Multivariate Testing - Too Much Information?




For most sites, yes. Contrary to what most gurus are saying today, for sites with under 1000 unique visitors per week, simple split testing will work better for you.


Why? The first is simple math. Without big, stable traffic numbers and a steady, measured conversion rate, your multivariate testing results are not much better than choosing a ‘winner’ sales page at random. Any freshman statistics student learns that for a statistical observation to be reliable, it needs to be derived from a sample size large enough to ensure reliability. If you are launching a new product or site with zero traffic to start - you are making a mistake if you are making copy writing decisions based on

Concealed Strategies Internet Marketing
...is the authority on what you offer and individuals will continuously visit. Not only should the information be up to date it should also be factual and not science fiction unless your site is a SCI-FI site. New Stories, Articles, ...
statistics based on just a few hundred visitors and a handful of sales. Keep in mind also, the more variables you test, the more observations you will need for a valid test.



The second reason I don’t like this approach is that the software used in multivariate testing puts your sales letter into ‘cookie cutter’ mode. You have to break your approach down into predictable chunks: “ok, here is the guarantee, here is the call to action”, etc. This flies in the face of good copy writing development technique.


When you are launching a new product - you need to test more than just variations on the same basic sales letter. You need a few different sales letters, that have

Internet Marketing Solutions To Promote Your Website
...and services and to get online business exposure. Various internet marketing strategies are also becoming popular these days. The main aim of internet marketing solutions is to attract more traffic to the business websites. Other is to improve the page ...
variations in headline, appeal, target audience and content order/format. You can’t be bound to what your ’software’ says you can do with your variations and ideas.


What to test? Start with the basics:


Headline - the most important part of your copy. The headline calls to the interested readers and pulls them in to find out more.


First paragraph - the second most important part of your copy. Now that you have the readers interest - the first paragraph is the gateway to enticing them to read on.


Appeal - what triggers are you using to get the reader engaged in your ad: the major appeals are:


Sex/Affection - including: Greed, Fear, Duty/Honor/Professionalism


Appearance/Photos -

Internet Marketing: Short-Term Solutions
...the long run. But, what about right now? What about the short-term? You might be wondering if you ll make it to the long-term portion of your entrepreneurial efforts. Any business startup is hard and you want to make sure ...
high-tech, glitzy, low-budget, minimalist. Try several approaches.


My advice: split test 3 or 4 sales letters first - in 30 sales or so, you’ll know what overall approach is working best - then you can work out the minutiae of what color, guarantee copy, etc. converts best.


Carl Ringwall is an internet consultant and programmer. http://www.datasystemsplus.net







Leave a Reply