How We Use AI To Delight Guests, Staff, and Owners

Featuring Roman Pedan, CEO of Kasa

This newsletter is brought to you this week by the Hotel Operator’s Weekly Briefing. Check it out if you’d like long-form analysis of what it takes to empower your teams to delight guests and drive profits.

Today, Roman Pedan, CEO of Kasa, a management company with 85 properties, shares how they have used AI to automate more than 50% of their guest message responses. It seems everyone talks about driving efficiency with AI, but I’ve found few people willing to share the details of what this looks like when you’re operating a hospitality business. Read on to learn how they’ve done this.

How AI Helps Us Delight Guests, Cut Costs, and Drive Profits

Roman Pedan, CEO of Kasa, a hotel management company with 85 properties - and quickly growing. He joined us earlier this week to share the story of Kasa, including how he learned from people like Danny Meyer, who shaped his thinking about how high tech should mean high service:

Today, he shares the three-stage process of how Kasa has used AI to improve service, operational efficiency, and profits.

Stage 1: Categorization

To begin, Kasa worked with OpenAI’s technology to automatically categorize guest questions.

The large language models would categorize every question a guest messaged us about. So if you text us about check-in, it would auto-assign a category that says this is a check-in question.

As you might imagine, this data is very helpful for understanding where operational changes, training, or investment are needed. I see many operators analyzing guest reviews and feedback, but quantifying where your guests are running into issues before the review can help you design a more efficient business.

Stage 2: Urgency Scoring

The second stage involved the AI assigning an urgency score to each message, allowing the Kasa team to identify and prioritize the most urgent things.

AI recognizes the urgency level with extraordinary accuracy and places it at the top of the queue.

This is really interesting to me because traditional forms of communication can only be processed in a first-come, first-served way. If you call the front desk, they can only take the calls in the order they’re received. But digitized communication allows operators to triage and prioritize high-urgency, high-importance issues.

Stage 3: Co-Pilot

Today, AI acts as a 'Co-Pilot' for the Kasa team, categorizing and prioritizing inquiries and suggesting responses to the team. These suggested responses can be used as-is or changed by staff before sending.

We provide all the context of a guest's reservation and the communications with us, as well as context about our property. The large language model recommends an answer to our team, and our team presses send.

The result?

We have seen over 50% of guest questions are answered with CoPilot - and that’s been growing.

You can listen to our whole conversation on this here if you’d like more detail on how they’ve done this and what they’ve learned.

What are you doing to prepare your hospitality business to take advantage of AI?

The guest experience and operational efficiency Kasa is driving through AI today is impressive, but it’s all based on the fact they’ve built a business that collects data, which makes this all possible. What will it take for you to take your business in that direction?

If you want more…

Watch Roman give me a little tour of one of Kasa’s San Francisco properties to get a bit of context into their operations:

Check out this guide to AI that I wrote on my other site, HotelOperations.com:

And that’s a wrap! This email was handmade for you—it took me 54 minutes (and a quad shot!) to write this and many more hours to set up the conversation with Roman 🙂 If you found it helpful, I’d be very grateful if you could forward it to a friend or colleague who you think might also enjoy it.

Click here to forward this email.

(Or, if it’s your first time reading this, you can subscribe here.)

A few final things:

  • If you have any thoughts, ideas, questions, or suggestions for Hospitality Daily, please reply to this email. I’d love to hear from you.

  • If you’d like a private community where you can discuss things like this and ask other hotel operators and investors questions that you can’t just Google or ask ChatGPT, the Hospitality Daily Huddle might be a good fit.

  • If you have a product or service that would benefit the hospitality professionals who read, listen to, and watch Hospitality Daily, let’s work together to help you help more people.

Thank you for reading, and I’ll see you again here tomorrow.

-Josiah

Subscribe to keep reading

This content is free, but you must be subscribed to Hospitality Daily to continue reading.

Already a subscriber?Sign In.Not now