Lately, I had the opportunity to examine the traffic source, content and keywords using Google Analytics for AltSearchEngines, a professional blog with heavy traffic and blog activities. Because this blog already has tons of traffic and few years of historical data I had enough depth of information to discover interesting blogging patterns. The specific blog structure is a single blog with multiple authors what that make using Google Analytics even more interesting. I’m not an expert in web analytics or SEO but I want to share some of my findings and I hope to hear and learn more from other web analytics users. In this post I also share some ideas for improving Google Analytics interface, those could make it even easier to detect some of the discussed patterns. Finally, I offer some ideas to act upon these patterns for improving future traffic.
Looking for patterns using Google Analytics
- Right now I’m using the blog URL convention (date) to isolate Information for posts that were posted during the selected duration.E.g. looking at the traffic during September 08 contributed by posts that were only posted during that same month. Today Google Analytics shows traffic for all the posts on this blog for the selected duration and it requires some work to isolate the new posts from the old one in order to see how they contribute to the selected duration traffic. Why is this important? Pattern # 1 progress – does your content gets better? Maybe the traffic is all based on old success. The inverse – same as #1 but for old post s- i.e. show me the traffic for the past month excluding posts posted in the last month. It would be great if Google Analytics can do that for you.
- I use spreadsheet to separate traffic for “generic” pages only– i.e. the home page, RSS button, page number etc… Why is this important? Each item on the blog other than blog posts has different importance and objective behind it so isolating these can help to fine tune each individually. It is also good to know if people are finding the root folder (i.e. your home page) maybe using bookmarks or blogrolls. Pattern #2 – are you building a great brand name? The inverse is interesting too – exclude “generic” pages from the report and allow focusing on blog’s posts impact only. It would be great if Google Analytics can do that for you.
- Support for blogs with multiple authorsor blog with many guest bloggers (not blog network). It will be a great help if you could see who is the blogger, in Google Analytics, next to the blog post traffic info. This could be automated (using a certain standardized way telling Google who wrote this post). Why is this important? Pattern #3 – finding the best bloggers on the team and learning from them.It could be the content, the style, the keywords, the community that these bloggers are building. I’m sure that this is useful in blog network too, yet over there it is easier to isolate bloggers traffic because each has his own domain name.
- Daily traffic – hourly – higher granularity. Currently if you want to see daily traffic you have to change the date to today. There is no way to see hourly traffic. I wish there was a separate page for daily traffic and the resolution is in hours. Why is this important? Pattern # 4 – timing the post – if you noticed that your recent post is just great, you may want to leave it a little longer on the top of the blog. You may find that one social media site like StumbleUpon, Digg, Delicious, Twitter, FriendFeed and more works better during different hours. It could also help learning when not to post (hint: 5pm eastern time during the work week:)).
- Pattern #5 – lost opportunities or finding hidden gems – Finding your best historical stuff (not traffic based) from the time that you were less known as a great blogger or with smaller supportive community around you – looking for pages with the highest Time on Page and the smallest Bounce Rate yet low traffic. Just be careful, not to judge too soon. If there was not a lot of traffic the Time On Page number may be high – Google average it, so too small of traffic may keep this number high. Once you find a post that match this pattern try stumbling it.
- Page view per visit– Google provide this number and you can track it over time. It shows if users stay longer on your web site reading more blog posts or leave quickly to find other more interesting blogs out there. It is important to build links between relevant blog posts on your blog (not excessively though). For blogs with multiple authors this is a real challenge. How other bloggers can see what to link to? This is when categories helps. If the team has visibility to exiting blog posts in each category, then they can think about how their post should be categorized, and then look for relevant blog posts to link to. Google Analytics can help here too, if it has access to the categorization information then it could show the success of one category over the other. If not then it would be great to annotate blog posts row’s with more info (having custom properties) . Pattern #6 – content type (category) success.It is also important showing blog’s recent and best posts on the sidebar, so the users can easily find them. The blog performance is another factor! People will not hang around slow web-site.
- Keywords– there is a ton of data out there about this subject so I may repeat other’s finding. My first finding is that people love the word “list”. People love list but they really like this word in the title more than “best” (I think that Google hate this word). I could not resist using the word “list” for this post title:) People like to see the benefit (value) from reading your blog post more than artistic titles, so “hot to” is a great start for a title. Speaking about titles it is important to think about them for the long run too.If you write about a company or product put the name in the title. Why? So, if someone will be looking for this company few month from now, using any search engine, there is a chance to find your blog. Simple, yet I do see the opposite happening even till this day. You’ll see the difference using Google Analytics very quickly. Some SEO experts tells you to optimize titles for the short run and then to modify them for the long run (beware from changing the permalinks) and Google Analytics can help telling you when is the right time to modify the title but I’m not sure if it will truly helps. I don’t think that Google come back to re-crawl this content. Pattern # 7 – keywords timing and tuning.By the way, blogs with multiple authors or blog networks that are providing training and guidelines can bring new bloggers up to speed quickly and avoid such simple mistakes.
- US vs. the rest of the world – Google Analytics provides regional segmentation’s. You can see where your audience is coming from and how fast that region grows. So you can use this information to keep them engaged. For instance European use Jaikumore than Twitter so post updates from your blog over there and build community on Jaiku if this is your target audience. This is true for other social networks too (e.g. Facebook is very weak in Japan). Write content about local products, news and companies relative to the regional success – pattern # 8.
- The right traffic source allocation and effort.Google Analytics dissect traffic into three categories: Direct Traffic, Referring sites, and Search Engines.As a blogger most of your impact is in the Referring web sites category. Especially in the beginning. This is where your work, building readers community is the key to your success. Again, Google Analytics shows you where your traffic is coming from and where it lands (landing pages). For blog with multiple authors (or for blogs network) it would be great to see the contribution by author, to reward the ones that expend their readership in this way. pattern #9 – are bloggers on the team growing their community? The Traffic Source percentage allocation can also tell you where you should invest your effort. It is different from case to case though e.g. if your blog talk about products (and sell them or use advertise on your blog for selling them) then SEO using the right keywords is the most important activity and the traffic source allocation should reflect that. For most other cases you should divide your time something like 30% reading other blogs or experimenting with tools, product, services, projects, 30% writing content and devote 30% of your time for community building. I really like to hear some feedback about this suggested allocation.
- mmmmnm, do I have to come up with the tenth? …OK, Pattern # 10 – what activities are helping you to find new readers?Google Analytics shows you two visitors types – new and returning. Look for what is it that you are doing bringing new users to the site. StumbleUpon is a great option. You can try tagging your blog using different tags. There are numerous places to tag these days: WordPress, Technorati, StumbleUpon, Delicious to name a few. Check hashtags.orgfor most popular and recently added hashtags. You can Digg posts under different categories (business vs. technology). You can use TwitterPack and Twellow for finding people from different categories and sub-categories to follow, so they can follow you back, then build new relationships. You can comment on blogs from different categories (using Technorati and recently the revised Google Blog Search). You can use backtypeto see where bloggers leave comments and follow their actions. Keep checking Google Analytics to see what works and how the new visitors vs. returning, percentage changes.
There are more capabilities in Google Analytics than described in here; like finding broken links, specific browser and OS optimizations and exit pages to name a few. I have a lot more to learn but I think that what’s in here are some of the more interesting patterns that you could discover using Google Analytics in the first few weeks using it.
Small disclosure: I work for a company that sells web analytics product but I don’t work with that team and I know very little about it.
Only started looking at Xoost – search sharing i.e. United we search, divided we fall (on spam)
I barely had a chance to explore Search Wikia – from Google: “open source Internet search engine, to which the community can contribute.” I call it read/write search. It seems like an interesting concept.
I’m honor by the coming requests to look at new technologies. I wish I had more time to do that.
I’m also waiting for Twingly launch (possibly soon) – check the spam free search – very cool and functional.
RecommendBox is getting more and more interactive. Having more members form my location can make it even more valuable. For some categories it is only useful for a community sharing local knowledge (restaurants, events, places). For others it does not matter so much (books, websites, video games). I can see many areas how this kind of a service improve and I believe it will over time. If this flies they have a chance to make good money from advertisement. People from MA, USA please come and check it out . It is still in a very early stage of development.
Other things that keep me busy – for one it is work! We have our own release coming soon.
I’m also working on a report that sums some of the progress I made in the “great blogger fishing project” (you can start from this post and keep going down the thread).
My @Blogmon discovered this great resource if you care about SERP: SEOTops, apparently Musa Aykac knows a lot about it and is willing to share (95.7% gain in Technorati rank in less 38 days, WOW! – now: 29898).
Bye for now.