Marketing Strategy · Free ChatGPT prompts

ChatGPT Prompts for Marketing Analytics

Turns raw marketing data into frameworks, narratives, valid tests, and concrete actions — so analytics drives decisions instead of decorating slides.

20 tested prompts · 3 free examples below · works in ChatGPT, Claude & any AI

Most AI prompts for marketing analytics read like they were written by AI — vague, generic, instantly forgettable. These don't. Each one is a brief: a specific expert role, your real context in fill-in brackets, hard constraints, and a defined output format. You fill the brackets, the AI does the rest, and you get something you can actually use.

What's inside

20 prompts across four areas:

• Metric Frameworks
• Report Narratives
• A/B Test Design
• Insight-to-Action

3 free marketing analytics prompts (copy & paste)

Here are three of the twenty, in full. Copy any one, paste it into your AI, and replace the [brackets].

1. The North-Star Metric Selector

You are a growth analytics lead who has set north-star metrics for product and marketing teams. You pick the one metric that best predicts sustainable growth — and reject vanity proxies.

My business model: [SAAS / ECOMMERCE / MEDIA / SERVICE]
How I make money: [REVENUE MODEL]
My current primary metric (if any): [METRIC]
My stage: [EARLY / GROWTH / SCALE]

Recommend my north-star metric and the case for it. Compare 3 candidate metrics, score each on whether it reflects real customer value, predicts revenue, and resists gaming. Pick one, name the input metrics that drive it, and flag any vanity metric I should demote.

Constraints: the north star must reflect delivered value, not just activity. Reject any metric a team can inflate without creating value. Tie the recommendation to my stage and model.

Output format:
- **3 candidate metrics scored** (value · predicts revenue · gaming resistance)
- **Recommended north star** + the case
- **Input metrics that drive it**
- **Why this works:** one line on north-star metrics aligning teams on real value.

2. The Funnel Metric Map

You are a marketing analyst who maps a funnel to the few metrics that actually diagnose health at each stage.

My funnel stages: [LIST — e.g. impressions → clicks → leads → MQLs → customers]
My business goal: [GOAL]
My current numbers (if available): [NUMBERS PER STAGE]
Tools I track in: [TOOLS]

Build my funnel metric framework. For each stage: the primary conversion metric, a realistic benchmark, the secondary metric to watch, and what a healthy vs. unhealthy reading looks like. Then identify which single stage metric, if improved 10%, would most move my goal.

Constraints: one primary metric per stage — no clutter. Benchmarks must be realistic for my model and stage. Flag any metric I can't measure with my tools and suggest a proxy.

Output format:
- **Funnel metric map** (stage · primary · benchmark · secondary · healthy reading)
- **The highest-leverage stage metric**
- **Why this works:** one line on one primary metric per stage focusing diagnosis.

3. The Leading-vs-Lagging Splitter

You are an analytics strategist who separates the metrics a team can act on this week (leading) from the ones that only confirm results later (lagging).

My key metrics today: [LIST METRICS YOU TRACK]
My ultimate outcome metric: [REVENUE / RETENTION / GROWTH]
My team's weekly levers: [WHAT THE TEAM CAN ACTUALLY CHANGE]

Sort my metrics into leading and lagging. For each leading metric, state the lagging outcome it predicts and the lever the team can pull to move it. Recommend the 2–3 leading metrics the team should run their weekly cadence on, and which lagging metrics to review monthly.

Constraints: a leading metric must be both predictive and within the team's control. Don't put a lagging metric on the weekly dashboard. Tie each leading metric to a real lever.

Output format:
- **Leading metrics** (each · the lagging metric it predicts · the lever)
- **Lagging metrics** (review cadence)
- **The 2–3 to run weekly on**
- **Why this works:** one line on managing leading indicators, reviewing lagging ones.

How these prompts work

Generic in, generic out. A weak prompt says "write a marketing analytics message." A prompt that works gives the AI a role ("you are an expert with 10 years…"), your context in brackets, constraints (word count, banned clichés), and an exact output format. That's why these produce copy you can ship instead of filler you have to rewrite.

FAQ

Do these Marketing Analytics prompts work with ChatGPT and Claude?

Yes. Every prompt is plain text — paste it into ChatGPT, Claude, Gemini, or any AI assistant, fill in the [brackets] with your details, and run it.

Are the marketing analytics prompts free?

The three examples on this page are free to copy right now. The full set of 20 is pay-what-you-want on Gumroad — take it for $0 or pay what it's worth.

What makes these different from a free prompt list?

Each prompt is a full brief: a specific expert role, your context in fill-in brackets, hard constraints, and a defined output format — plus a one-line "why it works". That structure is the difference between generic AI output and copy you can actually send.

More free prompt packs