How to hold a variable constant for an A/B test
-
For example, let's say you want to A/B test a title tag change. You are hoping to identify whether a title tag change increases CTR. But, position is always fluctuating a bit and that affects CTR, too. So, I'm interested in how you could hold position constant in order to isolate the change in CTR that is due to the title tag change. Does anyone know of resources/tools/tutorials for how to do this?
It's been... a very long time since I took statistics (-: I have access to Excel, MS Access, and R studio.
-
Thanks for the reply. I'll check out the links you included.
I thought it might be helpful to show an example of what got me thinking along these lines. This is a study about A/B testing of title tag changes. They don't say they accounted for position fluctuation; maybe they didn't, but it seems like you'd have to in order to have meaningful results.
-
I don't see a way of doing this accurately. With organic rank fluctuations, seasonality, competition it's near impossible to "hold" a position. However, if you have a position in serps for a specific keyword with little fluctuation and get decent traffic you can establish this as your baseline. Once the title is updated and appears in the serps, begin monitoring w/search console, analytics or etc. and see if you see any decent jumps. Some links for additional information: Finding the ROI of Title tag changes using Google's CausalImpact R package, SEO Split Testing & A Beginner’s Guide to A/B Testing: Effective SEO Landing Pages. Would love to hear if anyone else has a better way.
Browse Questions
Explore more categories
-
Moz Tools
Chat with the community about the Moz tools.
-
SEO Tactics
Discuss the SEO process with fellow marketers
-
Community
Discuss industry events, jobs, and news!
-
Digital Marketing
Chat about tactics outside of SEO
-
Research & Trends
Dive into research and trends in the search industry.
-
Support
Connect on product support and feature requests.
Related Questions
-
Site migration/ CMS/domain site structure change-no access to search console
Hi everyone, We are migrating an old site under a bigger umbrella (our main domain). As mentioned in the title, We'll perform CMS migration, domain change, and site structure change. Now, the major problem is that we can't get into google search console for the old site. The site still has old GA code, so google search console verification using this method is not possible, also there is no way developers will be able to add GTM or edit DNS setting (not to bother you with the reason why). Now, my dilemma is : 1. Do we need access to old search console to notify Google about the domain name change or this could be done from our main site (old site will become a part of) search console 2. We are setting up 301 redirects from old to the new domain (not perfect 1:1 redirect ). Once migration is done does anything else needs to be done with the old domain (it will become obsolete)? 3.The main site, Site-map... Should I create a new sitemap with newly added pages or update the current one. 4. if you have anything else please add:) Thank you!
Intermediate & Advanced SEO | | bgvsiteadmin0 -
SEO benefits/drawbacks of physical address on site.
Simple question - how much of an SEO impact does NOT having a physical address on your site make? I run a photography business based from home. For security reasons I don't really want our home address easily available to the public. I have a Google Maps listing (actually seem to have two at the moment, but that's a different matter) and the full address is logged there but is not viewable publicly - just the surrounding area. Any thoughts would be appreciated.
Intermediate & Advanced SEO | | robandsarahgillespie0 -
Google only indexing the top 2/3 of my page?
HI, I have a page that is about 5000 lines of code total. I was having difficulty figuring out why the addition of a lot of targeted, quality content to the bottom of the pages was not helping with rankings. Then, when fetching as Google, I noticed that only about 3300 lines were getting indexed for some reason. So naturally, that content wasn't going to have any effect if Google in not seeing it. Has anyone seen this before? Thoughts on what may be happening? I'm not seeing any errors begin thrown by the page....and I'm not aware of a limit of lines of code Google will crawl. Pages load under 5 seconds so loading speed shouldn't be the issue. Thanks, Kevin
Intermediate & Advanced SEO | | yandl1 -
Help article / Knowledge base SEO consideration
Hi everyone, I am in the process of building the knowledge base for our SaaS product and I am afraid it could impact us negatively on the SEO side because of: Thin content on pages containing short answers to specific questions Keyword cannibalisation between some of our blog articles and the knowledge base articles I didn't find much on the impact of knowledge bases on SEO when I searched on Google. So I'm hoping we can use this thread to share a few thoughts and best practices on this topic. Below is a bit more details on the issues I face, any tips on how to address them would be most welcome. 1. Thin content: Some articles will have thin content by design: the H1 will be a specific question and there will be only 2 or 3 lines of text answering it in the article. I think creating a dedicated article per question is better than grouping 20 questions on one article from a UX point of view, because this will enable us to direct users more quickly to the answer when they use the live search function inside the software (help widget) or on the knowledge base (saves them the need to scrolling a long article to find the answer). Now the issue is that this will result in lots of pages with thin content. A workaround could be to have both a detailed FAQ style page with all the questions and answers, and individual articles for each question on top of that. The FAQ style page could be indexed in Google while the individual articles would have either a noIndex directive or a rel canonical to the FAQ style page. Have any of you faced similar issues when setting-up your knowledge base? Which approach would you recommend? 2.Keyword cannibalisation: There will be, to some extend, a level of keyword cannibalisation between our blog articles (which rank well) and some of the knowledge base articles. While we want both types of articles to appear in search, we don't want the "How to do XYZ" blog article containing practical tips to compete with the "How to do XYZ in the software" knowledge base article. Do you have any advice on how to achieve that? Having a specific Schema.org (or equivalent) type of markup to differentiate between the 2 types of articles would have been ideal but I couldn't find anything relating to help articles specifically when I searched.
Intermediate & Advanced SEO | | tbps0 -
Are there any issues with search engines (other than Google/Bing) reading Protocol-Relative URLs?
Are there any issues with search engines (other than Google/Bing) reading Protocol-Relative URLs? Specifically with Baidu and Yandex?
Intermediate & Advanced SEO | | WikiaSEO0 -
What things should I consider if I am doing a 301 redirect on only 1 page/blog post?
I wrote a blog post on one of my websites and it got picked up by reddit and I got a bunch of nice backlinks and now that website got a nice boost overall, and especially that blog post page. I now wish I would have posted the article on a different website of mine. I would prefer if this other site was getting the traffic and the good backlinks that I've acquired. What are the pros and cons if I move the content over to my other website, and 301 redirect just that one article to the article location on my other website? The blog post I wrote almost instantly began ranking for certain terms in Google. Ideally I would like my other website to rank for those terms, but I realize there will be some differences as search engines look at the website as a whole and take many factors into consideration. I know there are tons of case studies and information about moving entire sites etc but I couldn't find much on this. Any advice, questions or comments would be greatly appreciated. Thanks,
Intermediate & Advanced SEO | | bradbowman
Brad0 -
Thousands of /img/img/img urls generated by website - where are they coming from?
Hello -just fed website into Screaming Frog and ended up crashing computer as these img/img/img urls went into the 10s of thousands (and the numbers of img/img/img/ in each URL ended up going into the dozens and probably hundreds and more per URL). Never seen anything like it! Any idea what might be going on with this website and why it's generating so many of these URLs - it is anything to worry about? Here's example of shorter URL... www.company.com/discover/img/img/img/img/img/img/img/img/img/img/img/img/img/img/img/img/photo-competition-winners
Intermediate & Advanced SEO | | McTaggart0 -
First Link Priority question - image/logo in header links to homepage
I have not found a clear answer to this particular aspect of the "first link priority" discussion, so wanted to ask here. Noble Samurai (makers of Market Samurai seo software) just posted a video discussing this topic and referencing specifically a use case example where when you disable all the css and view the page the way google sees it, many times companies use an image/logo in their header which links to their homepage. In my case, if you visit our site you can see the logo linking back to the homepage, which is present on every page within the site. When you disable the styling and view the site in a linear path, the logo is the first link. I'd love for our first link to our homepage include a primary keyword phrase anchor text. Noble Samurai (presumably seo experts) posted a video explaining this specifically http://www.noblesamurai.com/blog/market-samurai/website-optimization-first-link-priority-2306 and their suggested code implementations to "fix" it http://www.noblesamurai.com/first-link-priority-templates which use CSS and/or javascript to alter the way it is presented to the spiders. My web developer referred me to google's webmaster central: http://www.google.com/support/webmasters/bin/answer.py?answer=66353 where they seem to indicate that this would be attempting to hide text / links. Is this a good or bad thing to do?
Intermediate & Advanced SEO | | dcutt0