Comments on: Developing a Site Testing Process: Site Optimization 101

By: Patrick Thompson

Patrick Thompson — Wed, 04 Aug 2010 16:30:04 +0000

I totally agree with Alvin that many people do not consider the content of their site and invest a large portion of their marketing budget to optimization when they have a product that just wont sell. No matter how great your site not many people are willing to buy widgets so do your market research and adjust your product accordingly before doing large scale redesigns.

By: Alvin Tan

Alvin Tan — Wed, 04 Aug 2010 03:24:08 +0000

I think that anyone who has performed testing would have sometimes encountered what I would call a cognitive dissonance: they see that what they strongly believed to be a superior site version/copy/UI/etc. had performed worse than what they strongly believed to be an inferior counterpart, even accounting for statistical significance. So yes, testing has its own shortcomings, and still requires a great deal of interpretation and analysis to be actionable.

By: Mark

Mark — Wed, 04 Aug 2010 03:07:55 +0000

What you have to realize is that often these tests will uncover more questions than they answer.

So inevitably you have to carry out more tests.

You also have to apply some “marketers intuition” to the results.

Often the results will “smell a funny colour” and you’ll have to figure out why.

Of course this type of talk drives the left brained, logical types to distraction.

But you really do need a creative perspective to see what’s going on.

numbers will only get you so far.

By: Salman

Salman — Thu, 29 Jul 2010 16:38:21 +0000

Analyzing the sales funnel is more of an art than a science, but ofcourse you have to use all the quantitative measures you can to constantly tweak. One point to remember is not to change too much as figuring out exactly what change led to the increase in ROI becomes difficult.

By: Nick Stamoulis

Nick Stamoulis — Thu, 29 Jul 2010 12:26:16 +0000

I’m in agreement with the comment of perhaps it’s not your site (but instead could be what you are attempting to sell), but every site that is trying to achieve sales through conversion on their site, does need to dive into some SEO. If someone is passionate about what they are selling and wants to put the money and effort into an SEO campaign, in truth it should pan out. For every item people are selling, there are buyers out there in the online space, it’s a matter of reaching out to them.

By: Alvin Tan

Alvin Tan — Thu, 29 Jul 2010 07:24:46 +0000

Good post. I think that the issue that “It might not be your website” is more important than anything else and should be mentioned first to prevent people from plunging into site optimization.Generating hypotheses to test and exploring methods to test are important, but if the bottleneck is not a marketing one, all optimization efforts will be futile and a waste of resources.