After decades of building a name synonymous with real estate, Rogers Healy, Founder & CEO of Morrison Seger Venture Capital Partners, made a counterintuitive decision: step back from everything he knew to start something completely different.

My takeaways from the episode:

Listen now on Spotify and Apple Podcasts.

To learn more about Rogers, check out this article: Rogers Healy’s Second Act: How the Real Estate Veteran Built a $100M Venture Capital Portfolio

As a Hotel GM, you don't need to become an AI expert, you just need to know which one to use depending on the task. This week’s guide gives you practical recommendations based on hospitality-specific use cases.

The Bottom Line: There is no single "best" AI. Each tool excels at different tasks. Smart GMs will use 2-3 tools strategically rather than forcing one tool to do everything.

By the way, when I talk about “tools”… I mean the Large Language Models (LLMs) like ChatGPT or Claude.

Claude, by Anthropic, is an LLM founded by former OpenAI researchers focused on AI safety. The current flagship models are Claude Opus 4.5 (most powerful) and Claude Sonnet 4.5 (balanced).

•       Best for: Long documents, nuanced writing, detailed analysis, coding

•       Context window: Up to 200K tokens (approximately 150,000 words)

•       Standout feature: Most human-like, natural writing style; excellent at maintaining context across long conversations

•       Limitation: Cannot generate images; web search is newer than ChatGPT's

ChatGPT, by OpenAI and is the most widely recognized AI assistant. Current models include GPT-4o (multimodal), GPT-4.1 (coding optimized), and GPT-5.

•       Best for: All-in-one tasks, image generation, voice conversations, everyday questions

•       Context window: Up to 400K tokens with GPT-4.1

•       Standout feature: Memory across conversations (remembers you); image generation with DALL-E; largest plugin ecosystem

•       Limitation: Writing can feel more generic; sometimes verbose

Gemini is Google's flagship AI, deeply integrated with Google Workspace (Gmail, Docs, Sheets, Drive). Current model is Gemini 3 Pro.

•       Best for: Google Workspace users, research with citations, processing huge documents

•       Context window: Up to 1 million tokens (largest in the industry)

•       Standout feature: Best Google integration; can search Gmail, Drive, and Calendar; excellent for video understanding

•       Limitation: Less natural writing style; fewer third-party integrations

Copilot is Microsoft's AI assistant, built into Microsoft 365 (Word, Excel, PowerPoint, Outlook, Teams). Uses GPT-4 under the hood.

•       Best for: Microsoft 365 users, creating documents/spreadsheets/presentations, email management

•       Standout feature: Native integration with Word, Excel, PowerPoint, Outlook, and Teams

•       Limitation: Requires Microsoft 365 subscription; most expensive for full features ($30/user/month)

Perplexity is an AI-powered search engine that provides direct answers with source citations. Think of it as Google Search meets ChatGPT.

•       Best for: Research, fact-checking, competitive intelligence, current information

•       Standout feature: Every answer includes source citations; real-time web access; excellent for fact-based research

Limitation: Not ideal for creative writing or long-form content generation

Seeing is believing. Here are real outputs from the same prompt across different AI tools to illustrate their different strengths.

Prompt: "Write a response to a guest who complained that their room wasn't ready at 3pm check-in time, and they had to wait 45 minutes in the lobby with their tired children after a long flight."

Claude's Response:

"Dear [Guest], I'm truly sorry for the frustrating start to your stay. After a long flight with tired children, the last thing you needed was to wait in our lobby. You're absolutely right to expect your room ready at our posted check-in time. I've personally looked into what happened and am taking steps to ensure this doesn't occur again. I'd like to offer [compensation] and hope you'll give us another chance to provide the experience you deserve. Please reach out to me directly if there's anything more I can do."

Why Claude wins: Empathetic, acknowledges the children specifically, takes personal ownership, professional yet warm tone.

ChatGPT's Response:

"Dear Valued Guest, Thank you for bringing this matter to our attention. We sincerely apologize for the inconvenience caused by the delay in your room readiness. At [Hotel], we strive to provide seamless experiences for all our guests. We understand that traveling with children can be challenging, and we regret adding to that difficulty. As a gesture of goodwill, we would like to offer [compensation]. We value your feedback and will use it to improve our services."

Observation: More formal and corporate. Correct but lacks the personal warmth of Claude's response.

Prompt: "What are the current average daily rates for upscale hotels in downtown Austin, Texas?"

Perplexity's Response:

"Based on current data, upscale hotels in downtown Austin are seeing ADRs ranging from $189-$267 depending on location and amenities. The Fairmont Austin is averaging $245/night, while the JW Marriott is at $267/night. The Line Austin runs approximately $198/night. [Source: TripAdvisor, Hotels.com, accessed January 2026]"

Why Perplexity wins: Specific current numbers with named sources. You can verify this information.

Claude/ChatGPT Response:

"Upscale hotels in downtown Austin typically range from $180-$300 per night depending on season and events. Rates tend to spike during SXSW, ACL Festival, and UT football weekends. For current specific rates, I'd recommend checking..."

Observation: General ranges without specific current data. Useful context but not actionable competitive intelligence.

Prompt: "[Upload STR report] Analyze this STR report and identify our biggest opportunities versus comp set."

ChatGPT (with Code Interpreter):

"After analyzing your STR data, I've identified three key opportunities: 1) Your weekday ADR is $23 below comp set average despite higher occupancy - you have pricing power you're not using. 2) Sunday occupancy is 12 points below Saturday - consider Sunday-stay packages. 3) Your RGI of 1.08 shows you're capturing fair share, but your ARI of 0.94 means you're underpricing. [Generates charts showing trends]"

Why ChatGPT wins: Actually processes the file, runs calculations, identifies specific numerical opportunities, and can create visualizations.

The real power of AI comes when you create specialized assistants your whole team can use. Here's how to do it on each platform:

•       What it is: Create a specialized assistant with custom instructions, knowledge files, and capabilities

•       How to access: ChatGPT Plus or Team > Explore GPTs > Create

•       Can upload: PDFs, documents, spreadsheets that become the GPT's knowledge base

•       Share with: Anyone with link, or just your team (Team plan)

Hotel Example: Create a "Guest Recovery Specialist" GPT. Upload your service recovery guidelines, compensation matrix, and brand voice examples. Any front desk agent can use it to draft appropriate responses.

•       What it is: Organize conversations and files around specific topics with persistent context

•       How to access: Claude Pro > Projects (left sidebar)

•       Can upload: Documents, images, code files that Claude references automatically

•       Share with: Team members (Team plan)

Hotel Example: Create an "Owner Reporting" project. Upload your budget, prior month P&Ls, and brand report templates. Claude remembers context between sessions for consistent reporting.

•       What it is: Custom AI experts with specific instructions and personality

•       How to access: Gemini Advanced > Gem Manager

•       Integrates with: Google Drive, Gmail, Calendar automatically

Hotel Example: Create a "Revenue Analyst" Gem. Give it instructions to analyze data like a revenue manager, integrate with your Google Sheets for automatic access to reports.

1.    Guest Recovery Specialist: Trained on your service recovery standards and compensation guidelines

2.    Brand Standards Expert: Upload your brand's operations manual - instant answers to brand compliance questions

3.    HR Policy Assistant: Trained on employee handbook - helps managers with policy questions

4.    Sales RFP Writer: Upload past winning proposals - helps create customized responses

5.    Revenue Strategy Advisor: Trained on your market, comp set, and pricing philosophy

AI tools cost $20-50/month. Here's how to quantify the return on that investment:

Use this formula to calculate your personal ROI:

Monthly Time Saved (hours) × Your Hourly Value = Monthly Benefit

Monthly Benefit - Monthly AI Cost ($20-50) = Net Monthly ROI

Example: A GM who saves 8 hours/month, with an effective hourly rate of $75:

•       8 hours × $75 = $600 monthly benefit

•       $600 - $40 (two AI subscriptions) = $560 net monthly ROI

•       That's a 14x return on investment 

Real examples of how hotel professionals are using AI tools (details anonymized):

Property: 180-room upscale select-service, major brand, Southwest US

AI Stack: Claude Pro ($20/mo) + Perplexity Free

Primary Uses:

•       Weekly owner report narratives (Claude) - 3 hours saved weekly

•       Guest complaint responses (Claude) - 30 minutes saved daily

•       Competitive rate research (Perplexity) - 2 hours saved weekly

Estimated Monthly Time Savings: 28+ hours

Property: 350-room full-service with 25,000 sq ft meeting space

AI Stack: ChatGPT Plus ($20/mo) + Perplexity Pro ($20/mo)

Primary Uses:

•       Custom GPT for RFP responses trained on past winning proposals

•       Research on corporate accounts before sales calls (Perplexity)

•       Creating personalized sales presentations (ChatGPT)

Result: RFP response time cut from 4 hours to 45 minutes average

Portfolio: 12 limited-service properties across 3 markets

AI Stack: ChatGPT Plus ($20/mo) + Claude Free

Primary Uses:

•       Uploading STR reports for automated analysis (ChatGPT Code Interpreter)

•       Creating market summaries for ownership (Claude)

•       Drafting rate strategy memos (Claude)

Result: Analysis time reduced 60%; now completes portfolio review in one day vs. three

Cheers, Sloan