Hoorah! I don’t often say hoorah, let alone publish it in a blog post. However, I’m that chuffed about the latest development in the Google Analytics interface that it needed to be said.
I had been repeatedly asking Google when they are going to release multi-touch point conversion attribution in their Analytics interface for a long time, and have kind of given up hope of it being introduced. For me, the lack of multi channel conversion tracking (or ‘conversion journey’ tracking) was the principal reason web marketers used alternative analytics providers in addition to (or instead of) Google Analytics.
That’s all changed now with the release of Google Analytics’ new Multi Channel Funnel reports. They can be found in the new version, under the My Conversions tab.
Previously, Google Analytics had attributed conversions / goals / sales to the last known visit. So if a visitor interacted with a site like this:
• Visit 1 – Clicked through from an email campaign.
• Visit 2 – 2 days later, searched on Google and found the site again via an Adwords link.
• Visit 3 – 5 days later, searched for the brand name on Google and clicked through via the organic listings, and then converted.
… Then the final organic search visit using the branded keyword would get the credit for the conversion / goal / sale. When really, all visits of journey contributed to the conversion.
The way various channels interact with each other to generate conversions can now be analysed in these multi-channel reports. It can help answer questions like:
• On average, how many visits does it take before a visitor converts on my site?
• To what extent is my Adwords campaign driving or assisting conversions which were typically attributed to branded keyword searches?
• How many days are there between first visit and conversion on my site?
• What are the most popular times of day / week / month / year for converting visitors to initially find my site? E.g. they could be converting during lunchtimes, but doing their research in the evenings?
… There are so many conversion-led questions which can be answered now.
Marketers will need to ensure that goals are setup correctly and campaigns are properly tagged (with the URL builder) to make the most of these new reports.
We’re getting stuck into the new reports here and creating some best practice multi-channel analysis reports… and generally loving it.
Hoorah.
After reading Avinash’s brilliant post about the importance of segmentation, and how all data in the aggregate is crap (read it here – it’s great) – it’s inspired us here to start segmenting and analysing ourselves. Segmenting:
- Visitors how frequently visit certain parts of websites
- Visitors who’ve provided a specific answer when registering on a site (these guys can easily be segmented using custom variables)
- Visitors with high average order values
- Visitors how buy specific products
… there are loads of ways you can segment your website visitors.
A basic marketing strategy would typically involve identifying market segments, and targeting them with specific messages. Websites and landing pages should follow this basic marketing principle too. A marketer should think about how different segments are interacting with their websites, measure this activity (which is where Google Analytics’ advanced segments and custom variable functionality comes in handy), and adapt their sites and pages accordingly.
Here’s an example of how we’ve used segments to analyse site activity.
Segmenting all converting visits allows you to understand what are those valuable pages on your site which are contributing to conversions.
We’ve also segmentated visitors based on what they’ve entered into forms, by calling the setVar function. For example, one of our clients principally have 2 categories of visitor who register on a site. In order to register, they have to specify what category they belong to. When they do this, GA drops a __utmv cookie in order to classify them, which means that data for the category as a whole will appear in the visitor reports of Google Analytics. This is great, because it means that we can then analyse at the usual GA stuff:
- Bounces
- Top content
- Goals / conversions
… all by category! More information on how to implement this can be found here.
So, think about your different types of web audience, think about how you can measure those types of web audience on the site, and start segmenting, segmenting, segmenting!
