If you are a marketing manager, a CEO, or a stakeholder in a company, you are probably interested in measuring the results of your marketing campaigns.
With the internet being the most popular place to find business these days, internet marketing is a must. Perhaps even more importantly, measuring the results of your internet marketing is a must.
The purpose of this post is to help you take the next step in measuring the results of your web marketing campaign.
I have titled this post “Google Analytics Guide: Features for the Advanced Web Marketer” for a couple reasons: first, I am trying to explain how Google Analytics can help marketers, specifically.
Second, I say that this post is for the “Advanced Web Marketer” not because the material in this post is extremely advanced or technical, but because I am assuming that your campaigns have progressed to a point where you need more granular data and more quantifiable results.
In this post, I assume that you are somewhat familiar with Google Analytics, and thus I will be skipping the basics of navigation, terminology, and interpreting basic metrics. I do
If you would like a more introductory post of Google Analytics, check out the link provided above. Be warned! Some of the material in this post gets extremely advanced. I’ve ordered my points by difficulty, so the most advanced material is at the end.
The paradox every Analytics user runs into sooner or later is deciphering and tying results to specific actions. This is a major theme throughout: I talk about advanced segments, which help narrow down which traffic contributed to which results.
I’ll also discuss multi-channel funnels, which can help decipher lead data and attribute conversions to definite (actually, theoretical is more accurate) sources.
1. Intelligence Events
The intelligence events section of Analytics might be the most underappreciated (yet powerful) tools the software has to offer. As an overview, Intelligence Events projects expected values for measurements like Visits, Pageviews, Average Visit Duration, Goal Conversion Rate, etc. These values are calculated with classified algorithms, but they probably involve confidence intervals (click here for an introduction to confidence intervals).
Analytics then gauges the “importance” of each change, which could more accurately be described as the certainty or validity of the change. If the change is listed as high importance (the green bar is on a scale of 0-9, with red meaning negative changes), it is less likely that it is a fluke occurrence.
How valuable are these predicted trends? This depends on what view you are in: daily fluctuations (no matter what Analytics says the “importance” of the change is) may not last, and therefore may not actually be as significant.
Try and switch to a period where your data is broad enough to accurately represent a trend. Month views are certainly the most accurate for all measures, but weekly views can come in handy for measuring short term results from a specific action.
A noteworthy feature of Intelligence Events is that each metric is also followed by a segment. These include Source, Visitor Type, Landing Page, Keyword, Medium, Territory, and more.
Application: Custom Alerts
The real gold mine in the Intelligence Events section of Google Analytics is the custom alert options. If you are a marketer who wants to monitor changes through automated reports, this tool will serve you well.
Near the top of your Google Analytics page, you’ll see a tab titled “Custom Alerts.” Click here, and click “Manage custom alerts.” From here, you can add a new Alert that can even apply to multiple views (client sites). Click the “0 other views” drop down menu to select which accounts you want this Alert to be available to.
The first thing you have to choose for your alert is the dimension. As you can see from my example, I’ve chosen “Source” as the dimension. Basically, when the “Visits” from the “Source” matching the regular expression meets the set Condition, I will receive an email alert.
As a source condition, I’ve used one of the most powerful features of Analytics called regular expressions (I promise to write a full post on regular expressions sometime in the future, for now, see this guide).
Basically, my criterion says I want to watch visits with a Source matching Google, Bing, or Yahoo (regex for this is google|bing|yahoo). Furthermore, I only want to be notified if the traffic decreases by more than 10% over last week.
There are so many other metrics you can add, and they are pretty straight forward. The Value box doesn’t have to be as complicated as a regular expression; I just wanted to demonstrate the power these alerts carry.
Here are a couple more ideas for Custom Alerts:
2. Advanced Segments and Filters
If there is one tool that makes my life easier, it is Advanced Segments. Contrary to the name, Advanced Segments are not all that advanced, and every serious Google Analytics user should already be familiar with them. If you aren’t, then read on! Your life is about to become arbitrarily better!
Advanced segments allow you to parse traffic and find very specific types of traffic. Do you want to see the number of visitors who viewed three pages and then filled out a form? There’s an advanced segment for that! Built in segments include things like non-bounce visitors, tablet traffic, single visit users, referral traffic, etc.
The main thing I use advanced segments for is segmenting out organic traffic so I can easily see the audience reports. Near the top of your screen in Google Analytics, you’ll see an arrow pointing down. Clicking on this arrow will pull up a list of advanced segments. Go to the “Built-in” section, and choose a segment to serve your needs!
Once you’ve applied a segment, the magic begins. For example, if you have demographics and interest reports setup, you can easily analyze and profile your organic visitors. This can have an immense impact on targeting, content writing, and conversion optimization. You could also check conversion rates under the Conversions → Goals section.
Here’s how you can setup an example advanced segment. Let’s say you want to find out the number of mobile conversions. Let’s also assume you want to filter out illegitimate conversions because your company only services local/specific areas, so you want to segment out all the local form fills.
Finally, you want to check the results of your SEO campaign, so you’re going to want to segment out organic traffic. Here’s what you’ll do:
First, click the “New Segment” button. You’ll want to give this a descriptive name: my example segment could be called something like “Local Mobile SE Conversions.” Next, click on the Demographics tab on the left hand side. Scroll down until you see the Location options, select Region → exactly matches → Washington.
Click on the Technology tab on the left, and near the bottom click Yes next to “Mobile (Including Tablet).” Finally, go to the Traffic Sources tab, make sure the “Filter Users” option is selected at the top, then segment out Medium → contains → organic. Click test to make sure this works! Analytics should return a percent of users and sessions that this segment affects.
3. Profile Filters
A possible alternative to advanced segments is using advanced filters. You can find view level filters by visiting the built in report view on the sidebar (think the social overview, organic keywords, or referral reports).
From there, choose a dimension to reveal visitor traits like source, exit page, hour of the day, or city. Then, apply an advanced filter to view traffic meeting a criterion from the dimension. These filters will clear if you change your reporting view; you must reapply them each time.
Even more powerful are profile level filters. These serve a slightly different purpose: they are like permanent advanced segments, and alters which data is actually gathered through tracking. While advanced segments can be applied to historical and future data, profile filters only start filtering data after you build them; you must wait for results. You must have admin level access to implement profile filters.
One of the most useful profile filter applications is removing internal traffic based on the IP address. This returns a more accurate number when it comes to visitors. In other words, you are excluding “internal” traffic so as to avoid skewing data. Go to the Admin tab found at the very top.
On the right hand side, click “Filters” and then click “+ New Filter.” From here, choose to exclude traffic from the IPaddresses that are equal to whatever your IP address may be.
You can also use regular expressions if you are in the “View Filter” mode. For an overview of regular expressions used in Google Analytics, visit this tutorial.
Stay tuned for Part 2 of this post: multi-channel attribution, analytics experiments, and more!