How to Set Up Advanced Segments

/in /by

When you create a new view filter all of the data starts getting filtered from that day going forward. That’s all well and good, but what about all the data that was gathered before the filter went into place? Well let us introduce you to Google Analytics Advanced Segments.

What are Advanced Segments?

Without overcomplicating things, Advanced Segments allow you to segment or filter data based on a set of rules or filters. You can filter on something as simple as all traffic from Colorado or as complex as only show visitors who viewed page A, B, C but didn’t see D. When you implement the segment it applies to all the reports in Google Analytics, with a few minor exceptions, so you’re able to see how a particular segment navigates your website. Additionally, you can compare up to four segments at once, so you can compare and contrast behavior between different segments.

What to Segment?

There are tons of possibilities of data to segment. Google gives you several segment ideas out of the box, such as Demographics, Technology, Behavior, Traffic Sources, etc. I’m a little old school, so I usually build my own from scratch using “Conditions”. This basically allows you to use AND/OR statements to either Include or Exclude particular traffic. You have access to tons of dimensions and metrics on which you can segment. Additionally you can use “Sequences” to define particular paths or actions that visitors take and filter them out if they don’t follow those paths.

Why use Advanced Segments?

We use segments nearly every day as they allow a retroactive filtering capability to track down issues or to find your most engaged visitors. For instance, we created franchisee advanced segments for a business that had locations all across the country. We used the City dimension and included all of the cities that each franchisee covered. Another example is a school supplies company who offers both hard copy and digital products. We created advanced segments to show us visitors who purchased hard copy vs. digital products to see how they arrived on the site and if they navigated the site differently.

The opportunities for segments is endless (well, within 1,000 or so) so create them to your heart’s content and start going down the rabbit hole to discover your best visitors and optimize their website experience!

How to Set Up Custom Dimensions & Metrics

/in /by

Back in the day we used to get by with sending specialized user defined variables into Google Analytics. Along with Event Tracking we could really start to paint a pretty granular picture of site behavior and actions. Then came along a new player in town called Custom Variables. This guy allowed us to really take more control over our variables, such as specifying whether it was a page, session or visitor variable. While these variables were great, there were certainly some limitations, especially in the reporting aspect. However with Universal Analytics Google helped solve some of these gaps and brought us to a whole new world of tracking.

What is a Custom Dimension?

Say I have a news site that has multiple content sections. Within those sections I have several different articles. It would be really beneficial to know what site visitors are consuming most of, is it Sports or is it Entertainment? We could create a Custom Dimension for “Section” and fire a ‘hit-level’ custom dimension on each and every page. What about your users? Say I have a log in section where in my CRM I have information about the user, perhaps a “Home Owner” vs. a “Renter”. In this case maybe a fire a a ‘user-level’ custom dimension upon login and therefore I can segment site traffic between those different segments. Google spells out all of the gritty details about custom dimensions here.

What is a Custom Metric?

Without getting too in the weeds (Google explains custom metrics in confusing detail here) I want to paint a few pictures of how you might use a custom metric in the real world. First off a metric is always an integer that can be counted. Whereas a custom dimension is text, custom metrics allow you to start tabulating values. One example might be a real estate site. They could send the home “Asking Price” as a ‘hit-level’ custom dimension and then be able to tabulate the average price of a viewed home (eg: “Asking Price” / “Pageviews”). Or maybe you have a live chat feature and you want to measure the pages where visitors are having to engage with chat more often. You could send a “Live Chat” custom metric at the ‘hit-level’ and be able to view the pages generating the mot live chat engagement. Or better yet, do a calculated metric based on “Pageviews” / “Live Chat” to see the percentage of page viewers who had to engage with chat. Perhaps there’s something on that page requiring attention.

How to set up Custom Dimensions & Metrics?

In the Google Analytics Admin you will want to pop on over to the Property Admin settings and scroll down to “Custom Definitions”. After you click there you have the option to set up both Custom Dimensions and Custom Metrics. Remember to set your strategy first so that you are confident in the scope to use. We like to use a quick spreadsheet that outlines all of our custom definitions along with their scope and a detailed note outlining where/why they are being set.

The examples for custom dimensions and custom metrics go on and on, but our goal here was to provide some strategic guidance into how you might want to set them up. If you’re getting stuck or have any questions feel free to reach out to us!

How to Set Up Event Tracking

/in /by

Google Analytics’ Event Tracking allows you to easily track user behavior on your website by counting clicks and other actions that visitors do. While we have discussed Event Tracking in the past today we will focus on naming conventions and the strategy for setting them up.

What to Tag With Event Tracking

First off, did you know Google Analytics has a hit limit of 10 million per property per month? For most regular websites this is never an issue, but as you start adding a lot of advanced tagging you may start coming close, so just keep an eye on your usage statistics. That being said, we like to tag just about everything! You never know what you may want to do analysis on so why not track everything you can? Anything that can be clicked that is not an internal link OR is some sort of button or action we want to group together gets an event tag in our book.

How to Name Event Tracking Tags

There are four main naming variables within each event; Category, Action, Label and Value. All are text oriented except for value, this is an integer only. The goal for naming events is to do so in such a way that groups or aggregates them for reporting purposes. For example, “Shopping Cart Buttons” could be the Category for every button within the shopping cart funnel. The Action is often tagged as a “Click”. I despise this strategy because you only have three text fields so why use up one on something as useless as “Click”? In our shopping cart example, maybe Action could be “Continue” or “Add Promo Code”. Obviously this is vastly more helpful than knowing if someone simply clicked a button. Lastly there are Labels. We typically use this variable to track Page Path. We do this so that we can understand what buttons visitors are click8ng on individual pages. Remember that each event group can be viewed individually or together. This means you can look at all Actions that fired on the entire website or just the Actions within a single Category.

Hopefully this helps guide you through your Event Tracking naming conventions, but if you need more help feel free to give us a shout!

How To Set Up Google Analytics Campaign Tracking Variables

/in /by

We’ve spoken about Google Analytics Campaign Tracking Variables before, but today we want to discuss the strategy behind them. For a quick refresher, Google offers up five open text fields that allow you to specify and describe a link into your website. Remember, ONLY use campaign tracking variables for inbound links to your website and DO NOT use them for internal campaign tracking.

Strategizing Your Campaign Tracking Variables

It is amazingly important to list out your strategy for naming of all five variables. Typically we use a “matrix” of sorts using Google Sheets to make sure we always use the same naming conventions. If you do not use the exact same naming conventions for like variables then they will show up as separate line items. For example, “email”, “Email” and “e-mail” would all show up separately under a Medium report and thus you’d have to add them all together. A matrix allows you to keep track of them all and ensures they will always be properly grouped.

Channel Groupings

Google Analytics already creates a Default Channel Grouping for all your inbound traffic sources. However you may notice that sometimes a Traffic Sources report shows “other”. This is because Google is receiving campaign tracking variables that it does not understand. With the strategy above you’ll also want to make sure that all forms of variables are indeed identified by Google through these channel groupings. You may either just add new rules for any new variables into the Default Channel Groupings or you may add your own group. Keep in mind that you may want to create a group for “Paid” vs. “Non Paid”, or perhaps “Remarketing” vs. “Prospecting”. Google allows for up to 50 different Channel Groupings so filter away to your hearts content.

How To Set Up Google Analytics Accounts, Properties & Views

/in /by

The setup process is well documented by Google and other websites, so here we will discuss the strategies behind setting these up.

Account

This is the highest level and is the umbrella that houses your Properties and Views. You are limited to 100 Properties within an Account. Very rarely will you need to exceed this limit unless you are an agency or have tons of random domains. Although the latter probably contains SEO issues as well. Our advice is to start with a single Account and read on.

Properties

Here is where the strategy begins. Properties are typically used for a single website address. Often marketers will setup a separate Property for sub-domains as well. Our advice is to track all traffic under a single Property utilizing different Views UNLESS:

  • You want to limit login access to different data sets (You cannot do this for Views)
  • Traffic between sub-domains does not happen, so they are independent and visitors do not cross between them

Views

View filters allow some pretty granular targeting so we opt to use different Views when we can. The biggest problem is that Google Analytics restricts you to just 25 Views. To start we always recommend a Raw Data View that is completely unfiltered. This may help diagnose website issues down the line. From there we will discuss how we want to see our data. Here are some questions we typically ask to determine if a View is needed:

  • Do we need to replace URLs to better group webpages?
  • Do we have very specific site sections, such as News or Sports, that we want to easily see separately?
  • Are we focused on a particular traffic source?

Remember that Views are NOT retroactive and often you can utilize Advanced Segments for many of these reports. Typically we only advise different Views if you need to rewrite URLs (your or any other data for that matter). If not, Advanced Segments should do the trick, AND they ARE retroactive!

How to Set Up Google Analytics

/in /by

Over the last few weeks we have encountered several businesses that either do not have web analytics of any kind or they have analytics implemented but not well. This is an issue that is near and dear to my heart as a data geek but also as a marketer in general. How is it possible to know what is working and what is not when you don’t have proper tracking in place? So in the interest of making everyone smarter and better at tracking we have decided to do a six-part blog series that covers everything you will need to develop a strategy for setting up Google Analytics from start to finish. There are a million blog posts out there about the tactics of setup, but with this series we are trying to explain the how’s and why’s so you can develop a killer analytics strategy to get the actionable data you need!

How to Set Up Google Analytics

Why Analytics Projects Fail

/in /by

“Hey, we’re launching this Commerce site and we need to know if we made more money, can we track that?” I’m paraphrasing, but seriously, this happened. The first ‘shake my head’ moment would be that a site is nearly ready to launch without fully understanding HOW to track it. The second is looking at this project in a vacuum, assuming this is a piece of a larger strategy. There I go again, assuming.

How to win at web analytics

Ready for the simplest answer in the world? Plan! It’s really that easy. You don’t need a data analyst to tell you what the business goals are. You may need them to uncover deep metrics that lead to better sales, but your overall goals should be about the business as a whole. So when you start discussing a new project YOU should be the person that says:

  1. What metrics will we look at to deem this project a success?
  2. What is our plan for ensuring we can track those metrics?

We always like to do analytics tagging guides during the wireframe or design phase of website redesigns. This is because often one of those ‘necessary metrics’ will require some level of development assistance. Better the developer team knows about this before they get too far down the line. It also helps to understand what kind of advanced tracking you may need, such as events or custom dimensions.

Don’t do piece-meal analytics

It is a minimal effort to apply Google Tag Manager code across the entire site as it is to do it on a single page. Yet I have been approached numerous times to implement GTM on simply the order confirmation page for commerce tracking. This is reason number two of why a lot of analytics projects fail because it often is never thought of as a whole. Do yourself a favor, even if you are only implementing a new site section make sure to revisit your analytics strategy to ensure all of the pieces fit together nicely!

If you need help with this we are happy to help, but just by keeping the two simple questions listed above you will be lightyears ahead of many of your competitors!

Google Analytics Settings in Google Tag Manager

/in , /by

Do you have similar Google Analytics settings that you keep having to enter each and every time you enter a new tag in Google Tag Manager? Well fret not as Google has saved the day for you with “Google Analytics Settings” as a variable inside of Google Tag Manager. This variable allows you to set all of the typical settings you used to set at the tag level, but now you can save yourself some time by doing it once at the variable level.

How Do Google Analytics Settings Help?

Great question and I’m not so sure I would ever use them. Why? Well there are times when you may want to set the same Custom Dimension across every GA fire. Or, more likely, the cookie domain status field. The big thing to remember here is that whatever you set at the variable level will then be applied EVERY TIME a tag with that variable fires. If you want to set these to be more specific at the tag level then you can do that by selecting the “Enable overriding settings in this tag” option.

A Common Issue This Doesn’t Solve: Tracking Code

While using Google Analytics Settings variables will help make field settings easier they do not help with the problem of having different Universal Analytics account numbers per domain. Typically we will use a “Lookup Table” to serve up the right UA-### based on the hostname a user is accessing the site from. So for instance staging.bluefinstrategy.com vs. www.bluefinstrategy.com (please tell me you have different Universal Analytics accounts for these?!?!). This is important when you have tons of different Google Analytics Events and other things that you’d like to track. Why? Well, typically you’ll want to test them on a staging environment first then push them to production. Instead of having to change the tag when you do that push using a variable allows for the tag to work automatically on both domains!

Both the tactics above allow you to replicate settings using variables and should speed the entry of new Universal Analytics tags within Google Tag Manager. Happy Tagging!

You’re Doing Google Analytics Wrong

/in /by

I’ve had a few conversations the last few months where someone told me that we couldn’t tag something a certain way because Google Analytics isn’t set up to do it that way. And no, I’m not talking about personally identifiable information here. You must be able to think outside the box and break free from what everyone else is doing to ensure you’re getting the most value out of your web data!

Event Tracking

“Action means an Action”

An “Action” was defined to me by the person I was chatting with as the ‘exact action the user takes’. Therefore all of his event actions were labeled as “Click”. My argument was; well, if the event fired didn’t they click? Why do you need to know that? Instead why not use the field to answer better questions, such as Page URL, Click URL, Button/Link Type, etc. See, Google Analytics Event Tracking allows you to put whatever you’d like in those fields (except only an integer for Value of course), so why not take advantage of the flexibility they are offering? How many reports will you pull to see “Event Action=Click” vs. “Event Action=Page URL”. With the former there is no actionable data, but the latter allows you to know which pages may be driving the most clicks. Again, there are a ton of different use cases here, but just don’t get bogged down into thinking you have to follow Google’s naming conventions as law.

Campaign Tracking Variables

“Cost Per Click Advertising”

Campaign Tracking Variables are tough ones, because they do inherently drive a lot of out of the box reports within GA. However, you’re not necessarily handcuffed, especially outside the Pay Per Click walls. For instance, we usually only tag PPC advertising with the “cpc” medium, which is what Google does out of the box. But AdWords is not the only “Cost Per Click” medium in the world. You might be using display ads or social feeds to drive visitors and every time they click you get charged. For me, I’d like to see them all differently as separate line times, so I create separate “Mediums” for them. But alas, grouping is not lost. If you’d still like the group them in an easy report BUT have the flexibility to see them separately Google offers you Custom Channel Groupings. This gem allows you to group inbound traffic in any way you see fit while not losing the granularity you might need to make very specific decisions. This doesn’t just apply to “Mediums” but all of the other ‘UTM’ variables Google provides.

So don’t let someone tell you that “Goofle doesn’t work that way”. It can and it will. Google Analytics is a tool for you to use as you see best to make actionable decisions for your website. Any time you go to tag something just ask yourself, “Will I ever make a marketing or design decision based on this field?” If the answer is no, then get granular!