Here’s a question I got by email a few days ago.
I read your book Convert! and after had a question.
When I re-design my website to implement and test conversion, should I remove old content and test out the new content that was on the site or should I keep the old content and add new one?
Andrae Palmer, Remla Productions
There is no single answer to this question, but I can offer some pointers.
In Convert!, I talk about the old and new models of web design. In the old model, we use the “first best guess” approach. The dominant personality (client, designer, or manager) tends to get their ideas implemented, butÂ — if only one untestedÂ design is publishedÂ — it’s still someone’s first best guess.
The new model is to test the design, concentrating on broad strokes first. In brief, the right approach is to split-test all assumptions, starting with the stuff that matters most.
What matters most?
Well, it isn’t layout, button colours, typography, or any of those factors that can impact how nice a design looks. Nice doesn’t sell.
Here are the factors that most influence people to take action…
- Seeing that the website is for me, i.e. it’s relevant. Specifically, if it talks to the need that most motivates me right now, I’ll pay attention.
- A strong proposition. Your proposition should make people sit up and take notice. They may be suspicious, sceptical, or intrigued, they may think, “Yeah, I’d like to see that!” But, again, it’s all about attention. Get attention, build it, and keep it until you can convince me that I absolutely must take up your offer.
- Evidence that tells me I can believe your proposition. This could be soft evidence: testimonials or quotes; or you may have hard evidence: facts and figures.
- A strong call to action. A call to action simply means telling people to act now! A credible reason why I must act now will always increase conversions.
I argue that these factors are design factors. Your copy is just as much part of designing your communication solution as your graphics.
I would advise anyone who wants to test a website first to ask,
“Have I included all the most important factors for conversion?”
If not, this can’t be your best guess!
So I wouldn’t test a website that demonstrates relevance, a strong proposition, evidence, and a call to action against a version that doesn’t.
What you should really be testing is which visitor need… which proposition… which headline… which call to action… is the most effective.