5 Best Practices for Google Analytics for 2017

Share Article

WebCroppers reveals the 5 Best Practices for Google Analytics for 2017 to help small businesses use these tools to analyze their data and get more traffic to their site.

News Image
In an effort to help small businesses thrive, WebCroppers has put together some tips that should help even the playing field with the larger corporations on a local level.

Google Analytics is free cloud based software which helps to track activity on a person's website. It’s a very flexible reporting system that allows users to mold it to track e-commerce stores, blog sites and more. In an effort to help small businesses thrive, WebCroppers has put together some tips that should help even the playing field with the larger corporations on a local level.

Here are some of the best practices for 2017 when it comes to handling Google Analytics.

User-ID Coverage

This is a tool meant to unify users across devices and allows people to analyze visitor cyber steps as they travel through a website. The first step is to enable User-ID in a person’s account. Follow these steps for that.

Once it’s setup – navigate to the Audience > Behavior reports to use it. If the person is using the mobile app, it is under Behavior > Behavior.

To use it, WebCroppers suggests creating specific Analytics segments to identify pockets of visitors. This could be the business’s highest spenders, window-shoppers, users who consume the most content, or split them off by channel grouping. To yield more information such as device and time stamps, just click on the individual Client ID and make sure to adjust the date range to get a complete picture.

Weed Out Ghost Referral Traffic

Ghost referral traffic is when analytics data is indicating traffic coming from specific sources to a person's website, but in actuality, does not exist. This is done by pinging the Universal Analytics ID’s to the account. It’s truly a genius loophole people have been exploiting for a few years now – however it can really mess up a person’s data. The most in-depth “How to” guide has already been published; follow it here to clean up the data.

Expanding Attribution Models

The biggest strength to Google Analytics is the ability to see. See where clicks are coming from and where they are leaving, where sales are being generated and what content is really resonating with users.

Because technology is expanding at such a rapid rate – it’s difficult to say that AdWords exclusively got a business that one major sale when the User-ID report shows that the visitor interacted with Direct, Organic and Social before completing the purchase through Paid. In come attribution models.

They are not perfect, and one size fits all does not work here. But it’s a start.

Last Click Attribution – the golden child of attribution. Many marketers use this to quantify specific source/mediums. However, because consumers are being hit by so many different angles – it doesn’t provide the most complete picture. As the name implies, last click rewards the very last source/medium in a user’s interaction with 100% responsibility.

First Click Attribution – This is the opposite of last click; typically used for trends setting models. Whatever the first interaction in a user’s journey, it is rewarded with 100% responsibility for the goal completion.

Last Non-Direct Attribution – This model ignores direct traffic from being rewarded. The thought process is that most users will type the URL in from memory and pick up where they left off, but that doesn’t uncover what channel sparked the memory of the user going directly to the site.

Linear Attribution – Divide the percentage of sales equally over all touch points which were present through the users funnel. This is the model that Bernie Sanders ran his campaign on.

Time Decay Attribution – With time the beginning sources that interacted with a goal completion are minimized and the most recent sources are increased. Motion picture promotions are a good example for Time Decay, they don’t care that last year social media interacted with the most movie goers if the majority of users interacted with radio ads 5 days prior to the movie release date.

Position Based Attribution – Quite a bit more arbitrary, but a good model for rigid marketing funnels. Referrals start a user’s journey to a goal completion and website visits close. With their goal completion, people automatically allocate a percentage of the reward to referrals and direct channels. Any interacting channels in between will receive an equal portion of the remaining reward.

Weighted Attribution – Whichever channel converts at a higher percentage, receives a higher reward. Simple and can provide interesting insights.

Algorithmic Attribution – Custom, rule driven attribution. Very arbitrary and only recommended for those who have a large historical data set available.

Do Not Use Sampled Data

This should be a golden rule for anyone – sampling in Google Analytics happens when people cross a data threshold under a free account. What it does is average out all the data. The effect of this is that businesses do not get to see the whole picture. They will know if the data is being sampled if they see this in the top left corner of their Google Analytics screen.

This means that out of the date range selected, the report is based off only 12.33% sessions. At this point, people should minimize the date range and slowly but surely piece together the information. People can also upgrade to the paid version of Google Analytics or pay other Analytics platforms, but they would need to justify their department dropping $100,000 minimum per year for full access.

Leverage Affinity & In-Market Segments

Display targeting is going to be a hot commodity going into next year – there is no better time to experiment with building programmatic data set than now. Navigate to the Interests report on the left hand side and go into the Affinity Categories. Here, people will see groups divided into sub-types such as Shoppers/Shopaholics, Movie Lovers, so on and so forth.

Select the secondary dimension to ‘In-Market Segments’ and people will get a better picture of how to approach the display network. An affinity category symbolizes what type of individual the users are and the in-market segment highlights what products they are shopping for long enough that Google can compile historical data on them.

For example; if someone is selling flower vases, they would be very interested in attacking the ‘Home Décor Enthusiasts’ Affinity. People would also need to find out what in-markets these users are visiting to be able to put their display ad into a setting. If a user’s affinity is ‘Home Décor Enthusiasts’ and in-market ‘Home & Garden/Home Furnishings’ a good display ad would showcase a sitting room, furnished nicely, with the focus being on a brightly colored flower vase. This would invoke the users preference to home décor as well as speak to their in-market purchasing habits of having an eye for furniture while promoting the vase.

For help mastering the tools and data that Google Analytics has to offer, people can contact WebCroppers for a free consultation.

Share article on social media or email:

View article via:

Pdf Print

Contact Author

Eugene Feinman
+1 480-659-5240
Email >
Follow >
Follow us on
Visit website