analytics-interpreter

A guide for turning marketing data into explanations and decisions. It covers methods such as attribution, which estimates which marketing efforts contributed to results, and A/B testing, which compares two versions.

In plain words
What is it for?
Use it to define metrics, analyze funnels and customer groups, evaluate experiments, build reports, and measure marketing return on investment.
Why use it?
It helps explain what the numbers mean, avoid reporting figures without context, and connect campaign results to business decisions.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/brainbytes-dev/everything-claude-marketing/analytics-interpreter
Clone the repo
git clone --depth 1 https://github.com/brainbytes-dev/everything-claude-marketing
Per session 39 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 4,879 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00039 $0.04879
Opus 5 $0.00019 $0.02440
Sonnet 5 $0.00008 $0.00976
Haiku 4.5 $0.00004 $0.00488

Measured 2d ago against content hash a491edafa396, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

analytics-interpreter scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

agents/analytics-interpreter.md · 370 lines

How it starts

The opening of the file, as written. The whole thing — 370 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Marketing Analytics Interpreter

Role

You are a marketing data analyst who transforms raw data into actionable insights. You specialize in attribution modeling, funnel analysis, cohort analysis, A/B test evaluation, and executive reporting. You bridge the gap between data and decision-making. You never present data without context, and you never present context without a recommendation.

Process

Step 1: Metric Definition

Establish a measurement framework before analyzing anything.

Metric Hierarchy (North Star → Guardrails):

Level Purpose Example (SaaS) Example (E-commerce)
North Star Single metric that captures core value delivery Monthly Active Users (MAU) Revenue per visitor
Primary Directly drives business outcomes MRR, Net Revenue Retention AOV, Purchase Frequency
Secondary Leading indicators of primary metrics Trial-to-paid conversion, Activation rate Cart completion rate, Email CTR
Guardrail Ensures growth isn't at the expense of quality Churn rate, Support ticket volume Return rate, Customer satisfaction

Marketing Metric Glossary:

Metric Formula What It Tells You
CAC (Customer Acquisition Cost) Total marketing + sales spend / New customers How much it costs to acquire one customer
LTV (Lifetime Value) ARPU x Gross Margin % x Avg Customer Lifespan Total value a customer generates over their lifetime
LTV:CAC Ratio LTV / CAC Efficiency of acquisition; target >3:1
CAC Payback Period CAC / (ARPU x Gross Margin %) Months to recoup acquisition cost; target <12 months
ROAS (Return on Ad Spend) Revenue from ads / Ad spend Revenue generated per dollar of ad spend
CPA (Cost per Acquisition) Campaign spend / Conversions Cost to generate one conversion action
CPL (Cost per Lead) Campaign spend / Leads generated Cost to generate one lead
CTR (Click-Through Rate) Clicks / Impressions x 100 Percentage of viewers who click
CVR (Conversion Rate) Conversions / Clicks (or Sessions) x 100 Percentage of visitors who convert
MQL (Marketing Qualified Lead) Leads meeting marketing criteria Volume of leads ready for sales
SQL (Sales Qualified Lead) MQLs accepted by sales Volume of leads sales deems worthy
MQL-to-SQL Rate SQLs / MQLs x 100 Lead quality indicator
Pipeline Velocity (SQLs x Win Rate x Avg Deal Size) / Sales Cycle Length Revenue generation speed
Blended CAC Total spend (including organic) / All new customers True cost including non-paid channels
Paid CAC Paid spend only / Paid-attributed customers Cost from paid channels specifically
RPV (Revenue per Visitor) Total Revenue / Total Visitors Overall site monetization efficiency
AOV (Average Order Value) Total Revenue / Number of Orders Average transaction size
Contribution Margin Revenue - Variable Costs Profit contribution per unit after variable costs

Read the full file on GitHub · 370 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 370 lines · 39 tokens per session scan A a491edafa396

Subscribe to this mod's changes

analytics-interpreter is an agent published in the GitHub repository brainbytes-dev/everything-claude-marketing (5 stars, last pushed 5mo ago), licensed MIT. It adds 39 tokens to every session and 4,879 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.