# Build Marketing Prompt Library

> Ready-to-copy AI prompts from Build Marketing articles, for positioning, go-to-market, AI agents and your website. Each prompt links to the article that explains it.

## Add streaming to your AI feature

From the article [Why AI Answers Need to Be Streamed](https://www.buildmarketing.ai/why-ai-answers-need-to-be-streamed.md) · For your AI coding tool (Claude Code, Cursor, Copilot)

Give this to your AI coding tool to make your website's AI answers appear word by word, and to find other causes of slow answers.

```text
You are working in my website's codebase. Our AI feature feels slow: visitors type a question, see a spinner, and wait for the whole answer before anything appears. Make the answer stream onto the page word by word as the AI writes it.

Before you change anything:
1. Find the full path of one question: the page or component the visitor types into, the server endpoint or function it calls, and the AI API call (OpenAI, Anthropic, Gemini or another provider).
2. Tell me which step waits for the complete answer today. Check the AI call, the server response, the hosting platform and the browser code.

Then make these changes:
3. Call the AI provider's streaming API (for example stream: true, or the SDK's stream helper) and forward each piece of text as it arrives.
4. Return a streamed response from the server: a ReadableStream with plain text, or Server-Sent Events (text/event-stream). Do not collect the whole answer before sending it.
5. Make sure nothing between the server and the browser buffers the response: check the hosting platform's function settings, any proxy or CDN, compression and middleware. Use a function type that supports streaming responses.
6. In the browser, read the response with fetch() and response.body.getReader(), decode each chunk with TextDecoder, and append it to the answer on screen as it arrives. Keep the page scrolled to the newest text.
7. Show a small "thinking" indicator only until the first words arrive, then remove it.
8. Handle errors part-way through: keep the text already shown and add a short message with a way to try again. Let the visitor stop or cancel an answer.
9. Keep anything that needs the full answer (saving the chat, logging, lead alerts) running after the stream ends, without delaying the first words.

Also check for speed problems streaming will not fix:
10. List what is sent to the AI on every question (system prompt size, history, documents). Turn on prompt caching for the parts that repeat, if the provider supports it.
11. Tell me whether a smaller or faster model, or a lower reasoning or thinking setting, would give good enough answers for this feature.
12. Point out questions that do not need AI at all (fixed answers, lookups, forms) and could be answered instantly in normal code.

Finally, measure before and after: the time until the first words appear, and the time until the answer is complete. Report both numbers to me.
```

## Draft your first AI agent's job description

From the article [How to Build an AI Agent: A Practical Guide for Your Business](https://www.buildmarketing.ai/how-to-build-an-ai-agent.md) · For ChatGPT, Claude or Gemini

Turns a task you want to hand off into the one-page job description from Step 1, ready to brief any agent platform.

```text
You are helping me write a job description for an AI agent before we build it.

The job I want to hand off: [describe the task in a sentence or two]
Who the work is for: [team or customer]
Systems involved: [for example HubSpot, Google Drive, our website]

Interview me one question at a time until you have enough to write it. Then write a one-page job description with these sections:

1. The job: one sentence naming the single job.
2. What it may read: the data and systems it can look at.
3. What it may write: the one thing it is allowed to change or send.
4. Decides alone: decisions it can make without asking.
5. Hands to a person: when it must stop and ask, and who it asks.
6. Never does: hard rules.
7. Its boss: the person who reviews its work.
8. Measure of good: how we will know it is working, with a number where possible.

Keep it in plain English a new hire would understand. Flag anything that sounds risky to give an agent, and suggest a smaller first version if the job is too big.
```

## The CFO prompt: measure your marketing team’s AI maturity

From the article [My CFO Cut My Marketing Budget and Said: Use AI. Now What?](https://www.buildmarketing.ai/my-cfo-cut-my-marketing-budget.md) · For ChatGPT, Claude or Gemini

Paste this into ChatGPT, Claude or Gemini with marketing’s AI spend to see how mature the team really is: seats versus API spend, how many tools are integrated, which workflows AI supports, and how it links to people, leads and sales.

```text
You are an advisor to CFOs on AI in marketing. I am the CFO. I believe AI maturity drives better marketing execution, and I want to measure how mature our marketing team really is by looking at what it spends on AI, how connected its tools are, what that spend is used for, and what it produces.

What I will give you (use what I have; ask me for anything important that is missing):
- Every AI cost in marketing for the last 12 months: seat subscriptions (such as ChatGPT, Claude, Gemini or Copilot), AI features inside software we already pay for, API and usage-based bills for each tool, and agencies or contractors who charge for AI work.
- For each seat subscription: monthly cost, number of seats, and how many people actually use it and how often, if I have it.
- Every marketing tool we use (CRM, marketing automation, website, ads, analytics, data and enrichment), and whether each one is integrated with our AI through an API or an MCP connector.
- Whether the team has shared MCP prompts: saved prompts that use those connectors to pull data and run a workflow across our tools.
- The marketing team: headcount, roles and total marketing budget.
- Results by month or quarter: leads, qualified leads, pipeline, closed sales from marketing, and campaign volume.

Work through these steps:
1. Build a table of our AI spend: tool, type (seat subscription, feature in existing software, API or usage-based, agency), monthly cost, seats, active users, whether it is integrated, and share of the total marketing budget.
2. Answer these maturity questions with evidence from the data:
   a. Are we single threaded on one tool, or using the right tool for each job?
   b. Are we light users? Compare the seats we pay for to the people who actually use them, and how much each person uses them.
   c. Do we have API spend across multiple tools, not just seat subscriptions? API spend across several tools is the clearest sign that AI is built into our workflows and systems, not just used in a chat window.
   d. How many of our marketing tools are integrated with AI through an API or MCP connector? The goal is every tool. List the ones that are not.
   e. Do we have shared MCP prompts that run real workflows across our connected tools, and how many people use them?
3. Map each AI cost to the marketing workflow it supports: research, content creation, campaign production, personalization, lead scoring and routing, reporting and analysis, and customer insight. Flag spend that maps to no workflow, and important workflows with no AI behind them.
4. Rate our maturity on this five level scale and explain why:
   Level 1, Experimenting: a few people using personal or free tools, no budget line.
   Level 2, Individual use: seat subscriptions, mostly one general chat tool, light and uneven use, no API spend.
   Level 3, Team workflows: seats across several tools matched to specific workflows, steady use, API spend starting in one or two tools.
   Level 4, Integrated: API spend across multiple tools, most marketing tools connected through APIs or MCP, and shared MCP prompts running key workflows.
   Level 5, Scaled: every marketing tool integrated, API spend across all of them, MCP prompts running most workflows from start to finish, and results measured against spend.
5. Link spend to results. For each workflow AI supports, compare before and after: the hours or people it takes, leads and qualified leads, cost per lead, pipeline and closed sales. Say clearly where the link is strong evidence and where it is only a guess, and what tracking we would need to prove it.
6. Tell me what moving up one level would take: which tools to integrate next, which MCP prompts to build, what spend to add, what to cut (unused seats, overlapping tools), and what results I should expect to see.
7. Give me five questions to ask our CMO about AI maturity, and three to five metrics to track each quarter, such as share of tools integrated, API spend as a share of AI spend, active seat use, cost per qualified lead, and pipeline per marketing dollar.

Finish with a one page summary I can share with our CEO and CMO: our maturity level, the evidence for it, how our AI spend links to people, leads and sales today, and the next three steps. Where my numbers are missing, ask me instead of guessing.
```

## Find your position in five questions

From the article [Positioning: The Timeless Marketing Strategy](https://www.buildmarketing.ai/positioning-the-timeless-marketing-strategy.md) · For ChatGPT, Claude or Gemini

A starting point that begins with who you are in the customer's mind, then the article's five questions in plain words, ending with position sentences to test and how each could change your valuation.

```text
Help me find a clear position for my company. Positioning starts with who we are in the mind of the customer, not with our product. Keep it simple and use plain words.

About us: [what you sell and who buys it]
Our main competitors: [list a few, or "help me list them"]
What we have to work with: [team size, money, time, key strengths, partners]

First, confirm the list of competitors and alternatives with me. Then ask me these five questions one at a time. Wait for my answer before the next one:

1. Who are we in the mind of the customer today? What word or idea do they link to us, and how does that compare with the alternatives? Where is the market heading: which technology are buyers adopting, how fast, and are we riding that growth or fighting it?

Questions 2 to 5 look at the positions open to us, given the make-up of our company and the resources we have. The goal is the position that maximizes our opportunity, but do not suggest one we cannot afford to win.

2. How should we package what we sell: as an app, a platform others build on, or an API?
3. Which market do we go after first (for example education or finance), who pays there, and how do they buy (per seat, by usage, annual contract, grants or procurement)? Who matters most: the user, the buyer, or a future acquirer?
4. What can we honestly say we are first or only at? And what are we not, or not yet?
5. What will people doubt about that, and how do we answer?

Positioning depends on more than these five questions, so before you write anything, ask me about any other factors that matter for us, such as pricing, partners, regulation, timing or how fast buyers are changing. Then write our position in one sentence: "For [customer] in [market], we are the first (or only) [category] that [benefit]." Add one line on what we are not (yet), so the position stays honest. Give me two versions to test, and for each one, say how it could change how investors or acquirers value us: the packaging, the market and its growth all affect what a company like ours is worth. Do not invent facts about us; ask if you need something.
```