BUS572- Blog #2

                                                  Web Analytics and Conversion Optimization

Concept #1 - Log-File Analysis and Page Tagging

Log-file analysis is a very important concept that I took away from chapter 15 because it can provide my team with a lot of in-depth information that will be useful in our Google Marketing Campaign. Log-file analysis is basically an analysis of log records. This information can be used to show how often a potential consumer visits a certain site and it records all clicks that take place on the server. This software can also record where visitors are coming from geographically as well. When log-file analysis is used with internet cookies this software can provide even more in-deph information. Page tagging is the process of tagging web pages with embedded Javascript files. Page tagging will send this information to a third-party server where the statistics can be analyzed.

There are many pros and cons with both of these technologies as well. For example, page tagging can be less accurate than log files because log files record every click, but on the other hand if a browser doesn't support Javascript then no information will be recorded using page tagging. Log files are also in a standard format and can be easily switched between different vendors to analyze historical data, but if you were to use page tagging you might lose valuable information when switching to a different vendor. Log files also record search engine spiders which is extremely useful for search engine optimization and log files also record failed requests, while page tagging only shows successful requests. However, page tagging also has some advantages, such as, the software recording Flash movie interactions whereas log-files cannot. Page tagging can also be utilized by companies that may not operate their own websites. Finally, page tagging providers offer a higher level of support because of third party services. I'm not quite sure which method I would prefer more. At first, I like the idea of log-file analysis because it seems to be a lot more accurate when collecting and utilizing information, but page tagging seems to offer some great services that you cannot get with log-file analysis. However, page tagging can be so complex that it could take weeks or even months to complete and I severely question whether it would be worth it or not to use this type of service. What if you had a deadline to complete a project, like our GOMC project, would you prefer Page Tagging?

Concept #2 - A/B Split Testing 

source:http://www.emailloop.net/images/how_ab_works_diagram_695.jpg

A/B testing measures one variable at a time to determine its effect on an outcome while all of the other elements of the advertisement stay the same. You can create different versions of the variable and test them accordingly. For example, the phrase "buy now" on an advertisment can be moved around to see if there is an increase in sales or not. You can always test more than one version of the variable,  but it will take longer to find results from multiple versions. Next, traffic is then randomly distributed through the new versions and the outcomes are measured for each variable. Once you apply statistics to see which one is the most significant then you will know which advertisement works best and it can then be used in the future. This can be extremely important for digital marketing because you would be able to easily manipulate each part of your campaign or adverisment to see which placements work best. You will also be able to create more conversion rate lifts results. One disadvantage would be that you would have to plan carefully if you were to investigate interaction effects between two islolated elements. I really like the idea of A/B split testing because you are only testing one variable at a time, but would it be more efficient to use Multivariate testing to manipulate many variables at one time? I feel like Multivariate testing would be much more difficult to preform. It seems like you need a large sample of data to have any significant findings, but what if you were only working with a small sample of data? Here is a website that compares and contrasts the two different methods.

Skill- Organization of the AdWords Account

I believe that one of the biggest skill sets I've learned this week is the organization of the AdWords Account. In our first class, we discussed a lot about keywords , but I wasn't quite sure what to do with those keywords once I came up with them. After Sarah's presentation, I had a much clearer vision of how an Adwords campaign is set up and how the campaign functions. I realized that we had only one campaign we needed to run for our client, but I didn't realize that you can have many different Ad Groups. Under these Ad Groups, you can have hundreds of different keywords and the Ad Groups just allows you to organize your thoughts better and put certain keywords under the correct Ad Group. I'm wondering how many Ad Groups should our team have to maximize our return on investment? Should we narrow it down to only three or four Ad Groups or should we expand and make multiple Ad Groups?  This has also really helped with our second team assignment because we were all a little confused on how everything should be organized, but Sarah definitely painted a clear picture that we can all use in moving forward with our client's campaign.

Personal Journey Reflection:

This project is still very new to me, but with the help of Sarah's presentation I believe I have a much better grasp on our mission. I am still struggling with all of the different Google tools, but once I start physically messing around with AdWords I'll be just fine. It has been a struggle to really brainstorm ideas about this project, but I am confident that we will be able to do so soon. I'm really looking forward to the actual running of our campaign and to really see how it will affect our clients business.

Sources:
Stokes, R. (2011). emarketing: The essential guide to digital marketing. (V 1.0 ed.). Quirk Education. Retrieved from http://catalog.flatworldknowledge.com/bookhub/reader/19