Building a StoryBrand: the StoryBrand framework, and where it breaks

What are the seven parts of the StoryBrand framework?

Miller's argument is that a buying decision follows the shape of a story, and that most companies cast themselves as the hero when they should be the guide. The script, SB7, is seven parts in order:

1. A character. One customer, with one want.

2. Who has a problem. At three levels, which is the part people skip.

3. And meets a guide. Who shows empathy and authority.

4. Who gives them a plan. Three or four named steps.

5. And calls them to action. One direct ask, repeated.

6. That helps them avoid failure. The stakes, said once.

7. And ends in success. A specific picture of life after.

The three levels of problem are the most useful idea in the book and the most skipped. The external problem is the practical one. The internal problem is how it makes the customer feel. The philosophical problem is why it is wrong that things work this way. Most pages stop at the external one, which is why they read as accurate and dead.

Worked, for a B2B reporting tool:

• External: the weekly numbers take three days to put together.

• Internal: you walk into Monday's meeting unsure whether your own figures are right.

• Philosophical: nobody should spend a third of their week assembling numbers a machine could assemble.

The internal line is the one people quote back to you. It is also the one almost no company writes down, because it feels unprofessional to say out loud.

The other thing worth taking from the book is the grunt test. Can a stranger, looking at your page for five seconds, say what you offer, how it will make their life better, and what they have to do to get it. If any of the three is a guess, the page is not finished.

In what order should you build the page?

This is the sequence I use, with the lengths, because the framework tells you what goes on the page and not how much of it.

1. The one-liner first, before any layout. Problem, solution, result, in one or two sentences you could say to someone at a table. If you cannot say it, no page design will hide it. This is a day of work, not an hour.

2. The hero section. A headline carrying the external problem or the result. One sub-line of no more than 20 words. One button. Nothing else. Then count the sentences in the first screenful that start with "we" or your company name, and get it to zero.

3. The problem, in the customer's words. Three lines, one per level, with the internal one in the middle where it gets read.

4. The guide: empathy, then authority. One sentence of empathy. Then proof, and here you have to be stricter than the book is. Named customers, a number with a date and a source, or a piece of the work itself. Logos without context are decoration.

5. The plan, in three steps. Name the steps using the words for what the customer does, not what your system does. This is the slot that most often turns into a lie, which I come back to below.

6. One call to action, repeated. The same words every time it appears. Label it with what they get: "See your own data", not "Get started".

7. The stakes, once. One sentence. Two reads as a threat.

8. Success, specific. What Tuesday looks like after this works. Not "grow faster".

Where does the framework break?

It makes everyone sound the same. The pattern is now recognizable on sight: "Most [role] struggle with [problem]. We get it. That's why we built [product]. Here's our three-step plan." A reader who has seen that structure on 40 sites reads the shape before they read the words, and the shape says "this company read a book". Use the script to decide what goes where, then write the sentences as yourself.

It optimizes for clarity, not for difference. Clarity is the right first goal, and it is where most pages fail, so the book is right to aim there. But if your closest competitor runs their product through SB7 and lands on the same external problem and the same promise, you have both achieved clarity and neither of you has a reason to be chosen. The framework has no slot for "why this one and not that one".

It assumes one hero. In a committee purchase, the person who uses the product, the person who signs for it and the person who can block it have different internal problems. The user is worried about looking slow. The signer is worried about wasting a budget line. The blocker is worried about the migration landing on their desk. One character cannot carry all three, and a page written for the user alone stalls in procurement. I write one page for the hero and a second block, lower down, for the people who have to approve it.

The plan slot invites a promise the product does not keep. "Sign up, connect your data, get your first report" is a lovely three-step plan, and if connecting the data actually takes a developer and two weeks, the page has just written a check that activation has to cash. This is the single most expensive mistake I see with the framework, and it is why it sits in the activation stage of this library rather than in acquisition. Write the plan steps only after you have watched somebody new actually do them. The gap between the plan on the page and the first five minutes in the product is the thing user activation is mostly about.

It is silent on proof and on search. The authority slot sets no bar for evidence, and the book has nothing to say about the words people actually type when they go looking. A page can be a perfect StoryBrand page and contain none of the language of the problem as its buyers describe it, which means nobody arrives to read it.

Who should read it?

Founders writing their own site for the first time, and anyone who has inherited a page nobody in the company can explain. It is short, it is clear, and it will fix the most common failure, which is a page about the company. Read it once, build the page, then put the book down and rewrite the sentences in your own voice.

Skip it if your page already makes the customer the hero and your problem is that nobody believes you. That is a proof problem, and Making Websites Win is the better book for it.