Copywriting Riddle Tests Your Skills

[Stephen Dean]

If you test this headline:

FORMER BARBER EARNS $8,000 IN FOUR MONTHS AS A REAL ESTATE SPECIALIST

against this headline:

HOW I MADE A FORTUNE WITH A "FOOL IDEA"

And headline #1 wins, does it prove that the first headline is more specific and thus stronger, or does it prove that a 69 character headline is stronger than a 39 character headline?

Cheers,
Stephen Dean
Copywriting Dean | Copywriter Stephen Dean Talks Advertising

[Ken Strong]

Originally Posted by Stephen Dean

And headline #1 wins, does it prove that the first headline is more specific and thus stronger, or does it prove that a 69 character headline is stronger than a 39 character headline?

Strictly speaking, the only thing it proves it that Headline #1 beat the other headline in this specific test to that specific audience at this specific time (and whatever other specific test conditions there were).

Did this headline actually win a split-test, or is this a trick/rhetorical question?

[Ken_Caudill]

The second headline is pretty damned famous.

Did you actually test them?

[Cam Connor]

Originally Posted by Ken Strong

Strictly speaking, the only thing it proves it that Headline #1 beat the other headline in this specific test to that specific audience at this specific time (and whatever other specific test conditions there were).

+1. It's almost impossible to "prove" anything in copywriting... Having done a million tests like this, I still can't "prove" anything really. You can just get a TON of evidence, one way or another.

[travlinguy]

I can't imagine the first headline beating the second under almost any circumstances.

[Cam Connor]

Originally Posted by travlinguy

I can't imagine the first headline beating the second under almost any circumstances.

Lol, that's copy for ya.

[Stephen Dean]

They're both headlines from the same famous list, and no I didn't test them. So probably a poorly formed question. It was supposed to be a hypothetical about what knowledge you can obtain from a test.

People always say to test ONE THING. So they may choose to test the headline.

But then... you're still changing more than one variable. The length of the headline, the number of chars, the letters, the words... rarely are you changing just ONE variable.

This isn't a problem as long as you understand what the test means and what it does not mean. That's what I was trying to get at, maybe poorly It didn't help that the ex-cabbie knocked it out of the park on the first pitch.

Cheers,
Stephen

[nthmarketing]

Although there are many variables I think that as long as you keep testing it would prove whether it's length or not. I believe it would always be the message that is what is being tested. So if the shorter one wins and you bring a longer one that wins the next split test I think again you can conclude it's the message and has nothing to do with the length.

Of course I might be over analyzing it as well.

2 Cents,
Fred

[Scott Murdaugh]

It proves that headline #1 wins for that specific market.

I'd be very surprised to see a huge change in response from a 69 character headline to a 39 character headline.

Both are readable within ~3 seconds or less.

When you start talking LOOOONG headlines it's a different story.

As a general rule, specificity is always more powerful.

That's a *general* rule.

From that test, I personally wouldn't take length into consideration.

Although I may test...

FORMER BARBER EARNS A FORTUNE WITH A "FOOL IDEA"

Or...

HOW A FORMER BARBER EARNED $8,000 IN FOUR MONTHS AS AN "ACCIDENTAL" REAL ESTATE SPECIALIST

And variations upon that, combine that with tracking software (my headlines aren't great, it's late, thinking fast) and try to understand the psychology of what's working and why.

I'd say the market, the mindset, the psychology and the hook will overwhelmingly be more influential in the success of the ad than the actual length of the headline.

My .02.

-Scott

P.S. Stephen, long time no talk man, let's catch up soon!

[Scott Murdaugh]

Can someone ban this idiot? ^^^

We have plenty of off-topic non-nonsensical posts without him

(J/k, love this forum most of the time. But spam bot troll is obvious).

-Scott

P.S. Just in case his post (dabao) is deleted I'm NOT suggesting banning myself. But if I ever start replying like this guy, please ban me for my own good

[ThomasOMalley]

One of the easiest tests to do for your sales letter or web page is test one headline against the other. I would not be concerned about how many characters you are using etc.

If one headline beats another headline, use that headline.

[Scott Murdaugh]

If one headline beats another headline, use that headline.

Definitely true.

But I'd always try to analyze the reasons why one headline beat another. Once you understand the underlying psychology, motivators, appeal and other factors, it makes that success much easier to duplicate.

I've seen many headline tests, and more times than I can count, I've been completely shocked at the results. It's definitely worth a deeper look, beyond just "A" beat "B".

In one test (can't post the headlines, client) I saw a pretty vague headline, with lots of superlatives and pretty long headline lacking any specifics beat a shorter headline promising very specific results backed up with proof.

I had 2 guesses as to why...

1. The specific promise was true, but the results were too good for the market to believe they could achieve anything similar.

2. The longer headline was sort of a cliffhanger, getting those curiosity glands going.

Test, and go with the winner.

Understand WHY it's a winner, and gain invaluable insights into your market and human psychology in general, which in turn, makes you a much more intuitive (and effective) copywriter.

I get a big kick out of seeing what works and trying to figure out why.

-Scott

[RAGolko]

You'd have to know the market it was tested in to be sure. The same headlines in a diffrent medium could produce opposite results. That's why split-testing is so hard. If you are able to isolate all variables then you can be more conclusive but do it again in a month and it might change again.

[Daniel Scott]

But therein lies the rub, Scott (and from what I gather, this is Stephen's point...)

You can only guess why something does or does not work.

You can say with a fair amount of certainty x beat y... but saying WHY x beat y is a very difficult thing to do with any amount of surety.

Now... I'm not saying split testing isn't important.

But some people treat testing as the answer to all life's problems. And while it's important... it's only one of the things you need to be doing to become a great copywriter, marketer and salesperson.

I know YOU know all this... but people reading this thread may not (if the people I've talked to in the past is any indication). Some people think testing REPLACES educated guesswork, creativity and thoughtfulness. And, as you and I both know, nothing could be further from the truth.

-Daniel

Originally Posted by Scott Murdaugh

Definitely true.

But I'd always try to analyze the reasons why one headline beat another. Once you understand the underlying psychology, motivators, appeal and other factors, it makes that success much easier to duplicate.

I've seen many headline tests, and more times than I can count, I've been completely shocked at the results. It's definitely worth a deeper look, beyond just "A" beat "B".

In one test (can't post the headlines, client) I saw a pretty vague headline, with lots of superlatives and pretty long headline lacking any specifics beat a shorter headline promising very specific results backed up with proof.

I had 2 guesses as to why...

1. The specific promise was true, but the results were too good for the market to believe they could achieve anything similar.

2. The longer headline was sort of a cliffhanger, getting those curiosity glands going.

Test, and go with the winner.

Understand WHY it's a winner, and gain invaluable insights into your market and human psychology in general, which in turn, makes you a much more intuitive (and effective) copywriter.

I get a big kick out of seeing what works and trying to figure out why.

-Scott

[doingwrite]

I think it has less to do with the length of the headline than with what the headlines actually say. I know that "the shorter the better" is pretty much the rule, but these are words after all and it is words and how we put them together that do the selling.

[IM Gold]

thats the joy of split testing, when you change headline, body and image its hard to tell which has the greatest effect

[ThomasOMalley]

Originally Posted by IM Gold

thats the joy of split testing, when you change headline, body and image its hard to tell which has the greatest effect

The key is to test only one key component of your copy, such as your headline or offer, at a time against your control.

Focus on testing your key components of your copy.

[amo992]

Stats 101:

Testing the two headlines is an observational study, not an experiment, which means you can only claim correlation. Correlation is not causation.

In this case, RAGolko is right. The headline is correlated with the niche it's used in, and the conclusions would be as follows. "Title 1 worked well with the niche" vs. "Title 1 did not work well with the niche." Or "Title 2 worked well...you get the idea.

To perform an experiment you need independent variables, or variables you can control and change, and dependent variables, or variables that change as a result of your changes. And remember, nothing is ever 'proven,' it's 'supported.'

[darkonetoo]

Originally Posted by Scott Murdaugh

Definitely true.

But I'd always try to analyze the reasons why one headline beat another. Once you understand the underlying psychology, motivators, appeal and other factors, it makes that success much easier to duplicate.

I've seen many headline tests....

Test, and go with the winner.

Understand WHY it's a winner, and gain invaluable insights into your market and human psychology in general, which in turn, makes you a much more intuitive (and effective) copywriter.

I get a big kick out of seeing what works and trying to figure out why.

-Scott

I agree, but with a serious proviso.

It really matters how many results (actual numbers of conversions, etc) you are dealing with in your split testing (ST). ST is more about statistical significance than many testers realize. Statistical confidence numbers are calculated mathematically, and you need at least 90% "confidence" to choose a winner. There are free online tools to determine the limits you need to be watchful of.

Examples: (with data from: Landing Page Optimization: The Definitive Guide to Testing and Tuning for Conversions)

• If there are 100 responses in your sample, you need to see a 20% difference between the variations tested to be sure that they actually mean something.
• If there are 1,000 responses, you need a 6.3% difference.
• If there are 10,000 responses, you need a 2% difference.
• If there are 100,000 responses, you need a 0.063% difference.

So bigger-is-better, at least when split testing samples are involved. Otherwise, with small samples, the difference between test's success rates can easily be result of random chance.

Fortunately, the majority of split testing results differ by larger margins when dealing with a passionate audience.

DarkOneToo

[CopyWriteHer]

Ceteris Paribus. I agree, Thomas. If you are testing, you have to change just one element at a time, else you lose the valuable information of which thing made the difference.

Originally Posted by ThomasOMalley

The key is to test only one key component of your copy, such as your headline or offer, at a time against your control.

Focus on testing your key components of your copy.

[apolwar]

Originally Posted by Stephen Dean

And headline #1 wins, does it prove that the first headline is more specific and thus stronger, or does it prove that a 69 character headline is stronger than a 39 character headline?

Cheers,
Stephen Dean
Copywriting Dean | Copywriter Stephen Dean Talks Advertising

It isn't how long or how many words there are in a headline. Its how clear the message of the headline is.