A Trial Run

In November 1908 a printer called Antonio Comito, a Calabrian with no work and no money, was taken out of New York supposedly to Philadelphia, where he’d been promised a print shop of his own and twenty dollars a week. In fact he was brought to a stone house in the woods near the hamlet of Highland, fifty miles up the Hudson. There it was explained to him that what he’d be printing was money. The customer was Giuseppe Morello’s family. Almost everything we know about this story comes from Comito’s own statements: later he told the Secret Service everything, more than a hundred typewritten pages. The historian Mike Dash has reconstructed it in detail.

The story interests me not as a crime but as production. The criminal side was sorted out by the court long ago, and sorted out thoroughly. Look at how they worked.

Proof Prints

First they made a Canadian five-dollar note. The engraver, who lived apart and hardly spoke to anyone, etched five zinc plates, one for each colour. Every sheet went through the press five times. It turned out straight away that the ink ran between the fine lines, and however hard Comito struggled, he never got rid of it completely. They printed three thousand notes, and that took a month. Then they took on American two-dollar bills. Simpler, you’d think: only three colours. But they couldn’t get the right shade of green. They tried, gave up, tried again. Work stopped, and the printer was brought into town to meet Morello himself: the only thing Morello wanted to know was how to fix the problem with the two-dollar bill. In the end the engraver worked out additives for the ink and sent a technician to explain, and in February 1909 printing started again.

See what’s going on here? A trial batch. A mistake. A correction. Another batch. Another correction. This is what clever people call iterating, and what we just called “do it, look at it, fix it”.

Now look at what they did not test. They tested the ink, the paper, the shade. But they didn’t test the main thing: whether the whole business would hold up. The goods were fake, the key worker was a prisoner, and the distributors all over the country were men any one of whom might talk if he got caught. The gang meant to print two hundred thousand dollars’ worth of counterfeits. A year later the Secret Service found Comito, and his statements sent Morello to prison for twenty-five years. The most expensive test turned out to be the last, and it wasn’t they who ran it.

The Prototype

An honest Business starts the same way, only from the other end. Before you sink money into machines, make the cheapest version of what you mean to sell and show it to the people who’ll be paying.

When Don Vincenzo decided the laundry could make money on rush ironing of shirts, he didn’t go out and buy steam presses. He set up a single ironing board in a corner, put his best presser at it and hung up a sign: “Shirt by Tonight”. For a week we watched who came in and what they said. Clerks came in from the offices nearby. They all said the same thing: the collar. Nothing else bothered them, but the collar had to stand up. We retrained on collars in three days. Only two months later, when the queue had become steady, did the Don buy his first press.

A prototype isn’t a scaled-down copy of the future Business. It’s a question asked with a thing. It should be as simple as the question allows, and real enough that people answer honestly. A picture will do if you’re asking “do you like it?”. If you’re asking “will you pay?”, you need something people can buy.

Lap Time

Every lap—make it, show it, hear about it, fix it—teaches you something. Whoever does more laps in the same time learns faster. We did a lap a day: ironed, handed back, listened to the grumbling in the evening, changed things in the morning. A competitor with a new press didn’t do a single lap in six months; he just waited for the press to pay for itself. The press, I think, is still waiting.

So keep your laps short. Change one thing at a time, so you know exactly what worked. And don’t fall in love with the first version: its job is to break in a way that shows you how to do it right.

Whom to Listen To

Feedback comes in three grades. Praise out of politeness is useless, like an auntie’s kiss at a wedding. Abuse from a man who’s never going to buy anything from you anyway is next to useless. Complaints from someone who pays and keeps paying are gold. The clerk who grumbled about his collar and brought in another shirt the next day told us more than ten neighbours who “on the whole quite liked it”.

Go looking for feedback yourself; don’t wait for it to come. Don’t ask “did you like it?”, ask “what was wrong?” and “what else would you pay for?”. And say thank you for unpleasant answers. People who tell you the truth are rarer than people who tell it behind your back.

The Morello gang had feedback too, only it came late and in one form: a police raid. You can choose when to get yours and from whom. Choose early, and cheap.