When you pay to bring traffic to your website, you’re taking a big gamble. But if you don’t have good landing pages, it’s a near certainty your bet’s not going to pay off.
Along the way he’s learned a ton about what makes landing pages work. Do your landing pages do everything they should? Do they do too much? From knowing your audience to knowing when to A/B test, here’s Oli’s expert opinion on what marketers can be doing to make sure their paid search efforts hit the jackpot.
EB: Thanks for joining us, Oli! In your profile on the Unbounce website, it says your “disdain for marketers that send campaign traffic to a homepage is legendary.” What’s so wrong with homepages? And are there any good exceptions to this rule?
OG: Here’s the skinny. It seems that a lot of marketers are spending their time focusing on getting “the click.” They want a high Click Through Rate (CTR) on their ads or their emails, and then they break the experience by taking what was probably a very focused ad by sending the traffic it to a generic homepage.
To be clear, what I’m talking about here are marketing *campaigns*. Every marketing campaign has a single goal, to get someone to: start a free trial, download an ebook, request a callback, register for a webinar, subscribe to your blog, etc.
When you send someone to your homepage (or other internal website page), you are placing them in an environment designed for exploration and research. This is a completely suboptimal environment for conversion, and it’s underscored by the difference between User Centered Design (UCD) and Conversion Centered Design (CCD). The former is the practice of designing navigational, wayfinding, exploratory, and transactional experiences that are as simple as possible to complete. Conversely, CCD is about removing any and all distractions to make the experience laser focused on the campaign objective.
The first principle to note here is what I call Attention Ratio, which is the ratio of things you can do on a given page to the number of things you should be doing. For a marketing campaign, this equates to the number of links on your homepage (often 40-60), to the number of campaign conversion goals (which is always one). As Attention Ratio goes down (moving closer to 1:1) conversion rates go up:
A typical home page with 40-60 links:
A conversion-oriented landing page with just one call to action:
Another reason why a landing page dedicated to your campaign is effective is that you can easily match the ad copy to the headline of the page. This is called Message Match.
A strong Message Match is critical in communicating quickly to a visitor that they are in the right place – and made “a good click.” Anyone running campaigns (email, social, PPC, display) knows that there are many many variations of ad message used. Your homepage simply can’t adequately communicate to someone in a way that’s as strongly coupled as a page designed just for this purpose.
The only time when sending your campaign traffic to your homepage is acceptable is when you’re looking for low conversion rates. (Click to tweet.)
Joking aside, it’s an acceptable practice to send someone to your homepage when you are bidding on branded search terms (your product/service name). The reason being that you have zero context for their intent (that would normally be derived based on other keywords in the search terms – or the content of an email promo), so you should give them a generic experience.
EB: If you’ve made the effort to create more focused landing pages, what are a few telltale signs that your pages are either poorly designed or underperforming? Conversely, when can you pat yourself on the back and say a given landing page is pretty awesome?
OG: It can be incredibly hard to notice when you’ve created a poor landing experience because you’re too close to your own story. Rather than just saying “Your conversion rates will show how bad it is,” I would suggest performing a simple exercise that I’ve dubbed the BS detector.
The goal of the exercise is to uncover any issues with clarity—which is arguably the most important aspect of conversion.
First you assume the role of an impatient web visitor. Don’t know how? It’s easy. You quickly scan the page for the dominant elements that stand out from the design. This will include headlines, subheads, large images etc. Got it? Good. Now try these steps:
- Print out your landing page.
- Have a friend or colleague sit within earshot.
- Walk around the room in circles which scanning the page in your hands and reading aloud the main elements. The walking part is helpful to simulate distraction.
If anything you read out is confusing or doesn’t actually contribute to the communication of your campaign goal, then it needs to be re-written.
Attention-driven design is the act of using visual design techniques (hierarchy, dominance, encapsulation, whitespace, contrast) to make the reading experience as simple and fluid as possible. By doing this you illuminate the failings in your copywriting. Which is exactly the point of this exercise.
An effective landing page uses visual design to focus your attention on what’s important, and clarity of communication to get your message across quickly and easily.
EB: So when can qualitative research around things like page designs or messaging be useful?
OG: You don’t know what you don’t know. Which is great to know, so that you know you should be uncovering some of what you don’t know. But, at the same time, when you don’t know what it is that you don’t know, it can be difficult to know how to go about finding out those things. Know what I mean?
Qualitative insights are cumulative. By that I mean that once you start asking questions, the answers inevitably lead to more questions. So the secret to learning from your visitors and customers is just to start asking questions. Sometimes a broad question can help you find a way to the narrow, more poignant question which ultimately unearths the gold.
Take the landing page templates page on Unbounce.com, for example. I used a pop up survey to pose a very simple question to visitors: “What do you think of our templates?” After looking through 1,700 responses, I saw three primary recurring questions. “How much are they?”, “Where can I download them?”, and “Can I use them in WordPress?” You can’t do any of those things, so there was a lack of context on the page detailing how they could be used (inside the Unbounce landing page builder).
To answer this question we designed the graphic you at the top of the page (pictured below), which shows the templates inside the template library, and a template being edited inside the page builder:
This is an example of “context of use”, where you demonstrate visually (image or video) how the object in question will be actually be used in practice.
After A/B testing this new solution we saw an increase of 43% in new trial signups for people who viewed this page. That’s an incredible impact that came as a result of uncovering something we didn’t know we didn’t know.
EB: When you begin to plan a landing page test, when should you test new page designs, and when should you optimize by tinkering with smaller design or messaging elements?
OG: Testing a completely new design is typically reserved for trying to achieve the global maxima vs. small iterative tests that serve to reach the local maxima. (For an in depth thread on maxima/minima read “Don’t fall into the trap of A/B testing minutiae” by Rand Fishkin, and “In defense of A/B testing” which is a follow up from Paras Chopra – they are a few years old but still hold true).
In simple terms, small changes are unlikely to result in big results, but they are also much easier to implement, and for the most part less risky.
There are a couple of questions to consider when making a decision on how big you should go with your tests. First, how much do you actually know about your (potential) customers? Have you done any research? Do you think a high-level messaging change (headline/CTA) is most likely the answer to your conversion issues (based on having uncovered confusion through research)? What does the known data say (is the page converting really well already)?
We test our webinar registration pages every month. The inbound source is primarily email marketing, with the visitors being people on our “Conversion learning list”. Because they are well qualified, our conversion rates sit at around 70%, and no matter how much we play with the call to action, form length, small design changes – we’re unable to move the needle. This is a signal that we’ve hit a local maximum, and need to step back and take another look at the bigger picture, then reimagine a completely new design. We’ve done this once so far and it’s not yet had an impact, so it’ll be back to the drawing board.
This also raises an important question about test priority. There could be much higher gains available to us on pages that don’t convert so well, but I’m primarily interested in developing a solid theoretical solution that is repeatable in other circumstances. Stay tuned for more on that.
EB: Is there a pre-flight checklist you recommend to make sure you’re launching a landing page with a high chance for success?
OG: I like to run through my Conversion Centered Design checklist before pushing a new landing page design live.
Conversion Centered Design Checklist:
- Does your page have the 5 essential elements: Value proposition, hero shot, benefits, social proof and a Call-To-Action (CTA)?
- Is your CTA green or red or orange or blue? Yes, that’s a sarcastic trick question. The real question is about contrast. Does your CTA have a unique and contrasting color to the rest of the page?
- Are you using directional cues to point at your CTA?
- Is your lead gen form encapsulated in a bounding box that separates it from the rest of the page?
- Read your headline and subhead out loud. Does it describe what you will get by participating in the page’s goal?
- Do you have some areas of page content broken down into bullets?
- Does your form have a headline that describes what you will get for filling it in?
- If there is a photo of a person, are they looking either directly at you or at the CTA?
- Do your social proof elements talk directly about what you are offering on the page?
- Is there lots of white space to provide clarity and an enjoyable reading experience?
- Is there a way of demonstrating a preview of your offering? (Not always applicable). Previews can include: ebook previews in the form of a short PDF, HTML chapter, slideshare embed, or demo video.
- Information hierarchy: Is your eye naturally drawn to the headline first when you look at the page?
- If you are doing PPC, does the headline closely match the copy in your ad headline?
- If you’re doing display or Facebook ads, is the image in the ad repeated on the landing page?
- Are zero negative words (e.g. spam, gimmicks) placed close to your CTA? (Even if they are written in a positive way). If you did the congruence test you’ll have already fixed this.
- Framing: if you are presenting numerical data on your page, it should be numbers for good news, and percentages if bad news.
- If you have a form, do you ask for a second conversion (tweet, follow, subscribe) on the confirmation page?
- Do you split your value proposition into a headline and subhead? A subhead can be a good way to add clarity to your headline.
EB: Thanks for your perspective, Oli. Any parting thoughts or valuable resources for marketers who want to perfect their landing pages?
My pleasure Elliott!
Here are 3 things you should do next:
- Remember this: Never start a marketing campaign without a dedicated landing page. Otherwise known as the NSAMCWADLP principle.
- Read this post I wrote for MOZ: The Most Entertaining Guide To Landing Page Optimization you’ll Ever Read
- Take this free course: The Smart Marketer’s Landing Page Conversion Course