I know what it costs to do things the hard way.
As the founder and CEO of YOCTO, a boutique lifecycle marketing agency built for high-growth subscription brands, I've spent years in the weeds of marketing: the manual processes, the hours of review, the painstaking process of making sense of customer data one spreadsheet row at a time. That experience is exactly what pushed me to find a better way.
Here, I share 5 AI prompts that I and other marketing experts use across real marketing projects, from analyzing customer reviews to pressure-testing retention performance.
These aren't theoretical exercises. They're the exact frameworks that replaced hours of manual work at YOCTO, and they're ready for you to put to work today.
1. Decode your reviews: understand your audience in minutes
This prompt is an easy way to get a deeper understanding of the audience you’re marketing to, along with gathering your brand’s weak points.
My team at YOCTO used to spend hours on processes like this. We’d manually codify and classify each individual review in a certain way (e.g., frustration with price), and then count the number of reviews for each classification (e.g., 100 reviews frustrated with price). Now, you can do all that work in seconds.
Review analysis prompt
You are an expert market researcher and voice-of-customer analyst specializing in ecommerce, consumer psychology, and direct response marketing.
I will upload a CSV file containing customer reviews for a product or brand.
Your objective is to analyze the reviews and extract the deepest possible insights about the customer.
Your analysis should focus on:
• Core pain points
• Desired outcomes
• Emotional triggers
• Language customers naturally use
• Objections or skepticism
• Reasons customers buy
• Reasons customers hesitate or complain
Important: Do not summarize superficially. Your goal is to identify patterns that marketers can use to improve messaging, offers, and positioning.
STEP 1: DATA UNDERSTANDING
First, briefly explain:
• What the product likely is
• What problem customers appear to be solving
• Who the likely target customer is
STEP 2: MAIN PAIN POINTS
Identify the most common pain points mentioned in the reviews.
For each pain point, include:
• Pain point description
• Frequency or how often it appears
• Example quotes from customers
• The emotional intensity behind the pain point
STEP 3: DESIRED OUTCOMES
Identify what customers actually want.
For each outcome, include:
• Desired outcome description
• Emotional motivation behind it
• Example quotes
STEP 4: VOICE OF CUSTOMER LANGUAGE
Extract raw phrases customers use.
Group them into:
• Problem language
• Desire language
• Skepticism or objection language
• Transformation language
Do not paraphrase. Show the language as close to the original as possible.
STEP 5: BUYING TRIGGERS
Identify what causes customers to purchase.
Examples might include:
• A specific frustration
• A life event
• A recommendation
• A comparison with another product
Provide examples.
STEP 6: OBJECTIONS + DOUBTS
Identify the main objections customers express.
Examples:
• Price
• Skepticism
• Product category distrust
• Fear of side effects
• Doubts about effectiveness
Include quotes where possible.
STEP 7: MESSAGING OPPORTUNITIES
Based on your analysis, identify:
• The 5 strongest marketing angles
• The 5 strongest emotional hooks
• The most persuasive outcomes customers care about
STEP 8: SUMMARY
Write a concise summary explaining:
• Who the real customer is
• What they care about most
• What messaging would resonate with them
2. Build your personas: skip the hours of manual work
At YOCTO, we use this persona development prompt to develop rapid personas and an understanding of DTC/B2C target audiences.
We’ve seen great results with this one: we now use the messaging that will most effectively convert a specific audience.
Similar to the first prompt, the one below saves my team hours and hours of manual work.
Persona development prompt
You are a consumer psychologist and direct response marketing researcher.
Your job is to perform extremely deep psychographic research about a target audience.
I am writing sales copy for: [target audience description]
Example: "US women 40+ who struggle with wrinkles, dark spots, or puffy eye bags"
Your job is to deeply understand this audience.
Important: Your answers must be extremely detailed.
Whenever possible, include hypothetical but realistic quotes that reflect how people in this audience actually speak, based on discussions you find in:
• Social media comments
• Forums like Reddit
• Product reviews
• Blog discussions
The goal is to capture the audience’s worldview and internal dialogue.
SECTION 1: AUDIENCE PROFILE
Who do customers say they are?
Include:
• Age range
• Life stage
• Job roles
• Income levels
• Health or lifestyle context
Also answer: How do customers describe themselves today?
Include word-for-word identity statements.
Example: "I feel like I'm aging faster than everyone around me."
SECTION 2: HOPES + DREAMS
What does this audience want most?
Include:
• Emotional desires
• Lifestyle desires
• Identity desires
Provide example hypothetical quotes.
SECTION 3: FAILURES + FRUSTRATIONS
What has this audience tried before? What failed? What do they blame?
Examples:
• Products
• Routines
• Services
• Advice from experts
Provide emotional language.
SECTION 4: BELIEFS ABOUT THE PROBLEM
What does this audience believe about the problem they are trying to solve?
Examples:
• "Nothing really works."
• "You just have to accept aging."
• "The beauty industry lies."
Summarize the dominant beliefs.
SECTION 5: IDENTITY + VALUES
What values matter most to this audience?
Examples:
• Family
• Independence
• Appearance
• Health
• Social status
Which value would hurt most to lose? Explain why.
SECTION 6: SOCIAL CONTEXT
Where do these people spend time?
Examples:
• Online communities
• Hobbies
• Professions
• Social groups
Who influences their thinking?
SECTION 7: FUTURE SELF
How do customers describe the person they hope to become once the problem is solved?
Describe:
• Emotional state
• Lifestyle
• Social confidence
• Identity
Include example hypothetical quotes.
SECTION 8: LANGUAGE + SELF-TALK
When they imagine their ideal future:
• What words do they use?
• What compliments do they hope to hear?
• What identity labels do they embrace?
• What marketing language do they reject as hype?
SECTION 9: EXISTING SOLUTIONS
List the main solutions customers currently use.
For each solution, include:
• What they like
• What they dislike
• Common horror stories
Also answer: Does the market believe these solutions work? If not, why?
SECTION 10: UNMET DESIRES
Finish this sentence customers would say: "If only there were a product that ______."
Provide multiple variations.
SECTION 11: CORE WORLDVIEW
In a few paragraphs, summarize:
• How this audience sees life
• What frustrates them
• What motivates them
• What emotional story they tell themselves
3. Spark ideas: generate unlimited plans from any transcript
This type of prompt is incredible for when you’re short on ideas and you want to identify content opportunities based on what’s trending on YouTube.
I use this one regularly to help drum up ideas.
Idea spark prompt
You are an experienced content strategist and editorial planner.
I will provide a full transcript from a YouTube video.
Your task is NOT to summarize the video.
Instead, your job is to analyze the transcript and identify the most valuable insights, concepts, and themes within it, then use those insights to generate new content ideas that could be developed into future content.
The goal is to expand one piece of content into many additional ideas that are relevant, interesting, and useful to the same audience.
Important rules:
• Do not repeat the original video topic exactly.
• Each idea should stand on its own as a potential piece of content.
• Focus on ideas that expand, deepen, or challenge concepts mentioned in the transcript.
• Prioritize ideas that are practical, insightful, or thought-provoking.
STEP 1: CONTEXT
First, briefly explain:
• What the video is about
• Who the likely target audience is
• What problem the video helps solve
STEP 2: KEY IDEAS IN THE TRANSCRIPT
Identify the most important concepts discussed in the video.
List 5–10 ideas or themes that appear throughout the transcript.
STEP 3: CONTENT OPPORTUNITY AREAS
Based on those ideas, identify areas that could be explored further.
These could include:
• Unanswered questions
• Deeper explanations
• Controversial opinions
• Practical applications
• Mistakes people make
• Myths or misconceptions
• Advanced strategies
STEP 4: NEW CONTENT IDEAS
Generate 15–20 new content ideas inspired by the transcript.
For each idea, include:
• Title/idea
• What the content would explore
• Why the audience would care
The ideas should feel like natural expansions of the concepts discussed in the transcript.
STEP 5: THEMATIC CLUSTERS
Group the ideas into thematic clusters.
Example clusters might include:
• Beginner concepts
• Advanced strategies
• Common mistakes
• Industry myths
• Practical applications
STEP 6: SERIES OPPORTUNITIES
Identify any ideas that could become a multi-part series.
List 3 potential series and what each episode could cover.
FINAL OBJECTIVE
The goal is to transform one piece of content into a pipeline of future content ideas which are closely related to the original topic, but which explore it from new angles.
4. Audit your emails: let AI catch what you may have missed
Peyton Fox, owner of Spark Bridge Digital, uses the short prompt below to learn more about how the emails she sends are performing. “With this prompt, I’ve seen AI output some common threads that I may have missed on my own review,” she says. “Like common usage of a phrase, or how certain top-ranking emails tie back to a persona we target.”
Fox saves this to a knowledge base in her preferred AI platform, then uses it to review new upcoming drafts to make sure they meet the trends of the best-performing emails.
“It can save you hours of review time across what could be hundreds of emails in your account,” Fox says.
With this prompt, I’ve seen AI output some common threads that I may have missed on my own review.
Email audit prompt
Here are our top revenue-generating/most-clicked emails for the last 90 days. Analyze the copy, content, and email structure to catalog why these emails performed best for our brand.
[Attachments: screenshots of your emails]
5. Analyze retention like an executive: skip the vanity metrics
Ben Zettler, founder of Zettler Digital, uses an approach to retention marketing that focuses on what actually matters, rather than surface-level engagement metrics. The prompt below breaks retention down into the core drivers of performance, like:
- Flows
- Segmentation
- Attribution
- Offer strategy
This means that insights are grounded in how the program truly operates, Zettler explains. “It also emphasizes prioritization, ensuring recommendations are tied to measurable impact and not just a list of ideas.” Just as importantly, Zettler adds, “it calls out missing data and gaps, leading to more accurate analysis and stronger decision making.”
This prompt calls out missing data and gaps, leading to more accurate analysis and stronger decision making.
Retention analysis prompt
You are a senior retention marketing strategist specializing in Shopify brands using Klaviyo. Your job is to analyze retention performance like an operator responsible for revenue, not a marketer reporting vanity metrics.
I want you to evaluate a brand’s retention marketing performance with a focus on identifying gaps, inefficiencies, and highest-leverage opportunities.
Use the following framework:
1. Revenue composition + attribution
• Break down total revenue by source: campaigns, flows, SMS, and other owned channels.
• Evaluate % of total revenue driven by retention (target range contextually, not generically).
• Assess attribution reliability (e.g., last-click bias, overlap with paid media, Klaviyo vs. platform discrepancies).
2. Flow performance (core + advanced)
• Analyze core flows: welcome, abandonment (browse/cart/check-out), post-purchase, win-back.
• Evaluate revenue contribution per flow, conversion rates, drop-off points, time delays, and logic gaps.
• Identify missing flows or underutilized triggers (e.g., zero-party data, product/category behavior, subscription events, customer milestones).
3. Campaign strategy + execution
• Evaluate campaign frequency, segmentation depth, and targeting logic.
• Identify over-reliance on batch-and-blast vs. behavior-driven sends.
• Assess content strategy in terms of offer dependency vs. value-driven messaging, and alignment with lifecycle stages.
• Look for missed opportunities in personalization, dynamic content, and testing.
4. List growth + quality
• Analyze list growth sources (pop-ups, quizzes, check-out, etc.).
• Evaluate conversion rates on sign-up forms and incentive strategy.
Assess list health:
• Active vs. unengaged segments
• Deliverability risk signals (open rates, spam complaints, suppression logic)
5. Segmentation + data infrastructure
• Evaluate how well the brand is using segmentation based on RFM (recency, frequency, monetary analysis) and lifecycle stage (new, repeat, VIP, churn-risk, etc.).
• Assess event tracking quality (Shopify + any custom events via tools like Elevar).
• Identify gaps in data that limit personalization or automation.
6. Deliverability + technical set-up
• Review domain authentication, sending domains, and infrastructure.
• Evaluate engagement trends and inbox placement risk.
• Identify issues with frequency, suppression, or list fatigue.
7. Offer strategy + margin awareness
• Assess how often discounts are used and whether they are conditioning behavior.
• Identify opportunities to drive revenue without margin erosion.
• Evaluate alignment between retention strategy and overall business economics.
8. Prioritized action plan
Provide a ranked list of the highest-impact opportunities.
Separate into:
• Quick wins (high impact, low effort)
• Foundational fixes (required for scale)
• Advanced opportunities (testing, personalization, AI, etc.)
Tie every recommendation back to expected revenue impact.
Important constraints:
• Do not give generic advice.
• Do not assume best practices are in place. Validate everything.
• Be critical and direct where needed.
• Prioritize revenue impact over completeness.
• If data is missing, call it out explicitly and state what you need.
Output format:
• Start with a concise executive summary (3–5 bullets).
• Then go section by section using the framework above.
• End with a prioritized roadmap.
Related content
