Ads ROAS Forecast

Build ROAS forecasting and attribution-model assumptions for Meta (Facebook/Instagram), Google Ads, TikTok Ads, YouTube Ads, Amazon Ads, Shopify Ads, and DSP...

Security Scan

The skill is an instruction-only ROAS/attribution forecasting helper whose declared inputs and runtime instructions align with its advertising-planning purpose and it doesn't request extra credentials, binaries, or installs.

Purpose & Capability

Name, description, required inputs (forecast_target, planning_horizon, base_assumptions) and the outlined workflow are consistent with an ad forecasting/attribution planner. There are no unexpected requirements (no cloud creds, no unrelated binaries) that would be disproportionate to the stated purpose.

Instruction Scope

SKILL.md stays within modeling, scenarios, sensitivity analysis, and platform guidance. It does include a line about "escalate with a structured handoff payload" for high-risk issues but does not specify destinations or endpoints — this is ambiguous but not inherently malicious. Recommend clarifying where handoff payloads are sent and what data is included.

Install Mechanism

Instruction-only skill with no install spec and no code files. This minimizes on-disk risk and there are no external downloads or package installs.

Credentials

No environment variables, credentials, or config paths are requested. The skill does not ask for unrelated secrets or system access and its declared inputs are reasonable for forecasting tasks.

Persistence & Privilege

Skill is not always-enabled and does not request persistent presence or modifications to other skills. Autonomous invocation is permitted (the platform default) but there are no additional privileges requested.

Assessment

This skill appears coherent and minimal-risk because it's instruction-only and requests no credentials or installs. Before using it: (1) never paste ad-account credentials, API keys, or raw PII into prompts — provide only aggregated metrics and modeled assumptions; (2) ask the skill author or maintainer to clarify what the "structured handoff payload" contains and where it would be sent if the skill escalates an issue; (3) validate outputs on a small, controlled budget before acting on recommendations; (4) treat forecast results as decision support — confirm assumptions, monitor live performance, and implement explicit stop-loss controls; (5) if you plan to automate actions (e.g., via APIs), ensure the automation layer and endpoints are audited and authorised separately. Overall the skill is internally consistent, but safe usage depends on how you supply inputs and whether you connect it to automation or external endpoints.

Purpose

Core mission:

• forecast scenario modeling, attribution sensitivity, budget recommendation

This skill is specialized for advertising workflows and should output actionable plans rather than generic advice.

When To Trigger

Use this skill when the user asks for:

• ad execution guidance tied to business outcomes
• growth decisions involving revenue, roas, cpa, or budget efficiency
• platform-level actions for: Meta (Facebook/Instagram), Google Ads, TikTok Ads, YouTube Ads, Amazon Ads, Shopify Ads, DSP/programmatic
• this specific capability: forecast scenario modeling, attribution sensitivity, budget recommendation

High-signal keywords:

• ads, advertising, campaign, growth, revenue, profit
• roas, cpa, roi, budget, bidding, traffic, conversion, funnel
• meta, googleads, tiktokads, youtubeads, amazonads, shopifyads, dsp

Input Contract

Required:

• forecast_target: roas, cpa, or revenue
• planning_horizon
• base_assumptions

Optional:

• attribution_window_options
• budget_scenarios
• seasonality_factors
• risk_tolerance

Output Contract

1. Model Inputs
2. Scenario Outputs
3. Sensitivity Analysis
4. Attribution Impact Notes
5. Budget Recommendation

Workflow

1. Normalize baseline metrics and assumptions.
2. Build base, upside, and downside scenarios.
3. Run sensitivity on conversion rate and CPC assumptions.
4. Compare attribution windows and expected deltas.
5. Recommend budget path with confidence bounds.

Decision Rules

• If assumptions are uncertain, widen forecast intervals and reduce aggressiveness.
• If scenario spread is large, recommend phased budget release.
• If attribution window drives major variance, present dual-plan decisions.

Platform Notes

Primary scope:

• Meta (Facebook/Instagram), Google Ads, TikTok Ads, YouTube Ads, Amazon Ads, Shopify Ads, DSP/programmatic

Platform behavior guidance:

• Keep recommendations channel-aware; do not collapse all channels into one generic plan.
• For Meta and TikTok Ads, prioritize creative testing cadence.
• For Google Ads and Amazon Ads, prioritize demand-capture and query/listing intent.
• For DSP/programmatic, prioritize audience control and frequency governance.

Constraints And Guardrails

• Never fabricate metrics or policy outcomes.
• Separate observed facts from assumptions.
• Use measurable language for each proposed action.
• Include at least one rollback or stop-loss condition when spend risk exists.

Failure Handling And Escalation

• If critical inputs are missing, ask for only the minimum required fields.
• If platform constraints conflict, show trade-offs and a safe default.
• If confidence is low, mark it explicitly and provide a validation checklist.
• If high-risk issues appear (policy, billing, tracking breakage), escalate with a structured handoff payload.

Code Examples

Forecast Input

spend: 50000
cpc: 1.2
cvr: 0.035
aov: 68

Scenario Output

base_roas: 2.6
upside_roas: 3.1
downside_roas: 2.1

Examples

Example 1: Budget planning with uncertainty

Input:

• Next month spend doubled
• Baseline CVR unstable

Output focus:

• base/upside/downside scenarios
• sensitivity drivers
• safe budget release plan

Example 2: Attribution sensitivity

Input:

• 1d and 7d attribution produce different ROAS
• Need allocation decision

Output focus:

• attribution delta model
• decision thresholds
• channel-level impact

Example 3: Seasonal forecast

Input:

• Holiday promotion planned
• Historical CPC volatility high

Output focus:

• seasonality adjustment assumptions
• risk-adjusted forecast range
• final recommendation

Quality Checklist

• Required sections are complete and non-empty
• Trigger keywords include at least 3 registry terms
• Input and output contracts are operationally testable
• Workflow and decision rules are capability-specific
• Platform references are explicit and concrete
• At least 3 practical examples are included