“Test, test, test.”
We hear this saying all of the time, almost becoming annoying.
I’m here to tell you, I know of a little something that’s a it MORE annoying.
Ready? *drum roll*
Losing money by NOT testing.
One of my favorite quotes from David Ogilvy is:
“Never stop testing, and your advertising will never stop improving.”
Believe me I know it’s an extra step in your marketing funnel, but a very necessary one.
So let’s dive in what we SHOULD test and then I’ll go on to give some real examples of tests I’ve done.
Understanding What To Test
Before you decide what to test you really have to define your goal. What is it that you want to improve and measure? Is it more leads? Is it more sales?
One of the elements I like to test right at the beginning is the Call to Actions (C.T.A.). Testing button text will improve conversions no doubt, and is easy to set up and measure. Once someone clicks on your C.T.A. they will then end up on another page goal URL. Once someone lands on your goal URL, you’ll be able to start measuring conversion rates.
Here are examples of button text I like to test:
- Add to cart
- Add to bag
- Add to basket
- Buy Now
- Order Now
- Yes, I Want To (benefit) Now!
Button shapes can have a big impact too. Believe it or not the shape of your button (rectangle, irregular, representational, oval…etc.) can all have an affect on your conversion rate.
Also color. A red button has proven in many tests to have an advantage over many colors when it comes to generating leads. It creates a sense of urgency. Contrast makes a big difference. If you have a blue background try an orange button. Not only does it compliment the design, but it stands out.
Another obvious test is headlines. Remember your headline is the first thing people will read when they visit your site and it must stick out. Your main benefit of what you’re selling should be in your headline. Always keep in mind, when someone lands on your site, they immediately ask themselves: “What’s In It For Me?” (WIIFM) so to quickly answer this for them with your headline is critical.
Testing short vs. long headlines is what I do, and I’ve found that short, simple, sweet headlines tend to do well (about 8-10 words). Make sure your message is so clear, a 5th grader can understand it.
Other elements you can test could be adding a video sales letter vs. none. You can also re-arrange different elements. Adding security badges, images, price points, testimonials, countdown timers…etc.
It doesn’t have to stop there either. Beyond testing images, copy, and other elements on your pages, you should also be testing your funnel as a whole.
This means upsells, exit redirects, cross sells, and downsells. By re-arranging your funnel you may find that be switching around the order may increase conversions by a lot. Maybe you’re selling a front end offer at $7. The upsell you intended was $27, followed by another upsell of $97 (I like 7’s apparently). Testing could be you switch the $27 offer with the $97 because maybe that’s what people really want and the $27 is just getting in their way. Sure it’s cheaper, but it may not be as valuable to them.
Real Examples Of Testing I’ve Done
To do any of this sort of testing, you’ll need a split testing software. I choose Visual Website Optimizer (VWO). I like it because it’s easy to set up, and even gives additional testing ideas for you.
In this test I wanted to test 2 completely different designs, including different copy.
The end result ended up being 12.4% improvement rate:
This resulted in 44.41% conversion rate and I gained 81 brand new leads. Now you have to ask yourself, what is a lead worth to your business? In this case, a lead is worth a lot.
This is why it’s so important to always be testing because if I would have been too lazy, I would have lost out on 81 leads which could end up being converted into a customer(s).
Here’s another test I did where I ended up just testing 3 headlines (including the control):
Even though we didn’t have a ton of traffic, variation 2 was declared a winner with a 75.00% conversion rate. The result all from a headline change. This was a lead generation campaign so I was able to draw the conclusion on such a small amount of traffic. I would recommend at least 500 visitors for a sales campaign.
Let The Market Decide.
The key is, is to always be testing because you never know exactly what you’re going to find that works. The market will decide that : ) Coming up with your hypothesis of what will win and what will not is important, but testing it is where you’ll find the true results.
I will definitely keep more of my testing experiments coming and will give you more ideas in the next upcoming blog posts.
In the meantime….
I would love to hear your testing results in the comments below : )