AI for Marketing Capstone Exercise — AI for Marketing and Content Teams

What you'll learn

Apply skills from across this course in a single realistic professional scenario
Produce a concrete, role-relevant deliverable using AI tools
Self-assess your output against professional quality criteria

This course has covered AI for content creation, prompt strategies, SEO research, social media, brand voice, ROI measurement, paid media, conversion rate optimisation, and customer segmentation. The capstone tests whether you can coordinate all of those skills under real resource constraints: no agency, a ten-day window, and a campaign that has to work across multiple channels.

The difference between a professional marketer using AI and someone producing generic AI output is brief quality. A well-briefed prompt produces a first draft that needs editing. A poorly briefed prompt produces content that needs rewriting. This exercise rewards the former.

Capstone Exercise

Product Feature Launch: Six Campaign Assets in Ten Days

Context

You are the content marketing manager at a B2B software company that sells project management tools to professional services firms. Your product team has shipped a new AI-powered meeting summary feature that automatically generates action items from recorded calls. You have ten days to launch a campaign. Your budget is zero: no agency, no freelancers. You have one junior team member who can help with scheduling and image sourcing. Your target audience is operations managers and practice leads at firms with 20 to 100 staff who are already paying for your product but have not activated the new feature.

Your Task

Using Claude, ChatGPT, or Copilot, produce six campaign assets from well-crafted AI prompts. Asset 1: a campaign brief of 200 to 250 words covering target audience, key message, and three channel tactics. Asset 2: a hero blog post outline with a full introduction paragraph of at least 150 words. Asset 3: four LinkedIn posts at different funnel stages, one per stage (awareness, interest, consideration, decision). Asset 4: an email sequence of three emails (email 1: awareness, email 2: consideration, email 3: decision with a clear call to action). For at least two of the six assets, include the actual prompt you used to generate the first draft. The prompts must be specific enough that another marketer could use them to produce a similar output: no generic 'write me a LinkedIn post about X' prompts.

Your notes (optional)

Deliverable

Six marketing assets (campaign brief, blog outline with full introduction, four LinkedIn posts, three-email sequence), with the actual prompts used for at least two assets included in the submission. The prompts must be specific enough to be reusable by another marketer, demonstrating brief quality rather than generic instruction.

Quick check

The capstone warns that the model defaults to product promotion at every funnel stage unless the funnel logic is specified. Which of the four LinkedIn posts does that default damage most?

Select one answer.

Key takeaways

Campaign assets produced from generic AI prompts require as much rewriting as a blank page: the professional skill is in the brief, which specifies audience, funnel stage, constraint, and desired reader response before any generation happens
A good campaign brief is the single highest-leverage input in AI-assisted marketing: every downstream asset improves when the brief is specific about who the reader is, what problem they have, and what one thing you want them to do
Funnel logic must be applied by the marketer, not inferred by the AI: left to its own defaults, AI will produce promotional content at every stage, which is the wrong register for awareness and consideration audiences
Including your prompts in a deliverable is a quality signal: a prompt you would not want your manager to read is a prompt that was not doing enough work before you hit generate

Product Feature Launch: Six Campaign Assets in Ten Days

Try these prompts