There may be no greater testament to marital tolerance than this: Hennig Brand’s wife once allowed her husband to store hundreds of barrels of urine in their basement.
But I’m getting ahead of myself.
Brand lived in the 1600s during the transition between alchemy and chemistry. He was obsessed with a single idea: turning base metals into gold.
He wasn’t formally trained in any way. After fighting in the Thirty Years’ War, he reinvented himself as a physician despite having little education and only a vague understanding of human anatomy.
Still, Brand had something that matters in every era: curiosity, confidence, and persistence. This guy was the living embodiment of FAFO: F*** Around and Find Out.
Eventually he became convinced that the secret to gold might lie in the human body. His reasoning was simple:
Gold, after all, is yellow.
And urine is yellow.
Maybe the transformation begins here?
So he started collecting pee.
A lot of it.
Quite the medieval salesman, Hennig sold the military on the idea of gathering barrels of the stuff from its regiments and hauling it home to his basement laboratory.
There he evaporated it, let the residue rot, boiled it into paste, aged it for months, then reheated the mixture with sand and water.
Out of this fragrant process came something unexpected: a waxy substance that glowed in the dark and, unfortunately, occasionally burst into flames.
Hennig Brand—the hack, the quack, the alchemist with the remarkably tolerant wife—had discovered phosphorus, the first new element identified in more than 150 years.
This guy was wrong about almost everything.
But he was systematic about being wrong.
A few decades later the Royal Society of London formalized what we now call the scientific method: Observe, hypothesize, test, analyze, adjust.
Most of us learned it in fifth grade.
Few of us use it when adopting new technology.
The “Mess Around” Trap
Right now, AI tools are being deployed across companies everywhere.
Someone hears an AI testimonial.
Licenses are purchased.
And leadership says something like:
“Let’s give it to the sales team and see what happens!”
I know this happens, because this is what I’ve done.
Multiple times over the last three years.
After watching a demonstration of Claude at the LBM Sales Analytics Conference in San Antonio in September, I went home on fire and signed up.
My plan?
FAFO.
Without a hypothesis, experimentation becomes random exploration. Everyone uses the tool differently, results vary wildly, and leadership is left wondering whether anything actually improved.
AI adoption without structure produces more noise, less signal.
The goal of any technology experiment should be answering one question:
Did this actually help us?
Start With the Simplest Experiment: Email
Most AI initiatives begin with the tool.
“We bought ChatGPT.”
“We signed up for Claude.”
“We’re testing the Plaude AI hardware.”
That’s backwards.
The better starting point is a human frustration.
And one of the biggest is email.
If you are managing email the same way you did two years ago—let alone five or ten—you’re doing it wrong. You’re wasting barrels of time.
Most of us think best out loud.
The mind-to-finger-tips communication ain’t great.
The result is predictable:
• Large blocks of time behind a screen
• Rambling messages
• Unclear follow-ups
• Delayed responses
• Emails that never get written at all
AI finally gives us a simple fix.
The hypothesis
If salespeople can speak their thoughts and use AI to clean up the writing, communication will become faster, clearer, and more consistent.
The experiment (30 days)
For one month:
• Dictate the first draft of a minimum of 3 emails (or meeting summaries!) per day using voice.
• Use AI only as an editor, not the author.
Ask it to:
• Clarify your intent
• Tighten the messaging
• Tweak for tone
• Clean up grammar
What success looks like
• Faster replies to customers
• Clearer internal communication
• Fewer “can you clarify?” responses
• Fewer total emails
If you’re looking for the lowest-risk, fastest-payoff AI experiment available, start here.
AI Is a Lab, Not a Toy
From the outside, Hennig Brand’s basement probably looked like chaos.
Barrels of piss.
Rotting liquid.
Glowing, combustible chemicals.
But inside that mess was something important: Structured curiosity.
He observed something interesting.
He formed a hypothesis.
He tested it.
He adjusted.
Sales leaders should approach AI the same way.
Not as magic.
Not as hype.
But as a laboratory for improvement.
Before adopting the next AI tool, ask three questions:
1. What problem are we trying to solve?
2. What do we believe will happen if we try this?
3. How will we know if it worked?
Then run the experiment.
Because if we apply a little discipline to our curiosity, we won’t just adopt AI.
We’ll actually learn how to use it—without filling the basement with barrels of something that might burn the house down.
Thanks for reading.
I’ll see you back here next Thursday.
Subscribe here to get the next edition of The Craft of LBM Sales straight to your inbox—weekly stories and practical advice to master the craft of selling.
Want the straight analysis on QXO’s Kodiak acquisition and what it means for your sales team in 2026?
Jon Vaughn and I break it down in Episode 44 of The Craft of LBM Sales Podcast.
Listen here:
Apple Podcasts | Spotify
Copyright © 2026
Bradley Hartmann & Co.
All rights reserved.
Contact Bradley Hartmann:
bradley@bradleyhartmannandco.com
The Craft of LBM Sales 