Beyond the Perfect Prompt: How to Work Better With AI in Your Tourism Business

You ask AI to write a social media post for your tourism business.

It gives you beautiful views, unforgettable adventures and something about making memories.

The words sound polished. They could also describe almost any business in your destination.

What is missing?

In this Firecircle workshop, Nacho Monzon explores how clearer context and specific feedback can help you get more useful results from AI.

You can begin with a simple request, review what comes back and guide the next version. Each exchange is an opportunity to explain what matters.

Useful context makes the difference

During the workshop, Nacho demonstrated this process using a fictional kayaking business.

The starting request was simple: write a short social post for a kayaking tour.

The result was broad and interchangeable. So he added the details that would shape the message:

• The business was in Tofino.

• It wanted weekday bookings in September.

• It was speaking to couples visiting from Vancouver.

• Those guests preferred a quiet pace over adrenaline.

• The voice should feel warm and grounded.

• The post should invite people to ask about weekday departures.

• It should not invent prices, availability, wildlife sightings or guarantees.

Then he gave feedback on the draft, asking for an opening more specific to couples and less hype.

With each exchange, the instructions became more connected to the business and the people it wanted to reach.

Don’t prompt it. Brief it.

Think about how you would brief someone helping with your business.

You would explain what you need, share relevant information and make important limits clear. Nacho organized that brief around four elements.

1. Outcome

What are you trying to accomplish?

A social media post is a format. Encouraging enquiries for weekday departures is a purpose. Explain what the work needs to help you achieve.

2. Context

What does AI need to know about your business, audience, season and situation?

Include the facts that would influence your own decisions about the task.

3. What matters

What would make the result useful?

Consider the tone, level of detail, intended reader and action you want them to take.

4. Constraints

What must the result avoid?

That might include invented facts, unsupported claims, unsuitable language or offers you cannot deliver.

You do not need to anticipate every detail before starting. You can notice what is missing and add it as the work develops.

Make the first answer a starting point

The workshop introduced a simple collaboration loop:

Ask. Review. Give feedback. Refine.

“Make it better” leaves a lot open to interpretation. A more useful response might be:

“Keep the relaxed tone. Make the opening more specific to couples planning a weekday visit. Remove the claim about wildlife sightings because we cannot guarantee that.”

That feedback identifies what to keep, what to change and why.

Your role is to read critically, correct assumptions and decide whether the next version is useful. Check the facts and make sure the result reflects what your business can actually deliver.

Sometimes the next useful step is a question

A participant shared an accommodation-pricing example that began by asking AI to identify the information needed before making recommendations.

That led to questions about the property, current rates, historical performance, comparable accommodation, seasonality, business objectives and measures of success.

The exchange illustrated how AI can help organize the questions behind a task. The operator still needed to supply information, evaluate the analysis and make the business decisions.

For a complicated task, identifying what you need to know can be a useful first step.

Choose the right level for the job

Nacho also introduced his seven levels of working with AI:

1. Invisible: AI operates within tools you already use.

2. Ask: AI answers your questions.

3. Create: AI produces something for you.

4. Collaborate: You work through the task together.

5. Contextualize: You provide relevant background about your work.

6. Systematize: AI supports defined processes.

7. Delegate: AI carries out tasks on your behalf within agreed permissions.

This is a model for choosing how to work. You do not need to reach the most advanced level to get value.

The workshop focused on collaboration and context. A straightforward question may need a straightforward answer. A more involved business task may benefit from a richer brief and several rounds of feedback.

Try it with one real task

Choose something manageable from your to-do list: a guest email, an experience description, a social post or an outline for a staff checklist.

Before you begin, ask yourself:

• What am I trying to accomplish?

• What does AI need to know?

• What matters about the result?

• How will I know if it is good?

Ask for a first draft. Read it critically and make at least one specific revision.

What improved when you added context or gave clearer feedback?

That is a useful practice to build: communicating what matters, evaluating what comes back and staying involved in the work.

Watch the workshop

Watch Nacho’s demonstration and pause the recording to try the exercise with a task from your own business.

Watch Beyond the Perfect Prompt on YouTube

Start with a clear intention, share useful context and give honest feedback. You can build the next version from there.