Diagnose your stuck project fast

You know the one. It has a folder. Maybe a Notion doc with fourteen edit timestamps. Maybe a half-written plan you have read so many times the words have stopped meaning anything. It lives in that corner of your brain that goes off like a smoke alarm every Sunday evening and then politely shuts up by Monday lunch.

That project. Open the chat thread where you have been working on it. The actual thread. Not a fresh chat.

You know the fix: longer prompts, more context, clearer constraints. But typing all that takes five minutes per prompt, so you shortcut it. Every time.

Wispr Flow lets you speak your prompts instead of typing them. Talk through your thinking naturally — include context, constraints, examples — and get clean text ready to paste. No filler words. No cleanup.

Works inside ChatGPT, Claude, Cursor, Windsurf, and every other AI tool. System-level, so there's nothing to install per app. Tap and talk.

Millions of users worldwide. Teams at OpenAI, Vercel, and Clay use Flow daily. Free on Mac, Windows, and iPhone.

Try Wispr Flow free

One prompt. Suddenly you have a consultant who has read every message in that thread and knows the full context. Not a generic advisor pulling frameworks from a paperback. Someone who actually knows what you tried, what stalled, and the quiet assumptions you have been building the whole plan on.

Most people use AI like a search engine. Ask a question, get an answer, close the tab. That is fine for facts. It is useless for projects.

Drop this prompt into a long thread and something different happens. The model is not searching for information. It is synthesizing everything you have already said and reflecting back the shape of it. The shape is the part you cannot see, because you have been living inside it.

That gap between what you think your project is and what it actually turns out to be when analyzed cold, that is where the value lives. Everything else is just busywork dressed up as planning.

The DTC beauty category is crowded. To break through, Jennifer Aniston’s brand LolaVie, worked with Roku Ads Manager to easily set up, test, and optimize CTV ad creatives. The campaign helped drive a big lift in sales and customer growth, helping LolaVie break through in the crowded beauty category.

Learn more

*Ad

After the diagnostic, work these questions in order. Ask the AI directly if any of them feel fuzzy.

🔹 What is the biggest bottleneck? The one thing that, if removed, would make everything else move faster. Name it out loud.

🔹 What is the biggest unnecessary complexity? Complexity almost always hides inside the bottleneck. You cannot see what to simplify until you have named what is blocking you.

🔹 What is the biggest leverage point? Only makes sense once you know what is actually stuck. A leverage point on the wrong problem is just a faster way to circle.

🔹 What can actually be done with your current time, energy, and resources? The reality check. This is where most plans quietly die. Not because the idea is bad, but because the plan needs resources that do not exist yet.

🔹 What is the next concrete action you will take? Only honest after the first four. Before that, it is activity disguised as progress.

Most people skip straight to step five. Then they wonder why they have been moving for six weeks and the project still looks the same.

If the AI surfaces something that makes you uncomfortable, sit with it. The discomfort is the signal. The best diagnostics do not confirm you are on the right track. They name the thing you already knew was off and were not saying out loud.

If nothing surprises you, the project probably is not the problem. Your relationship to it is.

Pay attention to which word keeps showing up. If it keeps coming back to "assumptions," you are building on unvalidated ground. If it keeps flagging "sequencing," you are doing things in the wrong order and creating dependencies that did not need to exist. If "leverage" keeps appearing, there is almost always one move on your list that is ten times more effective than the others, and you have been treating it like just another task.

Also notice what the AI does not say. If you expected it to flag a specific part of the plan and it skips past it entirely, that is data too. Either that part is fine, or you have not given enough context for it to be evaluated honestly. Both are worth knowing.

AI agents already outnumber human visitors to your docs — now you can track them.

See your AI traffic!

*Ad

🔹 Do not run this every day. Save it for when new evidence shows up, something fails, constraints change, or you hit a real wall. Run too often and you start optimizing instead of building.

🔹 Run it in the working thread, not a fresh chat. A fresh chat has no context. The diagnostic needs history to be useful. The whole point is the synthesis.

🔹 When the AI says "delegate" or "automate," do not skim past it. That is usually where the real time is hiding. The thing you keep doing yourself because it feels faster than explaining it.

🔹 If you have been working alone on this for a long time, your thinking has grooves. You have worn certain mental paths smooth. This prompt is partly useful because it forces a perspective that has not walked those paths yet.

Open one stalled project. Paste the prompt. See what breaks loose.

Worst case, two minutes and you learn something. Best case, you stop circling the same problem and actually move.

Check out the original post

Run the diagnostic on your stuckest project today. When the AI's output comes back, one word will show up more than the others. That word is the diagnosis.