Segmentation is one of the most important features of any web analytics package, and Google Analytics is no exception. If you’re regularly looking at your overall site stats and user behaviour, then you’re looking at all visitors bundled together. Do you think it’s appropriate to be looking at visitors who are engaged and loving your site in the same set as visitors who bounce? Do you think you’re learning anything from the average?
There is a huge number of ways to use Google Analytics Advanced Segments to your advantage when you’re trying to improve your site or your marketing. You can separate visitors who convert from visitors who don’t and look for patterns of behaviour. You can remove bounces from your stats to get a better look at how many pages all other visitors view. You can directly compare visitors from one marketing source against another to find out if either is providing you more value. The list goes on. Below is a short video by the Google Analytics team on how to create and use Advanced segments.
Advanced Segment: “New to my site”
The Advanced Segment I want to talk about today is perhaps the most important segment in the arsenal of an online marketer, the “New to my site” segment. We use this on a near-daily basis when evaluating the performance of paid search campaigns or SEO to get a quick and easy look at how important each traffic source has been for driving new customer acquisition.
What we’re going to do is to try to remove visitors who already know about you, as those aren’t people you’re targeting with your online marketing efforts. We’re trying to exclude the folks who were going to reach your site whether you made any efforts to bring them back or not.
Setting up the Advanced Segment
The first step in any analysis of your online marketing is campaign tagging. This is a bit outside the scope of this article, so I’m going to assume you’ve got all your online traffic sources wonderfully tagged according to Google guidelines on campaign tagging.
Once that’s in place it’s time to set up the segment. There are three main parts to this process.
- Exclude returning visitors. These people already know about you since they’ve visited you before. Don’t look at these visits when trying to analyse your online marketing efforts.
- Exclude brand searches. Anybody searching on your brand has already heard about you elsewhere. Your PPC/SEO etc have not been the major factor in acquiring that visitor. The trick here is to exclude every variant people might use. If your name is easily misspelled then include a shorter version or several misspellings. Include something that will be matched by a person searching for your domain name. In these days of search boxes and omniboxes people can easily search for your domain by accident.
- Exclude direct traffic. If somebody can type in your web address then they clearly already know who you are. You haven’t reached a new potential customer here. If your traffic is correctly tagged then anything from a valid marketing source (e.g. email) should not show up in this category.
Analyzing the results with a Custom Report
The easiest and best way to view the results of this sort of segment is to use a custom report. There is one obvious type of report that unfortunately isn’t included by default (yet) in Google Analytics, and that’s the ability to show goal completions by traffic source.
So we’re going to create a new custom report that lets us:
- Visits
- Bounce Rate
- Pageviews
- Goal Completions
- Goal Conversion Rate
- Cost per Goal Conversion
- Compare between source/medium combinations, and drill down to different campaigns.
View the following metrics all in one table:
Add any other metrics that are particularly important to your KPIs: pages/visit, e-comerce revenue, etc.
The big picture Marketing Mix
This segment isn’t perfect. In these days of multi-channel funnels and multi-touch attribution, a simple segment like this isn’t going to give you a super accurate picture of how people are visiting your website. But for quickly and easily seeing a current snapshot of the big picture marketing mix, this is going to give you more honest results than what you’d otherwise see. No more brand searches making PPC and SEO look amazing. No more returning visitors making remarketing look like your best channel.
To analyse this kind of data you essentially want to do a straight comparison at a high level between the different traffic sources to work out exactly what kind of return on spend you’re achieving. Once you have an idea how profitable each source is at generating new visitors you can start to apportion budget between them.
Related Articles
- The Choice is Stark: Segment or Die by Avinash Kaushik
- Audience Segmentation – Giving Clicks a Personality by Emer Kirrane
- Test Segmentation for Higher Conversion Rates by Daniel Waisberg