Start with the boring task

Welcome to issue one. This newsletter exists for a simple reason: most of what's written about AI is either hype you can't act on or documentation you can't read. Each issue takes one tool, workflow or idea and explains it the way an operator needs it explained — what it's for, what it costs, and how to wire it into a real business.

So let's start with the question every founder asks first: "what should I automate?"

The impressive task is a trap

When people imagine AI in their business, they reach for the biggest, most visible problem — the sales strategy, the hiring pipeline, the "AI assistant that runs everything." Those projects stall, because the hard part isn't the AI. It's that nobody can describe the task precisely enough for a machine to do it.

The tasks that automate beautifully are the ones you'd never put in a pitch deck. Copying lead details from email into a spreadsheet. Sending the same three follow-up messages. Renaming and filing invoices. Turning a call recording into notes. Boring, repetitive, and precisely describable — which is exactly what makes them automatable this week, not next quarter.

Four filters for the first task

Run your week through these filters. A task is worth automating first when all four are true:

• It repeats. Weekly or more. A task you do twice a year isn't worth the setup cost.
• The steps don't change. If you can write the steps down for a new hire, a machine can follow them too.
• It moves data between two tools. Email → spreadsheet, form → CRM, calendar → invoice. The gap between tools is where hours quietly disappear.
• Nobody enjoys it. Automating a task someone likes creates resistance. Automating a task everyone avoids creates fans.

Most businesses that run this exercise find three to five candidates in twenty minutes. Pick the one that stings the most and start there — a Zapier or Make automation, or a well-prompted AI step, usually covers it without writing code.

Why starting small wins

The first automation isn't really about the hours it saves — it's about proving to yourself and your team that this works. A boring task automated end-to-end this week beats an impressive one half-built for three months. Momentum compounds; stalled projects don't.

Next issue: the actual toolkit — which no-code tools earn their subscription fee, and which are shelf-ware in disguise.

— Ugo, U-go Support Hub