User experience is becoming an increasingly important success factor.
It’s no longer enough just to have a website, or to just have a website that is well-designed, or even copy that reads well and delivers your message clearly. In the world of modern digital marketing, it is a synergy of all of the above and much, much more that will help you distinguish yourself in a crowd of similar websites and inspire your visitors not only to check you out once but to keep coming back.
In this post, we will be looking at how you can improve user experience of your website by using Facebook ads.
Before we actually get to the meat of this article (the main course, if you will), we need to address a brief point first (let’s think of it as our appetizer).
There is a principle marketers like to quote often, stating: 80% of your results will come from 20% of your effort (not to be understood as “only ever give 20% to achieve great results”).
This is, of course, not written in stone, but it is a great way to remind yourself that you need to identify and invest in your top-performing everything (content, marketing channels, ads, copy, landing pages, and so on), as opposed to spreading your effort and resources around equally.
Marketers sometimes forget that boosting what is already performing great is one of the best ways to boost results (be they clicks or conversions) – as opposed to building up an entirely new resource or campaign from scratch.
Paid ads are (apart from a way to drive traffic and sales and boost brand recognition) an incredible way to gather data. They are a great analytical asset if you choose to use them the right way.
As you can create two identical Facebook ads, directing traffic to two (or more) variations of the same landing page, they will allow you to test out all kinds of different elements of the page to improve it.
Not to mention that testing based on organic traffic will take time, and you will not be getting as relevant data sets – as you can never actually be in control of organic traffic as much as you can be of your paid ads. Yes, you could create different landing pages and promote both of them organically – but your data is likely to be skewed.
Now that we understand the principles behind the process, we need to cover one more important element before we dive into the actual how-to part of this article. We’re talking about the difference between split testing and A/B testing.
While the terms are sometimes used interchangeably, they are actually not the same.
A/B testing is when you change just one element of a landing page and compare the two different versions. For example, you can change the CTA and determine which one gets more conversions. The key is in showing your variations (your A and your B) to the same audience, so as to get the best results.
Split testing is quite similar, but it mostly focuses on design elements, and it can contain more than one change per test. For example, you can go for two completely different design variations and test to see which will work better for which kind of audience.
To perform your test successfully, you need to determine your target audience well (i.e., not test a landing page on an audience that is not interested in the first place), choose the element you want to test (more on that in a minute), and make sure your variations are clear.
You also want to collect your data so that you can actually make sense of it.
Digital marketing is all about collecting the right data. Without it, you will just be doing some very blind stabbing in the dark – true, you can hit the bull’s eye, but your odds will be rather slim.
One of the best tools you can use to collect your A/B and split testing data is Google Analytics, so making sure it is set up, optimized, and running as you want it to run is key.
Some of the data you want to track includes:
- Where your visitors are coming from
- How much time they are spending on each page
- What they are clicking on the most
- What percentages of your visitors are converting
- Where their attention is mostly spent (you can track this with the use of a heatmap)
You also want to ensure that the pages you are testing are at their very best. Apply the 80/20 principle here as well, and remove all elements that are performing poorly.
Consider your main goal for this landing page, and analyze all of its elements with that goal in mind. Read your copy carefully, change what needs tweaking, think about your supporting visuals and the formatting. Remove any CTAs that are not in line with the goal, or swap them with others that you think will work better.
Once you have a page you’re satisfied with, you can move on to the testing portion of the program.
Again, bear in mind what we discussed about split and A/B testing – and ideally, run a split test first if you are not sure about the design, and then move on to A/B testing.
Start by changing the one element you think will bring in the best results. Create two distinct landing pages, one being your page as is, and the other an identical page, but with the one tweak implemented.
Then run your Facebook ads – one for each page, ideally targeting the exact same audience (at different times of the day, on different days, etc.).
Run your campaign for as long as you need to – i.e., until you have enough data at your disposal to make educated decisions about the winning page.
And then just repeat the same process with the next tweak.
You want to test more than one element, until you come up with a version of the page that is performing at its absolute best, targeting your ideal audience and converting as well as it can. You also want to test more than two versions of the same element.
(And you can then start working on that organic traffic as well, knowing you are now prepared to adequately greet and convert it.)
Some elements will be easier to test than others, and testing some will be more important than testing others. Here are the four main elements you should definitely test on your most important landing pages (because remember, 80/20):
It’s certainly true that the visual elements of your website will have a more immediate, and most likely more powerful impact than text. However, your headlines and all of your other copy still play a vital role in converting, building rapport, and helping you stick in the minds of your visitors.
Headlines are your most important copy-focused page element, so start by A/B testing them. You can also test out different variations for different audiences.
Start with something more general, try different variations of power words, see what happens when you focus on different aspects of your service or product, and so on.
As for your copy, it might be a bit more of a challenge to test, as you can never be sure people are actually reading your words. The metric you want to keep an eye on here is time on page, and heatmaps can also help you determine how much of your copy has actually been read.
You can try switching your voice, your point of view, the language and rapport you use – but do make sure that the changes to your copy are palpable. Don’t just reword a bit of text which is essentially still the same and test right away.
Always test your headlines first, and leave the copy for later.
Here’s an example to illustrate our point. ZenMaid has the main headline at the top of their page that is dynamic and keeps changing. While this is certainly a great way to grab attention and to keep visitors engaged, they could run a test to see if this dynamic version works best, or if one of the versions being displayed might work better. Who knows, the constant movement at the top of the page might be distracting or annoying to some of their target audience.
image source: zenmaid.com
CTAs are one of the most important elements of every landing page. They inspire action, boost conversions, and speak to your audience more clearly and more loudly than your actual copy.
In fact, marketers have gone so far as to test out different CTA colors for different target audiences.
Some of the things you can test in terms of CTAs include wording, placement on the actual page, color, button size, button shape, font, and background.
Make sure to test out more than one variant here. Start with a base design and test it against a variation. If the variation proves to be more popular, come up with another variation and test it against the first variation. Keep going until you have exhausted all of your most popular ideas, and until you have tested all of the different elements of a CTA.
And remember, you are only changing ONE thing – so don’t use one color, wording, and font on one page, and a different font, color, and wording on another. Change these elements one by one.
Here is another example: Filecamp has a CTA at the very top of their page, yet it might be a bit on the small side and could do with being larger. The fact that it is green clashes nicely with the blue of the background, but maybe something white would work better?
image source: filecamp.com
If your landing page has no visuals, this might be the first thing you want to change about it. Visuals are simply a must, and they also must serve to provide more information about the product or service you are promoting.
You can use tutorials and how-to videos, you can show screenshots or photos of your product, but make sure you do have some sort of visual on the page. If you have no other option, go for an illustration at the least.
Knowing which images, in what size and in what color will work, and where exactly to place them is what your A/B testing should uncover.
For example, InFlow could do some testing to determine whether their images are too large, driving the right message home, being instructive enough, or whether a different variation on the same basic idea could possibly boost their conversions.
image source: inflowinventory.com
Finally, you want to look at the way your main elements are spread across the page and how they communicate with each other. Is there a clear story to be followed? Do the elements blend into one another seamlessly? Is there actual cohesion, or does your page look like it has been cut and pasted together by a somewhat talented five-year-old?
You should also test the formatting of your copy: bullet points, different fonts, the use of italics and bold, font sizes, alignment – even the smallest tweaks can prove to have surprising results sometimes.
For example, a site like Woven can do a lot of testing on their homepage: colors, the size of each element, font size and image placement, the use of negative space. If your page is reminiscent of this one, you can go to town with your UX tweaks.
image source: woven.com
Testing your pages is just another step in crafting, writing, and publishing them. Just because a page is live, it doesn’t mean it’s out of your hands.
With the clever use of Facebook ads, you can tap into a whole lot of UX-related data and improve your pages immensely – ensuring that your best assets are really doing their best.