Monday morning, my to-do list gained three more ideas

At 9 a.m. last Monday, I sat in a café in Shanghai’s Jing’an district with three windows open on my laptop: client delivery, content topics, and product development. One AI tool was rewriting copy, another was generating a product prototype, and a third turned “I might as well build a small tool” into a new project. By evening, all three had barely moved forward, and I had no energy left.

I used to think inefficiency was an execution problem. Later, I realized I was often pulling myself toward distraction on purpose. As non-technical founders, it’s easy to assume AI makes trying things “probably fast.” Each small idea still swallowed half a day.

The fix wasn’t more tools, but choosing one main battle

French developer Alejandro AO, writing an annual review on Hugging Face, said he once had several AI projects running simultaneously, even working with four to ten AI assistants at once. Many of those projects were eventually abandoned. His experience matched mine: plenty of visible output, but no time left to polish what actually mattered.

My approach is simple: before work starts each day, I choose one main project that “must get better today.” Everything else falls into three groups: continue, pause, or revisit later. AI still helps, but only to accelerate that main project—not to open new fronts. If I’m building a quote page for a client, I first clarify the pricing, the client’s needs, and the delivery date, then ask AI to help revise it. I don’t also work on a course, a community, and a new product.

Today’s version costs almost nothing

Money: RMB 0; use the AI chat tool you already have. Time: 15 minutes daily to set the priority and 30 minutes to review. Technical barrier: No coding knowledge needed—just a clear answer to “What must I deliver today?” First step: Open your AI tool, start a new chat, and enter: “I can only move one main project forward today. Help me break it into three steps and remind me not to start any new projects.”

I got stuck here too. At first, I interpreted “focus” as doing less, which made me anxious. The better approach was not to cut everything, but to decide what happens today, what happens this week, and what does not happen yet. Not everyone needs this method. If you need projects and have no clients yet, trying more things may make more sense right now; there’s no rush to try this.

Choose a focus that fits your current stage

If you’re just starting: I’d put 70% of my time into finding one real client and completing one delivery, then test ideas with the remaining time. AI can help write the first message and quote, but I wouldn’t build three products at once.

If you already have 1–2 clients: I’d turn recurring delivery steps into checklists and templates, then let AI handle rewriting, organizing, and replies. I’d use the saved time to improve quality or add services.

If you’re scaling: I’d identify which delivery work takes the most time and causes the most errors, then hand that process to AI while keeping human review. The danger of scaling isn’t being busy; it’s everyone producing shallow results.

Today, I can pick just one thing. For the next hour, I won’t open a new front. I’ll move one important piece forward.