Elias spends most of his Tuesdays in a shed that smells like cedar shavings and very old turpentine. He is a boat restorer, specifically specializing in the mahogany hulls of mid-century runabouts.

Last summer, a younger apprentice brought in a high-speed orbital sander, claiming it could strip a hull in a quarter of the time it took Elias with his hand-scrapers and gentle heat guns. On paper-and certainly on the apprentice’s stopwatch-the orbital sander was the “winning variant.” It was faster, it required less physical exertion, and the surface looked smooth to the naked eye.

But Elias wouldn’t let it touch the wood. He knew that the high-frequency vibration of the sander didn’t just remove the old varnish; it crushed the microscopic cellular structure of the mahogany. When it came time to apply the new stain, the wood would “take” the color unevenly. It would look perfect for three months, but by the second season on the water, the finish would begin to delaminate.

I think about Elias every time a growth marketer shows me a bar chart where “Variant B” has a 3.4% higher conversion rate. In the world of e-commerce, we have become obsessed with the orbital sander. We look at the immediate “click” or the “add-to-cart” as the ultimate arbiter of truth, completely blind to the fact that we might be crushing the cellular structure of our customer’s trust.

The Anatomy of a Digital Panic Attack

A few months ago, I was involved in a project where we tested a “Urgency Suite” on a checkout page. You’ve seen them: the little pulsating red dots, the “3 people are looking at this right now” notifications, and the countdown timers that reset if you refresh the page.

The data was unequivocal. Variant B-the one that looked like a digital panic attack-outperformed the clean version by almost 12%. By the logic of modern optimization, we were idiots if we didn’t roll it out to 100% of the traffic.

So, we did. And for 24 days, the revenue numbers looked like a mountain climber’s dream. But then, something strange happened. The “Repeat Customer Rate” began to sag. It wasn’t a sudden drop; it was a slow, rhythmic retreat.

Our regulars-the people who had been buying from us for years-were suddenly less frequent. When we finally ran a sentiment survey, one comment stuck in my throat: “I used to like coming here because it felt like a real shop. Now it feels like a used car lot where the salesman is trying to close me before I can see the engine.”

The Resonant Frequency of Resentment

This isn’t just a matter of “vibes.” There is an industrial precedent for this kind of optimization-driven failure. I remember talking to Hans D.R., an industrial hygienist who spent the better part of the 90s measuring air quality and noise pollution in manufacturing plants.

Hans once told me about a textile mill in North Carolina that decided to “optimize” its floor layout. By moving the machines closer together and removing the heavy, “unproductive” acoustic baffling between stations, they managed to increase throughput by 16% in the first quarter.

The spreadsheets were singing. But Hans was called in six months later because the turnover rate had spiked by 41%. The “optimized” floor had created a resonant frequency that was essentially invisible but caused low-level, constant nausea and cognitive fatigue in the workers.

They couldn’t point to why they hated being there; they just knew they felt “wrong” the moment they stepped onto the floor. The company had optimized for the movement of fabric and ignored the biology of the people moving it.

In digital spaces, we are the workers on that floor. We are constantly being subjected to resonant frequencies of manipulation-UI elements designed to trigger our lizard-brain anxieties rather than serve our adult needs.

When a store prioritizes a 1% lift in a weekly A/B test over the atmospheric calm of the shopping experience, they are effectively removing the acoustic baffling. They are making the store “noisy” in a way that the metrics can’t immediately see.

The problem with A/B testing is that it is a purely rational tool applied to a deeply irrational, emotional relationship: trust. Trust isn’t built in a single session. It’s built over a dozen unremarkable interactions where nothing went wrong and no one tried to hustle you. It is the absence of friction, not the presence of “engagement.”

The Accidental Trickery

I recently had a moment of profound social awkwardness that reminded me of this. I was walking down a crowded sidewalk and saw someone waving enthusiastically. I smiled and waved back, only to realize a second later that they were waving at a friend standing directly behind me. For a moment, I felt a flash of irritation-not at the person waving, but at the situation. I felt “tricked” by my own assumption.

Many A/B tests function on this exact mechanism of accidental trickery. We “win” the conversion because we tricked the user into a sense of urgency or because we hid the “No” button in a shade of grey that barely passes accessibility standards. We wave at them, they wave back, and then they realize we weren’t actually talking to them; we were just trying to get them to move toward the register.

This is why I’ve started to look at stores selling disposable vapes online as a sort of baseline for what “unhurried” commerce should look like.

When you are dealing with a specific, adult-oriented product-like the MT15000 Turbo or the MO20000 PRO-the customer usually knows exactly what they want. They aren’t looking for a “gamified” experience. They aren’t looking for a “mystery box” or a countdown timer telling them they only have 4 minutes to buy a pack of Off Stamp devices before the “deal” expires.

They are looking for authenticity and reliability. They want to know that when they click “Order,” the product is genuine and the shipping is fast.

The moment you start layering “growth hacks” onto a site like that, you begin to erode the very thing that makes a single-brand specialty store work: the expertise and the focus. If I go to a store that only sells one brand, I am doing so because I trust their curation. If that store then starts treating me like a data point in a psychological experiment, the curation starts to feel like a trap.

Implementing the Elias Method

We often forget that the “B” in A/B testing stands for “Better,” but “Better” is a subjective term. Better for who? Better for the quarterly report? Or better for the person who has to live with the interface?

Hans D.R. used to say that the most dangerous thing in a factory was a metric that didn’t have a counter-metric. If you measure speed, you must also measure heat. If you measure throughput, you must also measure vibration. In e-commerce, if we measure conversion, we must also measure “Digital Fatigue.”

We need to ask: how much did it *cost* us to get that extra 2%? Not in dollars, but in the “trust equity” of the brand.

I’ve started advocating for the “Elias Method.” Before we ship a winning variant, we ask if it “crushes the wood.” Does it make the site feel more “hyper”? Does it require the user to expend more cognitive energy to ignore distractions? If the answer is yes, we don’t ship it-even if the numbers say we should.

There is a certain dignity in a store that doesn’t try to outsmart you. There is a quiet authority in a product page that just gives you the facts, the flavor profiles, and a clear path to checkout. Whether it’s a high-end vape or a restored mahogany boat, the things we value most are the ones that don’t feel like they’re trying to “convert” us. They simply exist, reliable and authentic, waiting for us to decide in our own time.

We have to stop treating our users like laboratory rats and start treating them like guests in our home. Because eventually, the rats stop eating the cheese when they realize there’s a shock attached to the floor. And the guests? They just never come back.

Turning Off the Noise

I still think about that “Urgency Suite” we rolled out. We eventually turned it off. The revenue took a hit for about two weeks, a dip that made the stakeholders sweat.

But then, the repeat customer rate started to climb back up. The air in the “digital room” cleared. People stayed on the pages longer, not because they were confused, but because they were browsing. They were breathing. They were no longer being chased by a pulsating red dot.

Sometimes, the best way to win is to stop trying to force the victory. You just have to keep the hull smooth, the air clean, and the wood intact. Everything else-the clicks, the conversions, the loyalty-follows the silence.

“Everything else follows the silence. Or at least, that’s what Elias would say while he’s scraping the varnish, one slow, deliberate inch at a time.”

– Elias, Boat Restorer