TL;DR: When a so-called “AI improvement” adds steps, time, and friction while ignoring frontline and customer input, it is not innovation. It is waste, just wearing a modern label.
I recently stayed at a hotel in the Marriott family, one that positions itself toward the higher end of service and luxury.
In the past, getting my valet-parked car was simple and smooth:
- Pick up the phone
- Press the “Valet” button
- Talk to a valet employee
- Walk down and get the car
It worked. It felt personal. And it matched the brand promise.
That process has now been “improved.”
A New Step… or Several New Steps
At the direction of brand and corporate management, the hotel added a cheerful-sounding AI phone agent to the valet process.
I do not care for it.
To be clear, this is not anti-AI. It is anti-waste.
Here is the new process:
- Pick up the phone
- Press the “Valet” button
- Listen to an AI sound effect and short music clip
- Tell the AI agent you want the valet
- Confirm, again, that yes, you really do want your vehicle
- Wait while the call is “extended” (transferred)
- Wait for the call to be picked up
- Talk to a valet employee
- Walk down to get the car
That is more steps, more waiting, and more friction for the guest, and presumably more cost and complexity for the hotel.
The “Valet” button has effectively been downgraded to the same thing as dialing “0” for the main switchboard. Whatever convenience once existed is gone.
These kinds of “upgrades” do not exactly signal high-end, personal-touch service, the very thing this brand normally aspires to deliver.
Listening to the People Closest to the Customer
I mentioned this to a front desk clerk and said, with a smile, “I would like to offer some friendly feedback.”
She told me the system was new. Many guests had already complained. She also said she would pass along my comments, because more data points help make the case that this change was not really an improvement.
That response mattered.
The people closest to the customer already know this does not work well. They are hearing it directly and repeatedly from guests. And they probably could have predicted the reaction before the system ever went live.
Apparently, someone far removed from the gemba thought this was a good idea.
Was This Ever Tested as a Small Experiment?
I am left wondering:
- Was this rolled out everywhere at once?
- Was there a small test of change at one or two locations?
- Were hotel employees asked for input in advance?
- Were real customers observed using, and being mildly annoyed by, the new process?
Doing any of that would likely have surfaced the problem quickly, before turning it into a standardized “improvement.”
This reminds me of a post I wrote years ago about American Airlines rolling out digital signage that looked modern but made the customer experience worse. It looked innovative, but it added confusion and frustration.
That is not what PDSA (Plan-Do-Study-Adjust) is meant to be. PDSA helps us learn and improve existing processes. It is not a rationale for skipping careful planning or customer testing because “we can always fix it later.”
When changes are put in front of customers without enough upfront learning, the “Do” happens before we have truly Planned or Studied.
Key Takeaway
Adding technology is not the same thing as adding value.
If a change adds steps, delays service, or weakens the customer experience, it is worth asking whether it is really an improvement at all, no matter how modern or “innovative” it sounds.
One Toyota Way principle is clear: adopt technology only when it supports people and an already effective process. When technology adds steps instead of removing them, it is a sign that we automated before we understood.
A Lesson for Lean Leaders
Lean is not about automation, AI, or efficiency in isolation. It is about designing better systems from the customer's point of view and respecting the knowledge of frontline employees.
Before rolling out changes like this, leaders should pause and ask:
- What problem are we actually trying to solve?
- How does this make the experience better for the customer?
- What do the people doing the work, and serving the customer, think?
Small tests, real feedback, and a bit of humility go a long way. Skipping those steps is a reliable way to add waste while calling it “innovation.”
I would be curious to hear from others: where have you seen “improvements” that looked good on paper but failed at the gemba?
JOIN the LinkedIn discussion about this post.
If you’re working to build a culture where people feel safe to speak up, solve problems, and improve every day, I’d be glad to help. Let’s talk about how to strengthen Psychological Safety and Continuous Improvement in your organization.
