Hi there! My name is Jarryd Fillmore and I am from Grimsby and St. Catharines. I’d like to talk to you today about what to A/B test on your website, as this is extremely important when wanting to not only rank for websites on search engines (such as Google, Bing, Yahoo etc), but also for easy to use navigation for your users. Because ultimately, your users will be your end customers!
When dealing with Clients, many clients believe they know what’s best for their website. They may absolutely know what’s best for their own personal business, in terms of their clientele (and I don’t doubt that!), but in terms of their website and how it should be presented (layout, colour scheme, grid structure, etc), that should definitely be left up to the web designer and web developer. Even the SEO guy (heh). The client’s job is to know their own business and we can definitely learn from them (it’s our job to make project goals together, etc), but it’s also our job to present our Client’s business to the absolute best of our ability.
When we build websites, us web designers and web developers aim for perfection. We want to make sure our Clients receive the absolute best in terms of website design and even website functionality. However, there are certain things that may drive users away from a website.
Something as paramount as the large hero image that we end up choosing for the client’s site, or even something as minor as the colour of the buttons we use on their landing page, can absolutely and drastically affect conversion rates. That’s why this is absolutely imperative to get it right the first time!
This is why we conduct A/B tests when making not only our own website, but our client’s website as well. Similar to things that scientists do (science experiments), you may compare the effects of two different objects, things or liquids on something like a plant. Which works better? Which has effect A, and which has effect B, and even C? This is exactly what A/B testing does. It filters out what works, what doesn’t work – and ultimately, what you will put live on your website.
By conducting these A/B tests, you gain paramount insight that allows you to make the correct tweaks for your users, without sacrificing things like userability and reputation.
A/B Testing Basics
These tests require project managers, web designers, web developers and office bosses to work closely with the marketing team to determine exactly which results are most effective. Some of the most common areas to do A/B testing are as follows:
- Colours. Making the smallest change of even changing the colour of the hero image can have an absolutely massive effect on the influence of overall conversions.
- Headlines. It’s been proven that having headlines that are more of a statement, rather than a question for example, are FAR more effective in converting users.
- Images. Think about it, people respond to images differently. If you’re on a kid’s website and you photos of a bunch of kids being all sad and what not, well guess what! You’re going to have a bad time. However, if you have pictures of children smiling, having fun, and even playing on the playground, you will literally feel that vibe yourself when browsing the website. The psychology behind images are extremely powerful.
- Navigation. Having a clear-cut, easy to use navigation is absolutely essential when creating a website. This is actually why creating a sitemap is extremely important, and should be done for every single web project that you do! Figuring out different methods of displaying the navigation (including the functionality), all the way down to the content is absolutely essential.
- Offers. Offers on websites play a massive a role in whether or not your users will convert to sales. This is where the fun begins. You need to figure out exactly how your audience is (this research should have been done WELL before you started even designing the site), because this allows for your web team to properly understand and implement the exact language that your customers speak. Speak the language they do, and guess what! Leads and sales everywhere!
- Amount of text. This one is a bit tricky. Now, it’s known that people will respond better to text, at least initially, when there’s bullet points involved. It’s a much more organized and clean way of putting together your content. However – if people are genuinely hooked on the content that you provide them, people will read through that long ass wall of text or paragraph you’re presenting to them. That’s why news sites are so popular – people actually read the articles, and they’re super long. But people are interested, so that’s why they read them.
Setting Up Your Tests
Now once you’ve figured out exactly what you want to test, you have a few options to move forward with. You can either set it up for random (what this means is that you have a few different A/B options, and that element, functionality or page will be selected at random), or simply have two groups of users being directed at two different landing pages or websites.
The most important thing that you can get out of this is you need to tie into a specific goal. You need to ask yourself, “What’s the point of Test A? What’s the point of Test B?”. Having a goal is absolutely essential because without this goal – what are you trying to accomplish?
It’s extremely important to realize that not every single A/B test that you conduct will have a clear-cut winner. It’s just not realistic. Sometimes the results are so close that you’re like “what’s the point?” – well, even if they are super close, and A/B test A has 51% conversion while A/B test b has 49%, you still take the former! This is obvious as even though it’s only two percent better, it still brings home more business in the long run.
Well that’s it! I hope you guys enjoyed why should A/B test 🙂 Signing off, Jarryd Fillmore – Grimsby, St. Catharines.