Entry requires nothing. No technical background, no coding, no prior use of any AI tool. The course does not assume you are technical: if you can use a smartphone and navigate a booking system, you have the technical ability you need.
The finish line is a person who gets a professional-grade result out of these tools with a few minutes of skilled effort, on more or less anything they attempt. Sixteen video chapters across four phases. Every chapter produces one asset you keep, built from your own business. The sixteen assets assemble into a single Data Storytelling Playbook. No essays. No exams.
You arrive knowing nothing and leave able to open the right tool for a job and write a prompt that comes back with your property in it, not a generic list that would fit any hotel anywhere. This is the phase that closes the gap between owning the tool and getting value from it.
Open a free ChatGPT or Claude account, paste the prompt below, and under it paste the text of one real two-star review of your own business. Nothing to install. This is the chapter 3 strong prompt, unchanged.
1. Read it as the guest who wrote the review. Does it answer what that person actually complained about, or does it answer a generic complaint that happens to be nearby?
2. Strike every sentence that could sit under any bad review at any property. How much text is left, and is the remainder the part that matters?
3. It promises the guest something. Did you ever tell it what you are able to offer, or did it decide for you?
4. Would you publish this under your own name today? If not, name the one instruction you would add to the prompt, add it, and run it again.
This page shows no sample AI output anywhere, on purpose. The answer that teaches you anything is the one your own business produces, and a printed answer would turn the exercise back into reading.
After this phase you know what data your business already holds, where each piece lives, and what it means in your sector's language. You can put that data in front of an AI tool and get an answer back with your numbers in it rather than an industry average.
Collect recent reviews from your own listing, name your city where the prompt asks for it, and paste the reviews where it tells you to. This is the chapter 5 sentiment prompt, unchanged. Strip guest names before you paste anything into a public tool.
1. Pick ten reviews at random and label them yourself before you look at its labels. How many did you disagree on, and did the disagreements all lean the same way?
2. Find each quoted example in your own file, word for word. Anything you cannot find, the model wrote.
3. The percentages came out of a language model, not a spreadsheet. Do they sum correctly, and does the count match the number of reviews you actually pasted?
4. Take the single most common complaint. Is it something you could change this month, and what number would tell you six weeks from now whether the change worked?
Analysis nobody acts on is a hobby. After this phase you can carry one finding to an operations team, an owner, and an investor in three shapes, choose the chart that makes the point survive the room, and build a dashboard people open without being asked.
Take any observation you already trust about your business, drop it into the first bracket, and run this. It is the chapter 7 three-audience generator, unchanged. The point is not the drafts. The point is what you notice when you set them beside each other.
1. Set the three side by side. Wherever two of them say the same sentence, the adaptation did not happen and you are reading one draft in three fonts.
2. The investor version quantifies a return. Which assumption is that number resting on, and did you supply that assumption or did the model invent it?
3. The executive version claims a strategic connection. Is that your actual strategy, or a plausible strategy for a business like yours?
4. One of the three would change what a specific person does on Monday. Which one, which person, and what is the first thing they would do?
The last phase points everything you have built at pricing, pitching, competitive position, and marketing, then at the part most operators skip: getting a team to keep working this way after you stop pushing. It closes with the capstone, where the sixteen deliverables become one playbook.
Fill the brackets with your property, your market, and the five to seven places you actually lose bookings to. If you do not hold an STR report, paste the numbers you do have: your own occupancy, rate, and whatever you know about theirs. This is the chapter 13 competitive analysis prompt, unchanged.
1. It named a competitive set because you did. Is that list the one your guests actually shop against, or the one you have been using out of habit?
2. Trace the market share verdict back to the numbers you supplied. If the conclusion would flip on one number you were unsure of, you have found the number to go and confirm.
3. It read competitor moves as a signal about the market. What else could those same moves mean, and what would separate the two explanations?
4. It recommended one positioning change. What does that change cost, who has to agree to it, and what evidence would tell you within a quarter that it was the wrong call?
The founding cohort opens with the full Phase I release. Join the list for the launch date and founding pricing. Pricing is announced to the list before anywhere else.
Request accessStart chapter one freeGet the bookBack to Academy