At most one email a week. Unfollow anytime with one click.
Transcript
The full narration of this video. The current sentence is highlighted while the video plays; click any sentence to jump there.
Show transcript
Introduction
Hi, and welcome back to the Daily Productivity session. After weekly plans and meeting minutes, we now use Claude as a thinking partner. In this episode we learn focused brainstorming: not just a list of random ideas, but ideas with a goal, scored clearly, and ready to test.
Step 1: Why AI ideas often feel generic
Let's start with the problem. The first reason brainstorming fails is a request that's too general, such as simply writing, give me promo ideas, with no context at all. That leads to the second reason: cliché ideas. Discounts, giveaways, and influencer collaborations. They're not wrong, but everyone has already thought of them. Third, the ideas are beyond your means. An idea that looks great on paper turns out to need a big budget or people we don't have. Fourth, brainstorming stops at a list. There are twenty ideas, but none gets chosen, let alone carried out. Compare two ways of asking for ideas. The general prompt has no context. The focused prompt says this is a small coffee shop in Bandung that sells through WhatsApp. The goals are different too. A good promo is vague. Rainy-day sales up twenty percent within a month is clear and measurable. Then the constraints. A budget of five hundred thousand rupiah, and the promo must be something Dimas can run alone. Constraints actually spark ideas that are more creative and more realistic. Finally, the output shape: ten varied ideas, each one sentence long with an estimated cost. This format makes scoring easier later. Remember the principle: the clearer the direction we give, the sharper the ideas that come back.
Step 2: Give a clear direction
Step two is giving direction. Laras writes a prompt with four parts. The context: Sunset Coffee sells through WhatsApp and Instagram, and when it rains, orders drop by almost half. The goal is measurable: rainy-day sales up twenty percent within one month. The constraints are real: a budget of five hundred thousand rupiah, run by one admin alone, and no discounts above fifteen percent. The format is ten varied ideas. Laras also lists the ideas she wants to avoid, giveaways and paid endorsements, so Claude has to look for other paths. Laras sends that prompt to Claude and asks for ten ideas. The results are far more specific. There's a Rain Pack with free delivery when it rains, a WhatsApp reminder before the rain starts, a stamp card just for rainy days, and an offer to nearby offices for afternoon meetings. Every idea has an estimated cost, so Laras immediately knows which ones fit the budget. But these estimates are still Claude's guesses. Laras has to work out the real numbers herself. Sometimes the ten ideas turn out to be variations of a single idea, for example all about price cuts. That's a sign we need to force some variety. Here's how: ask for two ideas in each category: product, price, timing, sales channel, and service. That pushes Claude to look from different angles, not just change the discount number.
Step 3: Expand and sharpen
Step three is expanding and sharpening. There are four simple techniques. First, the customer's point of view. Ask what one particular type of customer needs when it rains. Second, flip the habit. Rain is usually seen as an obstacle. What if rain became the very reason to order? Third, combine two ideas. The Rain Pack plus a reminder before the rain can become one complete campaign. Fourth, ask for criticism. Ask Claude to name the weaknesses of each idea and who probably won't be interested. Criticism stops us from falling in love with a fragile idea. Laras tries two techniques at once. She names her two main customer types, students in boarding houses and office workers, then asks for criticism of idea number one. Ideas appear that she hadn't thought of: a shared one-liter pack for a whole boarding house, and an afternoon coffee subscription for office meetings. The criticism is useful too. Free delivery in the rain can overload the courier and eat into profit on long distances. So Laras starts thinking about limiting the delivery distance. These techniques work for any job. Nadia, a social media freelancer, uses them to find content ideas for clients, and Mr. Hendra uses them to find new ways to explain fractions to his students.
Step 4: Score and choose
Step four is scoring and choosing. Laras asks Claude to build a score table from one to five for three criteria: impact, low cost, and ease. The first idea, the Rain Pack with a three-kilometer limit, scores twelve. The reminder before the rain gets the top score, fourteen. It costs almost nothing and Dimas can easily run it. The rainy day stamp card scores ten, because its impact only shows after several weeks. The afternoon coffee plan for offices also scores ten. The impact is big, but it isn't easy, because Laras has to visit offices one by one. The boarding-house shared pack scores twelve, the same as the Rain Pack. Claude's scores are only a starting point. Laras corrects them, for example by raising the office plan's score because she already knows two offices near the shop. Your experience on the ground is still the most valuable. Before deciding, there are a few checks. First, the idea doesn't copy a competitor's promo exactly. Being inspired is fine; copying it wholesale is not. Second, you've calculated the cost yourself with real ingredient prices and delivery costs, not Claude's estimate. Third, the team can run it without overtime. Fourth, the promo doesn't promise anything uncertain, such as thirty-minute delivery in heavy rain. One thing to avoid: don't pick too many ideas at once. Pick one or two, so the results can be measured clearly. Laras picks the reminder before the rain and the Rain Pack.
Step 5: Turn it into a trial plan
The last step turns the idea into a trial plan. Laras asks for a two-week plan for the two chosen ideas, with the target, the steps, and the numbers to record. Claude builds the plan in the right panel. There's a target of twenty percent more on rainy days, a budget of five hundred thousand rupiah, and Dimas as the owner. Below that are the daily steps. Laras adds one important thing: when the decision gets made. Claude puts it in the last step. On day fourteen, the results are compared with the previous two weeks. This is what makes brainstorming focused. Ideas don't stop at a list; they become a small experiment with clear numbers. If it works, keep going. If it fails, we learn at a low cost. Let's sum up. First, direct it with a measurable goal and real constraints. Second, multiply into varied ideas, while naming the cliché ideas you want to avoid. Third, expand through customer viewpoints, then ask for criticism. Fourth, score with a simple system, and correct it with your own experience. Fifth, test on a small scale with clear numbers. Try this flow on a problem in your own business or job, such as getting more students for your tutoring, ideas for an office event, or content for next month.
Done!
We've learned focused brainstorming: giving a goal and constraints, expanding ideas from the customer's point of view, scoring them, and turning the best idea into a trial plan. In the next episode we use Claude as a personal tutor to learn a new topic quickly, without getting lost in too much information. See you there!
Practice exercises
Try these on your own after watching to lock in what you learned.
Write a brainstorming prompt with a measurable goal, a budget limit, and an idea format
Ask for extra ideas from the point of view of two types of your customers
Score your top five ideas for impact, cost, and ease, then design a two-week trial
Ask the class material
Type a question or keyword; we find the part of this episode that explains it and jump the video there.
My notes
Notes are saved with the current video time and are only visible to you.
Questions & answers
Stuck on a step? Ask here; the instructor and other learners can answer.
No questions yet.
Your instructor
Cerdas Inovasi Digital Team
Software developers, BIM & 3D engineers, and AI practitioners
53 video classes1,278 episodes
We build software, BIM models, games and AI automations for clients every day, and turn that hands-on work into step-by-step classes: every step is shown on screen and comes with a PDF guide you can keep.
