Case Study For Writing a Great Home Page
- Applied and tested PAS framework on the home page copy
- Produced a paid lift of 49%, with 99% confidence among female users
- Produced a paid lift of 46%, with 99% confidence among male users
Optimization is an iterative process.
Your copy is never “done.”
So although we more than doubled SweatBlock.com’s revenue by applying these 3 copywriting principles – messaging hierarchy, specificity and appropriate CTAs – we weren’t about to stop there.
This spring, we worked with the SweatBlock team to run another copy test on their home page.
To Figure Out What to Test, We First Had to Do the Unsexy:
Develop an Hypothesis About What Wasn’t Working
When you get a statistically significant paid lift of 108%, you might tell yourself it’ll be tough to beat that winning copy.
And you’re right: it probably won’t be easy.
But it’s never “easy” to beat any copy. (Anyone who says “this’ll be easy” is trying to sell you something. Slowly back away…)
And of course the goal is not to beat the copy. …This is the part where I robotically repeat what we all say and hear at conferences: the purpose of testing is to learn. I mention that grudgingly not because it’s not true but because I get sooo tired of people scaring people off testing with endless rules and guidelines. (Aren’t we a little early in split-testing for so many bloody rules??)
Back to the point: We knew some things about what moved SweatBlock prospects. We needed to know more. Hence, test.
By the time we got to the point of creating a new variation for SweatBlock, we:
- Knew the product well, having used it
- Knew the audience well, having watched dozens of testimonial videos and read through 1000s of their Amazon reviews
- Had a strong sense for what was working in the Control, given that we’d written it thoughtfully and intentionally
With so much more to learn, we needed to start by poking holes in the winner.
Picking apart other people’s work is always a fun exercise. Hence these fantastic comments on our earlier SweatBlock post.
So why not give it a try now? Go through the Control copy that follows immediately below, and think of what you’d do to improve it to the end that more people buy more SweatBlock:
Here is a fraction of the list we came up with:
Woman in hero is facing away so we can’t see her dry underarms – perhaps better to see her facing forward to see a grey shirt that’s free of underarm wetness.
Should the hero woman be in a crowd, where she’s more likely to feel “sweat shame”, rather than alone?
Should we get rid of the global nav?
Maybe people want more than to “control” their sweat.
Does the SweatBlock product shot come too soon?
Does the free shipping offer come too soon?
Is the call to value in the hero too value-y and not action-y enough?
Do the media logos need an introduction?
Do we need the media logos at all? What are they there to do? (We didn’t place these thoughtfully in the winner.)
Should we try to get our prospects to feel more of the pain of sweating before we intro the solution?
Does the product “demo” (i.e., 4-step nighttime application) appear too soon for prospects?
The 7-day calendar – is it believable yet? We show 7 days, but it’s actually 6.4 days.
Many SB customers have tried countless solutions and may be so frustrated they’re considering Botox. Should we address this or compare the ease and affordability of SB to the drastic measure that is a Botox injection?
Perhaps the photos showing the various types of sweating aren’t big enough?
- Maybe we should just cut that whole area with the 5 stacked photos and sweat types?
Perhaps those photos are too stock photo-y?
Are we minimizing the authority and influencer proof (i.e., doctor, Rachel Ray) too much?
By putting FDA compliant in the crosshead on the close, are we introducing concerns? Do we have reason to believe that prospects need to know about FDA compliance with such emphasis at this point?
If you get free shipping on 2 boxes, should we show 2 boxes in the close instead of 1?
Is there a better way to integrate the FAQs so they’re not so tacked on?
Where are all the testimonials???
Should we remove the CTAs in the footer?
- Should we add in new messages? – countless Amazon reviews raved about using SweatBlock on hands, necks, backs, knees.
For me, a lot of those questions and ideas were… smaller. It felt like we could and should address some of them, yeah. But, well, really, they were just elements. They spoke to execution rather than something bigger and possibly more interesting. Treating them would be treating the symptoms, not the disease.
Bigger research questions – based on bigger lessons to learn – might look more like:
- Do people landing here feel alone? Will a greater emphasis on social, authority and/or influencer proof better speak to our prospects and compel them to buy?
This would likely drive us to create a new variation that leads with and leans heavily on testimonials and shows of support from authorities.
- Are non-readers struggling to see themselves on the page? Will more solution-focused or happy-outcome photos better speak to scanning prospects and compel them to buy?
This would likely drive us to create a new variation that uses 1) more photos that 2) showcase wet vs dry underarms in a range of social situations.
- Are prospects struggling to understand how SweatBlock will work in their lives and, thus, will more story-focused social proof and demos help convince them to buy?
This would likely drive us to create a new variation based around a SweatBlock demo, a la infomercials and traditional consumer-goods long-form sales pages.
- Should we treat this more like the great “miracle cures” of past direct response excellence? Are we jumping too quickly to the solution? Will spending more time on the problem before presenting the solution pull more prospects in so we can convince them to buy?
This would likely drive us to create a new variation that uses the PAS framework, keeping most of the “solution” part of the page the same but leading with the problem and then agitating it.
Now comes the part where we separate the conversion copywriters from the conversion consultants.
(My favorite part.)
We decided to test the thing that would make our copy-geek imaginations light up. Because if you were a copy geek and could choose between indulging your copy geekery or not, what would you do?
Now before I get into what we did, you may recall that, at the end of the first SweatBlock post this week, I invited you to guess what we might have done to optimize our winning copy.
A big congratulations goes to Andre, who totally nailed it with his guess about what we’d test for this second round:
We Decided to Test a New Variation That Used the “PAS” Framework
The PAS framework is intriguing to moi because it’s an old-school copywriting formula / framework.
It’d be interesting to test on any page. But particularly on home pages. Because it’s not designed for home pages. …But that doesn’t mean it can’t work for ’em, right? Right!
Now combine the PAS framework with my frustration with home pages, which I mentioned in one of my recent posts.
Part of my problem with home pages is that they welcome so many different types of prospects – people in all stages of awareness, with varying degrees of intent – that they have to try to be everything to everyone. Which dilutes the message. Which makes the page nuthin’ to no one. (I know that’s a double negative. Work with me.) My general solve for this is to write for the 20 to 35% of the home page visitors that you actually stand a chance of converting. And that’s a good approach. But it makes for rather loose and generally ho-hum “I’m everything to everyone!” copy.
What I love is landing pages.
Landing pages follow lovely little rules.
Landing pages are meant for targeted audiences in a somewhat controlled flow, making them dramatically easier to write than home pages. And, BTW, it’s not like I’m not up for the challenge of writing a difficult page. I totally am. What I’m not up for is the moving target that is a home page. “Now it’s for you. Now it’s for you. Wait, now it’s for you. Forget personas and targets – home pages for everyone!”
Landing pages, on the other hand, benefit from lovely little layout frameworks like AIDA. And like my personal fave: PAS.
To use PAS when writing your copy, you simply start with the problem. Then poke at that problem so your reader can’t help but feel it. Then, when they’re itching for relief, present the solution.
Now let’s say you wanted to use PAS on your home page. That would require leading with the problem and then agitating it. (The latter is the hard part.) And that would mean you can’t do what we usually do on home pages: lead with your solution. Almost every home page you stumble across (except for large ecommerce and large publishers) leads with the product or solution, often phrased as a core piece of the value proposition. Take a look at the hero copy of a random selection…
Value props, all around! We’ve seen that leading with your value prop on a home page can work wonders, as we wrote about here. We recommend to our students that they focus their home page hero on a strong value prop. So it’s not like it’s a bad practice to lead with a value prop. Not at all. It’s good!
It just makes leading with something else… scarier.
Now, wouldn’t you just know it – when I was looking for screenshots to use as examples, I went to Basecamp.com. And there I saw the closest thing to PAS I’ve ever seen on a home page:
PAS on the Basecamp 3 home page. And they use an open letter on the page, too. Because of course they do.
The team at Basecamp has always represented to me the ultimate marketing team. As long as I’ve been reading their stuff – and it’s gotta be a solid 8 years now – they’ve promoted the idea that everyone on a product marketing team is a copywriter. If you work at Basecamp, you’re involved in copywriting. So little wonder they’re testing or they’ve arrived at the PAS framework for their home page.
Now let’s return to the home page in question: SweatBlock.com.
With a pain-aware reader in mind, I wrote the new top 800 or so pixels of the page using PAS.
Make No Mistake: Writing a PAS Home Page Made Me Uncomfortable
It seems easy enough to present a problem and agitate it, doesn’t it?
Until you sit down to do it.
What kept getting in my way was not just the possibility that the right prospects to convert might not be further up the ladder than Problem Aware, which PAS is ideal for. To convert Problem Aware prospects, you need to move them from Problem Aware to Solution Aware to Product Aware so they’ll buy. And I was gonna try to do that on the page. Not a tiny feat.
…But that wasn’t the problem. After all, the last round of tests we’d done on the home page had also tried to move prospects through multiple stages on a single page, but that version hadn’t even had the benefit of a framework. So using PAS to frame such a page made me feel more confident than when I’d used nothing at all.
What was getting in my way was the force of every single home page convention that PAS was forcing me to ignore. For the past 20 years, we’ve been carving into stone home page rules like:
- Start with a statement of what the product is and does (which we’ve been hearing since at least 2002)
- Emphasize your differentiator in seconds to reduce bounce (since 2003 or earlier)
- Get to the point (discussed here)
- Start with the most important point (as here)
- Your headline should say what the product is, say what the prospect gets or say what they can do with it (as here)
Imagine each of those conventions had a voice.
And they were all shouting at me as I tried using PAS.
It was extraordinarily painful to push their voices away and just focus on the prospect’s problem: sweating excessively. Please take this as the warning that it is. Should you decide that you, too, would like to test a PAS home page, know that you will struggle to quiet ye olde voices – the Jakob Neilsens, the Bryan Eisenbergs – but that you must push through all of that.
At least, that’s what I did. And here’s how that turned out.
And In This Corner: Our PAS Home Page
Lemme show you the wire I sent to Chase at SweatBlock, and then I’ll tell you wassup with it:
We led with a pain we’d heard a lot: excessive sweaters are sweating for all sorts of reasons. It’s almost never ‘cos it’s hot out.
We agitated that pain by reminding them that sweating isn’t just about ruined clothing. Excessive sweating makes you feel trapped. It can’t be solved with regular deodorant – that’s why you find yourself avoiding light-coloured clothing, layering your shirts, tucking tissues inside your shirt – the list goes on. (It actually goes on for 8 bullet points.)
We continued agitating with the true story of Brianna, a woman who struggled with sweating in social situations. Her problem culminated in an embarrassing event at her child’s school, following which she went home and Googled “excessive sweating.”
The perfect segue for the page.
At this point, we introduced the solution. This just so happened to be the same point at which Brianna found the solution: SweatBlock. In fact, Brianna’s whole testimonial follows the PAS formula – it’s almost like the page is a large extension of her story.
From there, we started to bring in much of the page that had already won. Here’s what changed:
- We revised the hero to more seamlessly connect the new PAS top with the rest of the page.
- We added more proof points, like 4 million towelettes sold.
- We removed the 4-step “demo” because I wasn’t entirely sure prospects needed to see that in order to buy. It may be worth testing the addition of it later.
- We changed the use case area (i.e., where we wrote about nervous sweating) to a 4-column area. This area didn’t make it into the final version we tested.
The SweatBlock team was keen on the test, God bless ’em.
But they weren’t sure if the hero image should be of a man or a woman.
So they ran this as a three-way split test (i.e., A/B/C), where these variations were tested against each other:
This is the easier-to-read closeup view of the PAS addition:
So, do you think this PAS addition has what it takes to beat the Control, which already proved itself to be a high-converter?
Could either Variation B (woman) or Variation C (man) outperform the Control?
If so, which one? Or could both do the trick?
Did our copywriter-geek gamble pay off?
Do We Have a New Winner?
The Results of Our Home Page Copy Split-Test
Indeed, we have a winner! Two winners, actually.
Both variations brought in more paid conversions than the Control.
Here are the results:
- Variation B (PAS, woman) produced a paid lift of 49%, with 99% confidence
- Variation C (PAS, man) produced a paid lift of 46%, with 99% confidence
So, at least for the SweatBlock business and audience, the PAS framework worked on the home page.
I’m particularly excited about these results because they suggest that a home page doesn’t have to be an absolute guessing game or designed to speak to a broad audience. If you have reason to believe that the people you can convert on a page are in X stage of awareness – even if they’re landing on a catch-all home page or one-pager – you should test messaging for X stage of awareness, not for all the stages.
You should give it a shot with your next test.