We’re pretty excited about the possibilities here and proud that two of our videos are the top-watched so far.

Check them out:

• Escaping the Dark Ages of Online Marketing with Tom Leung – We invited Tom Leung up to Vancouver to talk about GWO and we had to get overflow seating. What an event!
• Website Workout: Colonial Candle – A project we did for the Ecommerce store, Colonial Candle, where they saw an incredible 20% lift in purchase conversion rate!

Maybe we’ll start creating more Conversion Optimization Strategy videos. What do you think? Would that be helpful?

Here’s a screenshot of an experiment we have in progress with the new report features showng:

1. Remove unwanted combinations
You can now selectively remove variations or combinations from a Google Website Optimizer experiment that is in progress. This is a helpful feature I’ve been asking for from the beginning, but it’s also going to be dangerous for those inexperienced in testing and must be used carefully.

At some point with virtually every client, usually on the first experiment, we have to have a conversation about when to make decisions about ‘winning’ and ‘losing’ variations. It’s natural to try to make decisions early when you see the red and green bars begin to grow. It takes great discipline not to panic when one variation is tracking at a negative conversion rate and you can quickly calculate how much that could be costing the business.

But, just like a good investment advisor, this is where we have to coach our clients to be rational. You do not know which combination is going to win until we have achieved statistical significance. I’ve lost count of how many experiments completely reverse results before achieving a surprising finish.

That is why I also think the Google Website Optimizer team was wise to roll out the second feature at the same time.

2. Enhanced reports
The red and green bars have been replaced with yellow when the confidence level is low. It’s amazing how much colors influence decision-making (which we’ve proven through many conversion optimization tests) and this change will help calm nervous testers now that the big red bars aren’t glaring back at us early in a test.

3. Offline validation option for A/B tests
In many cases the Website Optimizer script can’t be validated online and the offline validation option for Multivariate tests is a feature we use often. It’s a welcome addition to have this feature available for A/B testing for beginners, but until it’s easy to copy A/B experiments and re-start them without changing the Control page html we’ll continue using the Advanced A/B/n method.

You can get more detail on the Google Website Optimizer blog. And if you haven’t taken Website Optimizer for a spin yet, what are you waiting for? It’s free!

Note: If you’ve been considering increasing your conversion rates with Google Website Optimizer but don’t know where to start or have hit technical challenges, our conversion optimization experts would be happy to help.

Sign up here

You can read their post here that includes screenshots or read the full case study showing 270% conversion rate improvement here.

Tom Leung, Product Manager for Google Website Optimizer, spent a few minutes with us discussing the updates, the opportunities for improving conversion rates with Google Website Optimizer, and how to use conversion optimization to advance your career – especially in a down economy.

Click the play button below to listen to the audio file of the conversation:

Or, download the mp3 here.

What does GWO do?

In case you’re unfamiliar with the tool, Google Website Optimizer allows website owners and marketers improve website effectiveness by testing which page variations (images, headlines, layout, etc.) persuade the most people to take action.

At WiderFunnel, we’ve been using Website Optimizer since it was in private beta in early 2007 and have used it to improve our clients’ Conversion Rates (or Action Rates) by up to 270%. Let me emphasize that we’re talking about an immediate 270% ROI improvement. So, the tool can do powerful things.

Making the tool even easier

• Signup outside of Adwords. In addition to accessing the tool through AdWords, you can now access it with any Google Account, such as the one they may use for Google Analytics or Gmail.
• Improved multivariate experiment setup. If you’re a do-it-yourselfer, the instructions for tagging your pages have now been improved.
• More resources. They launched a new website last night that has more educational information. Check out our Tourism BC case study under “Success Stories”.
• New Website Optimizer product blog. If you want the latest news about GWO, you’ll get it there.

Don’t let Website Optimizer’s ease-of-use and $0 price tag mislead you. In the hands of a skilled Conversion Optimization team it can be used to dramatically improve almost any website.

Last week Google quietly introduced an update to Website Optimizer that I believe is a great advancement for the tool’s accuracy and user experience, and something that we’ve been asking for. The update extends the cookie life to two years (previously 30 minutes) and across multiple browser sessions (previously only within the same browser session).

We’ve been happy to help Google out in Beta testing the new cookies before their public release and are very happy with the results.

Previously, when a person entered your experiment and then returned to the experiment page a second time after more than 30 minutes or after opening a new browser window, they had a good chance of seeing a different combination of variations than their first visit. The new cookies eliminate this issue.

There are two main benefits:
1. You can now be sure that the variable combination that the user converts from is the one that persuaded them to act (unless they’ve cleared their cookies in the mean time, but there’s no way to avoid that).
2. Arguably more important is that there’s no Distraction introduced if the person is expecting the combination they’d seen previously and now see a different combination.

Both of these factors can have a significant impact, especially in pages involving longer decision-cycles, like some B-to-B lead generation experiments we’ve run recently.

Probably not exciting news for most marketers, but it’s good news for us Website Optimizer fans.

Here are the details from Google:

The first online seminar, Introduction to Website Optimizer, will be geared towards those who are unfamiliar with website content testing and optimization. Tom will discuss the importance and benefits of optimizing your website design and content, and he’ll provide a detailed introduction to Website Optimizer and review the product’s latest features.

The second online seminar, Website Optimizer: Creating & Launching Experiments, builds on the first and is designed for those who have previous
experience with Website Optimizer or other site testing tools. Tom will deliver a step-by-step demonstration of how to successfully launch multivariate and A/B Split experiments, and he’ll also answer your questions.

Schedule and registration information:

Introduction to Website Optimizer (New or inexperienced users)
Tuesday, October 30th, 2007 10:00 – 11:00am PDT
Register to attend.

Website Optimizer: Creating & Launching Experiments (Intermediate and advanced users)
Thursday, November 1st, 2007 10:00 – 11:00am PDT
Register to attend.

Then when you want to develop a Funnel Experiment™ plan to maximize GWO’s potential

Contact one of our Google Authorized conversion rate optimization experts.

You may also want to check out our free resource downloads.

I visited the famous Googleplex in Mountain View last week and very much enjoyed meeting the Google folks and the other pioneers in the conversion rate ‘industry’.

One of the things that has impressed me the most about the Google Website Optimizer team is how receptive they have been to our suggestions on ways to improve the product. Tom Leung and I had an in-depth discussion last week about the ways WiderFunnel is using the tool and how we think it can be improved. In fact, some of the issues I raised with Tom several weeks ago are already being addressed and will be released soon. We’re excited that WiderFunnel is already helping Google with beta testing some new features as we speak (although I can’t talk about the details just yet).

I also very much enjoyed meeting with the other WOACs. Robbin Steif of Lunametrics was forward-thinking enough to organize a sushi dinner including myself, the squirrel guys from Conversion Rate Experts and Tim and the gang from ROI Revolution. I think we all found it helpful to learn from each other about how to build the Conversion Optimization industry and specifically how to build a successful consulting business. I found it most interesting to see how each of our organizations has very different core strengths and areas of focus.

I came away feeling optimistic about the growth prospects for the industry and pleased that our unique Kaizen Method™ planning process and Website Optimizer development feedback will be able to play a key role in the industry’s development.

Thanks again for putting so much effort into organizing a successful event, Tom!

Unfortunately, there is no simple answer.

The number of test combinations and traffic level are the variables that you can calculate. And Google has provided a calculator that does a ballpark job for you.

However, there are variables to reaching statistical significance that can’t be predicted, which are:
1. The conversion rate of the control
2. The spread between the conversion rate of the control and the alternative/test

If the conversion rate is high and the spread between the control and the variations are large, the test will complete relatively quickly. If, however, the test alternatives are very similar in conversion rates to the control, it will take much longer to reach statistical significance.

That is why our advice to clients is to let us design dramatic tests rather than piddling around with a word here and there. There’s much more upside to be had in being bold with the variations.

It’s also tempting to stop an experiment before it’s complete. And that’s fine to do if it’s not completing fast enough and you want to modify it and try again. But be careful in taking learning from an incomplete test. It’s easy to infer too much into all that green on the bar when it is really not ready to tell you anything.