Analysts, marketers and decision makers need insight into the online behavior of customers and users over periods of time and across multiple devices, while most analysis packages are based on measuring individual sessions on a single device. The engineers at Google are busy measuring visitor behavior over multiple visits and devices with a new version of Google Analytics.
Main Engineering Paul Muret shared his views of the distant future of Google Analytics during the Google Analytics User Conference in Amsterdam.
This distant future soon could become reality. Insiders talk about a launch in the fourth quarter of this year. There has even been a successful pilot of this new method of measurement on Youtube and Youtube Channels.
What are the changes?
Google makes the new version of Google Analytics 'Customer Centric Analytics platform’, with a central role for the user. The new version of Google Analytics has the following two major innovations:
·Visitors are measured with a unique user ID instead of multiple cookies. This would make the new platform the first to monitor individual users across multiple devices and sources.
Google Analytics users can themselves add external data, integrate it with web statistics output and design the output of the reports.
Measuring across multiple devices and sources
The online purchase process is now very diverse and consumers often use multiple devices during the orientation process. Here’s an example of the journey of a multi-device customer:
On Monday Judith notices an offer for a holiday to Portugal when browsing on her computer at work. In the evening she shows the offer to her husband and children on her iPad. They also check out the sights around the hotel in Portugal on her iPad. During the weekend they decide to book the trip on the laptop. An hour later, Judith opens the e-mail confirmation on her iPhone, and downloads the voucher.
Most web analytics packages register 4 unique visitors in the above exampleon 4 different devices, while there is really just one unique visitor. With the addition of a new methodology with a unique ID, a user can be better monitored, across devices.
Disadvantages of measurement through cookies
To measure visitor behavior Google Analytics now drops at least 3 or 4 cookies. On each page view the cookies are being read, updated the information is sent to Google. The Google Analytics servers receive and process this information for millions of websites simultaneously, which even for Google is a fairly intensive load.
In addition, we know very well that in the European Union, by introducing the new cookie legislations, measuring cookies will become increasingly difficult and less reliable.
A unique visitor ID instead of cookies
The solution to both problems that Google is working is a visitor ID. They didn’t share how they plan to pull this off. The visitor ID can be generated by Google, but it is also possible for a website owner to determine their own unique feature within Google Analytics and and add visitor ID’s to them. Examples include an ID of users logged on, a campaign code or other unique code
Add external data processing at Google
Google will also allow us to add external data to the data set and link to it with web analytics data, or other data from a CRM database. This makes conversion attribution easier and allow for deeper market segmentation based on visitor behavior. Google will only offer this functionality as long as we host the data on the Google Servers.
On May 1st, Google launched Big Query, enabling businesses to benefit from the computing power of the Google servers for complex calculations.
Center Parcs in the Netherlands had the scoop. They have deployed their reservations system and marketing activities to the server for faster analysis.
The expectation is that a link is created between Google Analytics and Big Query. "Because the data processing takes place afterwards, the user can shape the information as he wishes." Said Paul Muret. This means that Google can provide customized reports. Also, users can define data processing and visualization configuration and customization.
The new capabilities also raise additional questions. The changes will demand a certain level of online maturity from users. Analyzing with the new tool is a lot more complex than before. Planning a useful datastructure will become a lot more challenging.
For companies that do not want to use the new method and possibilities, it remains possible to follow the current way of online measurement.
This article is a translation from the Dutch article written by Wilbert Pot on webanalisten.nl.