When It Comes To High-Performing Landing Pages, It’s Not Long Or Short. It’s This…
If you’ve worked in the conversion optimization industry for longer than say, fifteen minutes, you’ve almost certainly stumbled across someone making the argument for longer sales pages. Or shorter sales pages.
It usually sounds something like this…
“People have short attention spans. Who’s going to read something this long?”
“Long pages are ugly. What’s with the yellow highlighter?”
“You should never make customers scroll to find what they’re looking for.”
Or, from the other side…
“Basecamp uses a long sales page. Shouldn’t we follow best practices?”
“Longer pages let you tell the whole story.”
“Long pages give busy people information to skim so they can pull out what’s important to them.”
There are smart arguments (and smart people) on both sides of this ongoing debate. Which can make it hard to figure out who’s right. And who’s not.
When The Daily Egg asked several experts for their thoughts about long sales pages, they got responses like…
“If long copy sales pages are not a thing of the past, they should be. It seems evident that people have short attention spans, so the best way to convert them is to sell them on the product as quickly as possible.”
—Kristi Hines, Freelance Writer
And, advocating for long pages…
“…the sales letter must support several different modes of reading, including: scanning, partial attention, and engaged reading… the methodical reader must be captured and taken though the page in a more traditional sales letter approach.”
—Bryan Massey, Conversion Sciences
And, when push comes to shove, the long copy advocates pull out the biggest gun of all:
“Direct response advertisers know that short copy doesn’t sell. In split-run tests, long copy invariably outsells short copy. But I must warn you that if you want your long copy to be read, you had better write it well. In particular, your first paragraph should be a grabber…
Long copy sells more than short copy, particularly when you are asking the reader to spend a lot of money. Only amateurs use short copy.”
—David Ogilvy, Founder Ogilvy & Mather
“Only amateurs.” Argument settled, right?
What Do Scientists Say About Long And Short Sales Copy?
Given that the argument about long and short copy dates back to at least 1923 when Claude Hopkins published Scientific Advertising, you’d think that academics would find this a rich area for study.
But you would be wrong.
Maybe academics aren’t interested in studying these critical questions. Or maybe they’re too busy doing even more important work, like identifying gravity wavesand finding the Higgs Boson particle.
But that’s not to say that no one has studied this.
One team of researchers found that longer messages on banner ads are associated with higher click-through rates. But even long banner ads are barely long enough to qualify as short sales messages.
A second team of scientists found that readers of an industrial magazine were significantly more likely to read ads containing long, rather than short, copy. The researchers “found that short copy is less effective in arousing the interest of readers. The results suggest that longer ad copy is needed to communicate the type of information sought by industrial prospects, empirically confirming beliefs held by the advertising industry.” But, single-page ads aren’t that long.
A third study, published in the Journal of Interactive Marketingtook a look at two sales letters in a direct mail campaign and found that a two-page letter performed no better than a single-page letter. But again, most marketers would argue that a two-page letter is “short.”
Sadly, academia doesn’t have much to offer here.
Fortunately, online marketers do.
Internet Marketers Have Been Testing This Stuff For Years
Forget cat memes, streaming video, and Facebook. The best thing about the Internet is data. For marketers, the ability to measure everything from scrolling and clicks to conversions and revenue has changed the game.
And that means we can easily see how short pages perform when compared to long pages.
Here’s what a few tests have revealed…
Example 1: Crazy Egg Kills It With Long Copy
Readers of The Daily Egg will certainly be familiar with this test. In fact, thanks to Neil Patel, it may be the most well-known test of long and short copy ever conducted.
In an effort to increase conversions on Crazy Egg, founders Neil Patel and Hiten Shah approached Conversion Rate Experts for help. CRE took a look at the existing page and identified several areas for improvement.
They discovered that visitors weren’t always clear on what Crazy Egg did or how heat maps worked. Potential customers also couldn’t tell whether the data they would get was worth the price they would pay. There were questions about features and more questions about how Crazy Egg differed from free tools.
To address these issues, CRE hypothesized that customers need more information. So they made the page longer—20 times longer—and filled it with detailed information that addressed customer concerns.
Then they ran an A/B test to see which performed better.
The result? The new, longer page outperformed the control by 30%.
Example 2: Crazy Egg Gets Even Better With Shorter Copy
But Neil and Hiten weren’t done. (They’ll never be done.) So Crazy Egg’s founders approached Copy Hackers with the challenge of beating their already optimized and high-performing long sales page. The Copy Hackers team did a lot of additional research (surveys, user testing, heat mapping) to figure out which messages were important for conversions and which were just getting in the way.
Copy Hackers opted to use a similar design because they were testing the message, not the layout. They also did exclusion testing, removing certain elements to see what impact it had on conversions. New headlines were pulled verbatim from key phrases customers repeated during interviews. Then, Copy Hackers proposed four variations of the new page—all significantly shorter than the control and ran the test.
The result? Version D, the short page on the right, pulled 13% better than the long sales page control. The winner isn’t as short as the original (in Example 1), but it is a lot shorter than the new control.
Example 3: Highrise Gets A Big Bump With Longer Copy
The team at Highrise (once part of Basecamp, but since spun off into its own company) had a landing page design that they liked and was performing relatively well. But they realized that some potential customers didn’t convert because they didn’t understand what exactly Highrise did or whom it was for.
After some heated internal discussions about what information needed to be added, several design options were created, tweaked, and reworked until the team was convinced they had a winner.
The new home page was significantly longer and added in-depth information about each of the product’s features and how real customers were using them. It also included a long list of things people accomplished with Highrise as well as customer testimonials and ideas for getting more from the software. Once the page was ready, they ran an A/B test with 42,000 visitors.
The result? The longer page delivered a 37.5% increase in net signups compared to the control. Proof that long copy works.
Example 4: Highrise Gets An Even Bigger Bump By Cutting Copy
But just because version B beats version A, doesn’t mean there’s a version C that isn’t better than both. So the Highrise team went back to the drawing board. They were challenged to test something better. Something radically different. So the designers proposed a new layout featuring a huge picture of a user along with a testimonial headline. The new page was significantly shorter than the control. Again, they ran an A/B test (or an A/C test in this case) with site visitors.
The result? The shorter page increased paid signups by 102.5%. Then the team ran another test, adding back much of the content from the longer page into the new design. The long page version of the photographic design, reduced signups by 22.72%. Shorter copy was a big winner here.
Example 5: 911 Restoration Cuts Landing Page Copy And Increases Conversions
Disaster recovery company, 911 Restoration, relies heavily on SEO and PPC to drive traffic to its website. This traffic is responsible for 90% of its leads. The homepage was relatively long in part because it included important keywords customers search for.
911 Restoration decided to test a shorter, more focused landing page against the longer control after they realized that callers dealing with an emergency didn’t want to read in-depth content about their services. They just needed basic information and a number to call to get things fixed now.
They tested a shortened down page that only hit the highlights against the original, longer home page.
The result? The shorter treatment generated 35 calls from 133 clicks ($4,982.74 in ad spend) compared to 34 calls from 177 clicks ($7,177.96 in ad spend) for the longer version. That’s a 37% increase in conversion and a simultaneous reduction in cost per conversion of 33%—a big win for shorter copy.
Example 6: Moz Makes Their Page Bigger And Ups Conversions
Like Crazy Egg, Moz knows its stuff when it comes to page optimization and conversions. The company has been optimizing its pages for a long time. But they knew they had to do better if they were going to achieve their growth goals.
Working with Conversion Rate Experts, Moz surveyed customers at all stages of the sales funnel, including free trials. Using an iterative process of usability testing, page modification, then testing again, Moz created a new page that added information about the service and spelled out exactly what users get when they purchase the tools. The new page was seven times longer than the control.
Then they ran an A/B test with more than 5,000 site visitors to see how the two pages performed against each other.
The result? The longer page achieved a 51.83% increase in sales of Moz’s PRO memberships, which translated to more than $1million in new revenue. That’s a nice win for longer content.
Example 7: Qi Networks Improves Conversions 70.1% With Shorter Content
Qi Networks is a Brazilian Google Apps reseller. They partnered with Brazilian CRO specialists, GoSupersonic, for help improving conversions on their homepage.
The original page included a lot of information about the reseller’s services. After interviewing customers, they realized that most customers arrived at the website in a “dramatic” moment, when their email would go down (because they used a cheap, ineffective solution). They just wanted to get things back up and running fast.
So the team proposed a shorter page that didn’t explain Google’s Apps in great detail, but instead had a simpler message: Google Apps has the best email in the world. It will solve your problem now.
The result? The shorter page increased conversions by 70.1%. Short copy fans love this kind of proof that short copy works.
Example 8: Quality Training’s Longer Content Gets A Lift
Quality Training offers executive placement services for professionals looking for new career opportunities. The company wasn’t thrilled with the performance of their website and wanted to try something different. Surveys of their customers found that they thought the content on the site was excellent.
However, heat map data suggested that customers were clicking on words in the text as if they were links, looking for additional information. Noticing that, Quality Training decided to test a longer page that gave significantly more information to site visitors. Providing this additional information would also help position the company as the expert in the executive placement field.
The team created a new home page that was three times longer than the original, pushing significant information down the page and below the fold. Then they ran an A/B test to see which would perform better.
The result? The longer page saw an increase of 662% in conversions versus the original page and a record beating sales month following its publication.
By Now You’re Seeing A Pattern. Or Lack Of One
Sometimes long copy wins. Like here. And here.
Sometimes short copy wins. Like here. And here.
We have a tie.
Which is completely unsatisfactory for those who want to declare a winner.
So when do you write short and when should you write long?
Bob Kemper, Director of Sciences at MECLABS, analyzed hundreds of unpublished A/B tests of long and short copy to identify which factors determine whether your copy should be short or long. He found three of them:
The nature of the visitor’s motivation.
Are site visitors familiar with your product and ready to buy or just looking for information about how to solve a general problem? In other words, are they hot or cold? Prospective customers who need your solution now or those who already use your product, usually don’t need a lot of information (or long copy) to convert.
Initial level of anxiety about a product or company.
Has the customer heard of your product and does she trust your brand? Visitors who are familiar with your product or brand will respond better with a short page, while those who are unfamiliar will likely need more content in order to get to know you and build trust.
The cost or commitment associated with conversion.
Is your product cheap or expensive? Does it require your prospect to do something easy or hard? Expensive products often require additional information to help buyers understand the value of what you offer, while inexpensive purchases requires less thought (and less copy). Asking your customer to do something that takes a lot of time and effort may also need longer copy.
Kemper proposed a matrix outlining how conversion decisions align with emotional needs and expense considerations that may prove helpful as you try to decide whether your content should be long or short (I’ve revised the original matrix just a bit):
Conversions that are emotional or impulse-driven, or have a low cost (in time or money) tend to require less information and shorter content. While conversions that are need-driven and more rationally considered (or appear to be on the surface) usually require more information and longer content.
Compare Kemper’s matrix to the State of Awareness Scale that Joanna Wiebe shared on Copy Hackers and you get an even better picture of when you should consider longer or shorter content:
Prospects who have little awareness of your product or brand need a lot more information and explanation than those who already understand that your product solves their problem.
According to Joanna,
“There’s no point in saying “long copy always beats short copy”… or “long copy doesn’t work on me”… or “web users will only tolerate short copy”…
Rather, your page needs to be as long as is necessary to make the argument that will address the prospect in their state of awareness. If you don’t know how aware they are, you need to find out in [before you] shape your argument.”
Now The Really Hard Question…
How long or how short should your copy be?
Of course, as Joanna suggests, it depends on the visitors to your site and their state of awareness. To find out what your prospect needs, you’ll want to conduct the appropriate research and analyze what visitors are looking for on your site.
Remember the Crazy Egg examples above? Look at their home page today. It’s crazy short. Just a headline, a few words about what the product does, and a call to action. (There’s a bit more information if you want it, but only if you click a link.)
How do they get away with such short copy?
Today Crazy Egg is the authority in the heat map space. Potential customers arriving at the site already have a good idea of what they do. Their state of awareness is high. Crazy Egg doesn’t need to educate prospects like they did five years ago. Longer worked then. Shorter works now.
The same has happened at Moz. Five years ago they were just finding traction. Today they are an established leader in the SEO tools space. Many visitors already know how valuable their product is. So the home page has gone from short to long then back to short again.
You can bet that neither of these conversion-obsessed companies made the change without testing to ensure the new pages met their customer’s needs.
It’s Not Long Or Short. What To Measure Instead
Copy length tests are informative, but not because longer or shorter copy is better, but because the elements that change the page’s length tell us what’s going on.
In fact, length isn’t the thing to test at all.
Where and why do customers drop off your page?
What is their state of awareness?
What are they looking for that your page isn’t giving them?
What are you telling them that they don’t care about?
What page elements don’t matter?
What features and benefits should be upfront?
How can you say things more effectively?
Can you say things more concisely?
What words do they use to describe your product (and should you include them)?
Study your heat and click maps. Conduct user tests with real customers. Do meaningful voice of customer research with exit surveys, on-page surveys, and one-on-one interviews. Figure out what competitors are doing better than you.Then decide how to approach the problem.
Test improvements to your USP, user eye flow, form placement, visuals, headlines, testimonials, third-party validation, and so on.
With long pages, you’re essentially testing the addition of information, extra subheads that demonstrate value, more product visuals, and added elements like social proof or testimonials. By virtue of being longer, long content answers more buyer objections and includes more details that customers can use to justify their purchase.
So instead of testing page length, test those individual elements. Based on your research, build a hypothesis about adding social proof or explaining the most important feature. Either of these (and dozens more) tests will require a longer page.
Shorter sales pages, by their nature, are almost always simpler than long pages. They’re easier to read and digest. And focused on a single objective.
So rather than testing a “shorter” page, run exclusion tests to see what happens when you remove certain elements that may distract from your CTA. Simplify and clarify your message. Shorten headlines. Focus on a single conversion objective.