The Official Google Website Optimizer Blog - The most up-to-date product news, industry insights, and testing strategies

Top Myths about Google Website Optimizer

Wednesday, September 03, 2008 9/03/2008 02:30:00 PM


A lot of people ask us similar questions about what Website Optimizer can and can't do, so we thought we'd share the answers with everyone. If you've always wondered, or you've forgotten our answer ;), here's our own Top Myths List:

1. Does Google Website Optimizer work for organic traffic?
Yes! We alternate your content regardless of where visitors come from. You can optimize a page whose traffic is all organic, multiple paid keyword sources, all direct, or any combination of those three.

2. Can I optimize pages that aren't landing pages?
Sure. You can test product detail pages, purchase funnels, any place where there is HTML. You can even use it within web applications.

3. What about dynamic pages?
You can optimize those as well. The multivariate tags simply insert the alternate code as the visitor's browser renders the page being tested. Those tags can be part of a template or dynamically added by your web server using fancy logic. Furthermore, alternate content can be code that pulls dynamic content from a database.

4. Does Google Website Optimizer perform fractional analysis?
Yes. We can serve variations and combinations equally or provide main effects, fractional analysis. This assumes no interactions and can lead to initial insights much faster than full factorial analysis. More detail is on our Help Center.

5. Can I get dedicated account management, consulting or support?
Absolutely. You can get in-depth professional services from one of our Authorized Consultants who are experts at Website Optimizer experiment research, design, instrumentation, troubleshooting, live support, analysis and training. These consultants can be hired on retainer or by the hour.

6. Can I look at experiment data in conjunction with my web analytics data?
Yes. If you're doing an A/B test, simply look in your content report for each page.
If you're doing a multivariate test, look at the cookie we set for each visitor, which will indicate the combination to which they're assigned. Then segment your web analytics report by combination. Check out ROI Revolution's post on using Google Analytics and Google Website Optimizer together.

7. Can I use Google Website Optimizer with my own goal tracking?
Yes, you can split your traffic, alternate the content, and then use another method to perform your statistical analysis. For example, you could run an A/B test, which offers a different telephone number for each page version. In this case, you could probably tell which page is most effective by simply counting the number of calls you receive on each number.

8. Can I run a test across more than one page?
You can put the tags for an experiment on an unlimited number of pages. For example, you can put our tags in a common header file or put them individually across multiple pages. Read more on our Help Center here and here.

9. Does Google Website Optimizer require me to use Google Analytics?
No. If you're using a non-Google web analytics service and are happy with it, you can still use Google Website Optimizer. Just add our tags to the pages where you want to run experiments and leave your existing web analytics tags as they are.

10. Does Google Website Optimizer require me to use Google AdWords?
No. If you already use AdWords, you can conveniently access Website Optimizer from the Campaign Management tab within AdWords. If you don't use AdWords, you can use our standalone version.

11. Can Google Website Optimizer be used on an enterprise website?
Sure. We use it on most of our own sites, such as AdWords, Gmail, and Google Earth. Fortune 500 companies have also used Google Website Optimizer since we launched in late 2006.

12. Can I track non-page goals?
Sure, goal tags don't need to be implemented as part of a separate goal page. You can record a conversion when someone submits a form, clicks on a link, or dwells on a page after a certain number of seconds.

13. Do I need to retag my pages to run a new experiment on the same page?
If it's a multivariate experiment, you can use our copy feature to re-use tags and launch a new experiment directly.

14. Is it really free?
Yes! We think better designed web content makes the web more efficient and benefits users, Web search and advertisers.