In banking, customer complaints get logged with care. Regulators expect it, and nobody wants the same problem landing on the CEO's desk twice.

For years, I watched those logs do one job very well. Each complaint got a ticket, someone resolved it, the ticket closed, and the monthly report showed a tidy resolution rate that made everyone feel good about the month.

What the report rarely showed was the patterns of what was really happening. Three branches might each hear from customers about the same confusing fee notice, and because every complaint was handled politely and closed on time, nobody connected them. A fourth branch would get the same complaint the next month, and the cycle started over.

Treating complaints as individual problems has a hidden cost. Your team gets very good at apologizing for something that keeps happening, but nothing really changes to stop it from happening.

AI is well suited to this kind of work. Give it a batch of complaints with the personal details removed, and it can sort them by theme, separate what customers said from the likely cause, and point out which issues show up across locations, products, or stages of the customer journey. It also reads the hundredth complaint with the same attention it gave the first, which is more than most of us can say by Thursday afternoon.

Do note that we are not handing any decision-making to the AI. It can show you that a dozen complaints point to the same notice, but you decide whether the fix is a rewrite, a training session, or a conversation with the people who designed the process.

That is our 90/10 rule at work. Most of the effort is human preparation, such as choosing the right complaints, removing personal details, and deciding what a useful answer looks like. The bounded sorting work is what you hand to AI, and your review continues after it responds.

Today's brief turns a pile of complaints into a short list of patterns worth fixing.

Let's go raise your AIQ.

If you have been reading AIQ with Paul for a while, today's Master Prompt will look a little different. Starting this week, every prompt is written as an AIQ Brief, the same format I now teach in the AIQ Advantage™ program.

The older prompts opened by giving AI a role and then spelled out every step. Today's AI tools plan their own steps well, so the most useful thing you can give them is what only you know, such as your situation, your standards, and the facts that cannot change. A brief covers what any capable new partner would need to know before starting the work.

You will also notice the prompts are shorter. Much of the old length went into telling AI how to work, and the brief spends those words on your situation instead.

Fill in the brackets the same way you always have, and give the Check section extra attention, since it tells you what to confirm before you use the result.

The Master Prompt is just the beginning of what you are missing as a free, non-subscriber.

AIQ with Paul™ subscribers are currently reading the rest of today's intelligence brief, which includes:

> Join the network of everyday people raising their AIQ. And even better: It’s completely free!

Click the button below to get tomorrow's brief delivered straight to your inbox.

Oh, and you will be able to see this and all of the previous Daily Prompts here on the website!