…not all at once, anyway.
When it comes to internet marketing, testing is the name of the game. Testing is an invaluable tool, whether you’re testing a new link building strategy, PPC landing page or eMail marketing campaign. In order to better hone your craft & improve your results you need to find out what works and what doesn’t.
The problem is, some people get so excited at the thought of testing that they try and test everything all at once.
- Two totally different email templates.
- Two totally different landing pages.
- Two totally different sets of keywords.
When the results start to come back, the first thing most people do is look for the clear ‘winner’. The problem is – do you really have any idea what caused test A to outperform test B?
When you test everything (all at once) you’re essentially testing nothing at all.
If you use two totally different designs that bare almost no similarities it will be next to impossible to determine which elements caused the win.
When testing, it’s always a good idea to focus on specific elements, leaving the rest of the marketing piece identical (or close). Consider it your ‘control group’ of sorts. This way, when the results come back in, you’ll have a much better understanding of what actually caused the results that you’re now reviewing.
For example, let’s say you’re currently running an email campaign. Let’s also assume that your open rates look solid from mailing to mailing. Your click through rates, on the other hand, aren’t looking so hot. Since open rates are good – we can assume that our subject lines are doing the job and getting your audience to at least open your message. The problem is, once they get there, they’re ignoring the call to action that you’ve put in front of them. In this case, reviewing your templates is a good place to start. That being said, don’t go out and test your current template against a completely different template. Instead, focus on a single element at a time. Are all of your links currently simple text links? Try replacing them with larger, more interesting buttons. If your business model allows for it, test various dollar amounts against precentages when dealing with offers & sales. Although the actual discount may be the same, the perceived value may be greater one way or the other.
Test image placement, headline placement, the location of your CTAs – test everything! But only one at time, please….
Whenever a test yields a clear winner – implement the winner into your next campaign and start testing a new element.
Test. Implement. Improve.