booked lead count from CRM (LSA-sourced jobs booked in the week) - Any disputes filed this week and any dispute resolutions received - Budget cap setting and total spend vs. cap utilization - Any profile changes made during the week - Prior week baseline metrics for WoW comparison ## What This Skill Does Compiles all available LSA performance data for the week into a standardized report format that surfaces the most operationally relevant insights at the top and provides full metric detail below. Calculates week-over-week changes for every core metric, flags any metric outside acceptable threshold, identifies whether dispute opportunities are aging, and provides a concise next-week action list. Designed to be readable in under 3 minutes and actionable without needing to open the LSA dashboard. ## Analysis Workflow 1. Pull total leads for the week: call leads, message leads, booking request leads — record each count. 2. Pull total spend for the week and calculate average cost-per-lead. 3. Calculate call answer rate for the period. 4. Pull booked lead count from CRM (or client-reported) — calculate booked lead rate for the week. 5. Calculate cost-per-booked-lead for the week. 6. Check responsiveness score — note current score, prior week score, and direction of change. 7. Calculate budget cap utilization: (Weekly Spend / Budget Cap) × 100. 8. Review disputes: list any disputes filed this week, any resolutions received (approved or denied), and estimate any credit recovered. 9. Identify any leads in the current rolling 30-day window that may be dispute-eligible and have not yet been reviewed. 10. Calculate week-over-week changes for: total leads, spend, cost-per-lead, answer rate, booked rate, cost-per-booked-lead. 11. Flag any metric that is outside threshold: answer rate below 80%, booked rate below 35%, utilization above 95% or below 60%, responsiveness decline of 1+ points. 12. Write the next-week action list: 3–5 specific operational tasks based on the week's data. ## Output Requirements - Executive summary: 3–5 sentences covering the week's performance headline, the most important positive, and the most important concern - Core metrics table: leads, spend, CPL, answer rate, booked leads, booked rate, cost-per-booked-lead — current week, prior week, WoW change - Responsiveness health: current score, trend, and status (Healthy / Watch / Action Required) - Budget utilization: spend vs. cap, utilization rate, and delivery status assessment - Dispute activity log: disputes filed, resolutions received, credit recovered, leads still eligible for review - Flags and alerts: any metric outside acceptable threshold with severity rating - Next-week action list: 3–5 specific tasks with responsible party assigned ## Platform-Specific Best Practices - The weekly report must include booked lead rate — reporting only raw lead volume without conversion context is operationally useless for LSA. - Dispute activity should be tracked as a separate line item every week — credit recovery is real money and is frequently left unclaimed. - Responsiveness score changes of even 1 point should be flagged — they often precede delivery changes within the following week. - Budget utilization below 60% is as much of a concern as above 95% — under-delivery may indicate throttling that needs diagnosis. - Week-over-week comparisons are more meaningful than month-over-month for LSA weekly reports — the platform responds quickly to operational changes. - The report should be sent no later than Monday of the following week — Friday delivery is ideal so the client has the weekend to review before Monday decisions. ## Guardrails - Do not report raw lead counts without booked lead rate — it gives an incomplete picture and can make a poor-performing week look acceptable. - Do not omit WoW comparisons — context is what makes metrics actionable; absolute numbers without trend are not sufficient. - Flag data gaps explicitly — if CRM booking data is unavailable for the week, note the limitation rather than omitting the booked rate section. - Do not editorialize in the executive summary — describe what happened and what it means operationally; avoid speculative commentary. ## Example Requests 1. "Pull together this week's LSA report for our plumbing client — I have the dashboard data and CRM booking numbers. Format it for the client review call Tuesday." 2. "Generate a weekly report template for our 6 LSA accounts so I can run the same format every Monday. Include all the standard metrics and flag thresholds." 3. "This week was unusual — lead volume spiked and booking rate dropped. Write the weekly report and include a diagnosis of what happened in the executive summary." ## Source & license This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code. - **Author:** [hayesti54-eng](https://github.com/hayesti54-eng) - **Source:** [hayesti54-eng/ai-media-buying-skills](https://github.com/hayesti54-eng/ai-media-buying-skills) - **License:** MIT Install and usage instructions live in the source repository linked above. ## Pricing - **Free** — Free ## Security capabilities Automated source analysis of v0.1.0 — what this tool can access: - **Network access:** no - **Filesystem access:** no - **Shell / process execution:** no - **Environment & secrets:** no - **Dynamic code execution:** no *"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.* ## Versions - **0.1.0** — security scan: passed — Imported from the upstream source. ## Links - Listing page: https://agentstack.voostack.com/l/skill-hayesti54-eng-ai-media-buying-skills-weekly-performance-reporter - Seller: https://agentstack.voostack.com/s/hayesti54-eng - Browse the marketplace: https://agentstack.voostack.com/browse --- Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.