Front-load the opening seconds

1. The rule

Put 80 percent of your attention on the first five seconds.

The objection comes immediately, and is answered in advance: that cannot be right, because five seconds does not take that long to make. It should take that long. If you are doing enough research and enough homework, and you are talking to the right people, nailing the first five seconds makes almost everything after it work.

Statements that sit with it:

• People noticing your advertisement is the most important part of it, by a lot.
• In one practice here, the first impression is the part tested most.

2. Why the front matters that much

Every second of an advertisement has one job: selling the section that comes after it. Read an advertisement as many content units spinning inside one package, each one earning the next.

The headline is the first sale.

The maxim. Once you have written the headline, you have spent eighty cents of your advertising dollar. The view here is that it is one of the most impactful things learned about advertising, for the same reason: if you do not get them at the beginning, nothing after the beginning matters.

3. The production method: separate the hook from the value

The method that produces the volume is separating the hook from the value and recombining them.

Step	What you record
1	Maybe 10 value pieces — the meat. Each one breaks one belief, or demonstrates one skill, or unpacks one nuance the audience did not understand, drawn from experience
2	30 different hooks, recorded separately
3	Paste each hook onto each value piece

That gives 300 advertisement variations from one sitting.

Check the arithmetic: 30 hooks across 10 value pieces is 300 combinations.

Then test in order. Find the best-converting hook first. Then, behind that hook, find which value piece converts best. That sequence is what produces winners and occasional home runs repeatedly, and it is what makes scaling possible at all.

What one value piece contains. Each of the ten is one thing, not a summary of several: one belief broken, or one skill demonstrated, or one small point the audience did not understand as well as they thought, taken from something you have actually done. That is what makes it the meat of the advertisement rather than the setup.

Record the value separately from the hooks. The separation is the whole trick, because it is what lets thirty openings be tried against the same body without re-recording the body. This is a tactical note from practice, not part of the formal framework.

Hooks per body, set against SOP 119. Here, about ten bodies share thirty hooks. SOP 119, section 7, puts ten or more hooks on each body. This page's ratio costs more recording and less hook writing; SOP 119's costs more hook writing and less recording. This page does not settle which reading is right.

4. How to spend against 300 variations

Having 300 variations is only useful if you can afford to find out which one works, so the budget runs the other way from the usual instinct:

Step	What you do
1	Spend very small budgets across the variations
2	Let the test tell you quickly which ones are working
3	Scale the ones that are doing well

The precondition on every step above is tracking. You can do this only if you are tracking, because without it you cannot tell which of the 300 produced the result, and the whole method collapses back into spending money on advertising in general.

5. Why advertisements stop scaling

The common belief is that the platform has capped you — that you cannot get past a thousand dollars a day, or ten thousand, or a hundred thousand, whatever number you have hit.

The platform has not capped you. It wants your money. What is capped is how good the advertisement is, and that is probably because you are not making enough of them. An advertisement reaches a certain warmth of audience, and as soon as it goes beyond that audience it stops converting. The ceiling you are seeing is the quality of the advertisement, expressed as a budget.

The measured difference, with the condition it carries. The top 1 percent of advertisers on one large platform split test 11 times more content than everybody else, and for that reason they scale 10 to 100 times more. That figure is about one platform's advertisers and is a comparison between groups; it is not a promise that eleven times the testing returns a hundred times the scale for you.

SOP 40, section 2.1 puts the same finding at the top one tenth of one percent. Read as the top 1 percent, the elevenfold testing describes a broad leading group; read as a tenth of a percent, it describes only the very top, and says nothing about how much the merely good test. This page does not settle which reading is right.

The diagnosis you can run in one question. Ask an advertiser what their cadence is — how many advertisements do they make? The answer is usually "tons." Ask what that means and it becomes something like once a month, at least ten of them. That is the problem, and no further investigation is needed to find it.

The exception, stated as a condition. Unless your targeting is very wide, scaling requires significantly more split testing across the hook, the value itself and the call to action, in order to find the one package that goes on scaling and stays profitable at scale.

6. Where the effort actually goes

Part of the advertisement	Share of your attention
The first five seconds	80 percent
Everything after it	The remainder

Part	How many you make
Value pieces	About 10
Hooks	30
Combinations available	300
Which you test first	The hook
Which you test second	The value, behind the winning hook

7. The checklist

Question	The answer
Who runs this	You, or whoever makes the ads
Where the effort goes	80 percent into the first five seconds
What the headline is	The first sale
What each second is for	Selling the next section
How you get volume	Record value and hooks separately, then combine them
How many hooks	30
How many value pieces	About 10
How many variants that gives	300
What you test first	The hook, then the value behind the winning hook
Why ads stop scaling	The advertisement is capped, not the platform
How much more the best advertisers test	11 times more content, and they scale 10 to 100 times more
How many tests before a winner appears	Not established on this page
How long a test should run	Not established on this page
What a very small test budget is, in money	Not established on this page

8. What this page does not cover

What a call-out is made of is on SOP 32 for words and SOP 33 for images, sound and setting. Which platform to run on, and how to research what already works there, are on SOP 35. Getting the advertisement in front of the right people is on SOP 36, and where they land afterward is on SOP 37. Turning content that already performed into an advertisement — which skips most of this page's production problem — is on SOP 38. The reading level to write it all at is on SOP 31.

Budgets, bidding and the cost arithmetic behind a scaling decision are not covered on this page.