What Horst Schulze Understood About Excellence That Every Hotel Investing in AI Needs to Hear
Drawing on Horst Schulze's service philosophy at Ritz-Carlton and Capella, the piece argues hotels abandon AI too early by measuring the wrong outcomes on the wrong timeline.
Photo by Are Morch, Digital Transformation Coach
The most respected hotelier alive solved the expectations problem decades before AI existed. The hotels getting impatient with AI today are making the exact mistake he built two legendary companies by refusing to make.
A great many hotels are investing in AI right now, and that is a good thing. But a harder question arrives about ninety days after the contract is signed, and it is the question of quietly deciding whether all that investment succeeds or sours. What results should we be seeing by now, and why does this feel slower than we hoped?
That moment, ninety days in, is where most AI investments are abandoned. Not because the technology failed, but because the hotel measured the wrong result on the wrong timeline and lost faith during the exact phase that was always going to look ordinary.
I want to answer that question. But I do not want to answer it with a vendor's timeline or a maturity model. I want to answer it with Horst Schulze, because the man who built Ritz-Carlton and Capella solved the expectations problem long before AI existed, and almost everything hotels are getting wrong about AI results, he got right about excellence.
The Hotelier Who Refused to Chase the Number
For those who do not know the name, Horst Schulze co-founded the Ritz-Carlton Hotel Company in 1983 and led it to become the only hotel company ever to win the Malcolm Baldrige National Quality Award, and to win it twice. He grew up in a small German village with no hotels, told his parents before he was a teenager that he would work in hospitality, and began as a busboy. He built the most respected service culture in the industry from that beginning.
Then he retired, for exactly one day, and founded Capella to prove the philosophy could reach even higher at a smaller scale. Capella Bangkok was named the number one hotel in the world in 2024. Capella has been ranked among the best hotel brands on earth. It did this with roughly a dozen properties, no race for market share, no giant portfolio. Culture, not size.
And underneath all of it sat one conviction that Schulze repeated for forty years. Concentrate on the things that make money, he said, rather than on the money itself. Chase excellence, and profit follows. Chase the number, and you get temporary wins while your competitor quietly overtakes you.
Hold that, because it is the whole answer to the AI expectations question.
The $2,000 Decision That Explains Everything About AI Results
Here is the most famous thing Schulze ever did, and it is the clearest lesson any hotel investing in AI could receive.
At Ritz-Carlton, he gave every single employee, from the newest housekeeper to the general manager, the authority to spend up to $2,000 to solve a guest's problem without asking anyone's permission. People were stunned. It sounded like a license to throw money away.
Schulze was emphatic that it was the opposite. It was, in his words, an economic decision, not a decision to throw away money. His reasoning was simple. A loyal, repeat guest was worth a fortune over a lifetime of return visits and referrals. Measured against that lifetime value, $2,000 to save the relationship was not a cost. It was one of the best investments the company could make.
Look closely at what he did there, because it is the exact thing hotels investing in AI are failing to do. Schulze refused to measure $2,000 by its immediate cost. He measured it by the compounding outcome it protected. The immediate number looked like a loss. The real result, a guest who returned for a lifetime, was worth many times more, and it only revealed itself over time.
Now put AI in place of the $2,000. Hotels are measuring their AI investment by its immediate return, the fast number this quarter, and getting impatient when that number is not dramatic yet. Schulze would recognize the error instantly, because it is the precise mistake his entire career was built to avoid. The real return on AI, like the real return on his $2,000, is not the immediate figure. It is the compounding outcome, and you cannot see a compounding outcome by staring at the first ninety days.
Why AI Is Schulze's Philosophy Wearing New Clothes
There is a deeper connection, and it should shape how every hotel thinks about what it is actually buying.
Schulze's defining move was to empower the human being in front of the guest. He gave the person the authority to act on what they saw, without waiting for permission, because he trusted their judgment and wanted it unleashed rather than constrained. He did not script his people. He equipped them and got out of their way.
That is precisely what AI, done right, is for. The correct role of AI in a hotel is not to replace the judgment of the person in front of the guest. It is to give that person better information, so their judgment is sharper, and their response is faster and more personal. Schulze empowered his people with authority. AI, used well, empowers them with intelligence. The philosophy is identical. Only the tool is new.
This is why the hotels setting the wrong expectations for AI are so often the ones who would have misread Schulze too. They imagine AI as a machine that produces results on its own, the way they imagine service as a system that runs on its own. Schulze's whole legacy is proof that the durable result comes from the empowered human, never from the system beneath them. AI is simply the most powerful tool yet invented for empowering that human. Expecting it to deliver the result by itself is the same error as expecting a rulebook to produce loyalty by itself. Neither can, because that was never where the result came from.
The Convergence Worth Noticing
There is something worth naming here for anyone who works, as I do, with Blue Ocean Strategy.
Schulze never, as far as I know, followed Blue Ocean Strategy. His methods predate the framework. And yet listen to how he described building Ritz-Carlton. He asked what the customer wanted, what his market wanted, and built every process around that rather than around beating competitors on their terms, and as a result he could charge more and fill more rooms. Listen to how he described starting Capella. He saw luxury splitting into affordable and ultra-luxury, identified the ultra-luxury space as an emerging market that few were serving well, and moved into it deliberately.
That is Blue Ocean reasoning in everything but name. Stop fighting competitors in the crowded existing space. Find or create an uncontested space defined by what customers actually want and build there.
Schulze did not arrive at this from a research framework. He arrived at it from four decades of experience. Kim and Mauborgne arrived at the same conclusion from studying hundreds of companies. When a master practitioner and a body of research reach the identical truth by completely different roads, that convergence is the strongest evidence there is that the truth is real. And it is the same truth I keep returning to independent hotels. The winning move is rarely to fight the giants on their terms. It is to build in the space your specific character and your specific guests define, where scale stops being the deciding advantage.
The Process Any Hotel Can Follow to Set the Right Expectations
This is not reserved for luxury or for large operators. It is a process any hotel can follow, because it is built on outcomes rather than tools. Here is how to set AI expectations the way Schulze set every expectation.
Name the outcome, not the output.
Decide what result matters before you measure anything. Not the output, the count of inquiries answered, but the outcome, the guest who felt genuinely responded to and chose you directly next time. Schulze measured guest satisfaction and employee satisfaction as leading indicators, not just the bottom line, because he knew those were what produced the money later. Define your outcome first. You cannot set a fair expectation for a result you have not named.
Expect relief before you expect revenue.
In the first ninety days, expect operational relief, not financial transformation. The overnight inquiries handled before morning. The cleared front desk queue. The reclaimed hours. The errors that stop recurring. This is the equivalent of Schulze removing the friction that kept his people from being fully present with guests. It is not the final result. It is the condition that makes the final result possible, and it comes first for a reason.
Let the compounding result arrive on its own timeline.
Between three and six months, expect the deeper returns to begin. The conversion gains, the revenue effects, the guests returning because the experience was quiet, reliably excellent. This is the loyalty layer, and loyalty compounds rather than switching on. A hotel that abandons AI at month three is Ritz-Carlton cancelling the $2,000 policy because the first month's spending looked like a loss. The return was always going to arrive later, and it was always going to dwarf the early cost.
Measure like Schulze, not like a spreadsheet.
Judge the investment by the outcome that compounds, not the figure that appears on day one. Concentrate on the things that produce the result, Schulze would say, rather than on the immediate number. The fast metric tells you whether the tool ran. The outcome tells you whether the hotel became better. He built the most respected culture in hospitality by measuring the second thing while his competitors measured the first.
What This Means for Independent, Boutique, and Management-Group Operators
Now return to the fact I asked you to hold. Schulze proved with Capella that culture, not size, is what makes guests come back. Roughly a dozen properties beat portfolios of hundreds to become the best in the world.
That is the entire case for the independent and boutique hotel, made by the most credible figure in luxury hospitality, and AI extends it. What Schulze proved a small hotel could achieve with empowered people, AI now lets a small hotel achieve with empowered people who also carry better intelligence. The boutique property using AI to make every returning guest feel genuinely remembered is doing what Capella does, at its own scale, in its own market. The management company that builds this outcome-first expectation across a portfolio is building Schulze's philosophy into a repeatable system, one standard applied everywhere while each property keeps its own character.
The chains are spending fortunes trying to manufacture at scale the culture and care a well-run independent hotel already has. AI, expected correctly and deployed to empower rather than replace, is how the independent hotel scales the one thing the chains cannot simply buy.
The Expectation Worth Setting
So here is the answer to the question every investing hotel is quietly asking.
Do not expect AI to hand you a fast number. Expect it to do what Schulze did with authority and with his famous $2,000, to empower your people and compound into loyalty over time. Expect relief in the first ninety days, deeper returns across the following months, a lasting capability over the year. Measure it by the outcome that matters, not the output that is easy to count. And do not lose faith during the phase that was always going to look ordinary, because that is exactly where the foundation of the real result is being laid.
Schulze proved that excellence, not the pursuit of the number, is what builds a business that lasts. AI, used the way he used trust and authority, is how a hotel of any size scales that excellence without losing the humanity that makes it worth having.
The hotels getting impatient with AI are chasing the number. The hotels that will win are chasing the outcome, exactly as Schulze taught. And as he promised those who commit to excellence even when it is hard, the success you are hoping for is not merely possible. It becomes inevitable.
The compass is ready. The direction is yours.
Comments
Comments for this content
0 comments available