This post is actually about the Website Optimizer, the multi-variate testing tool for landing pages. I’ve taken a look at the conversion optimizer here.

Tell me about Omniture. Expensive, but seems worth it. We are in negotiations now at the company I work for since we really need a better solution. We have lots of sites, and it’s been a nightmare trying to look at analytics data for everything and put together funnels for conversion. Better analytics reporting could save me hours a week, so I’m really looking forward to it.

I may test it myself on one of my accounts. Currently I am managing the PPC efforts for a major newspaper organization with $100,000 monthly spend across 10 newspaper sites. Those sites range from $5,000 to $20,000 monthly. I also manage several smaller accounts (between $500 and $1,500 monthly spend). Which do you think would be better to test on?

I can’t really complain about my bid management software. I got in early with Omniture SearchCenter and negotiated a pretty good price (before they understood what bid-management was worth). It’s been working pretty well for me considering it doesn’t offer portfolio bidding. Still, it’s a little cost prohibitive for the smaller accounts.

I guess I was hoping there was something that would automate the process on these small accounts, which I handle 100% manually. Of course, manual means labor, and labor means lower profit margin. I want to continue to provide a great service for these small accounts, but there’s a point of diminishing return where you end up working for free …and I hate turning down someone who could really use the help.

I’m not sure about that. So far it’s a race with 42 and 60 going back and forth daily. I really think that I’ll never know, as I’m most likely to end this experiment early. It’s my first test with Website Optimizer (although I’ve done manual A/B testing for years). I like the platform, it’s easy to implement, and gives you a nice simple report.

Although it’s probably not enough for the kind of work that we want to do. Too simple, not enough data. And it’s not fractional. It tests every combination, so on pages with less traffic, it’s going to take forever to test more than one or two variables. Nice for the small business to get started with, but the professionals are going to leverage the more complete solutions.