Enter a search term such as “mobile analytics” or browse our content using the filters above.
Check your spelling or try broadening your search.
Sorry about this, there is a problem with our search at the moment.
Please try again later.
As reported earlier, Google announced some shiny new analytics features yesterday, with 36 more to come over the course of the week.
I asked a few analytics experts for their views on the new features, which are most significant, and how they could be put to use by online businesses.
What are the biggest changes announced by Google at the summit?
Dan Barker, EBusiness Consultant:
It's still early on. Google has said it'll be announcing 50 updates at this year's summit, and ran through 14 in the first livestream broadcast alone.
They are all great from my point of view, but perhaps the biggest three that every single website could make use of are:
- Tool to migrate from 'classic' analytics to 'universal' - Google has hinted about this for a while, and I think some website owners are getting a bit fed up of the mixed messages around classic/universal. It will be great to get to a point where you can migrate across and retain your historic data.
- Demographic segmentation. The data here is not perfect, but is broadly indicative. This is the sort of data that people pay thousands for at present.
Improved educational content. This will be boring for most readers, but I really like that they're focusing more on helping people use & understand Google Analytics. They're adding walkthrough videos to the tool itself, and setting up a proper set of educational resources.
The existing Google Analytics educational resources are surprisingly good, but few people use them. Hopefully this will change that & open up the tool and the information it unlocks to far more people.
Aside from those, some big ones that won't apply to everyone, but will be huge to those who can make use of them are:
- Event tracking through tag manager a little more easily. I implement events on almost every website I work with, so this will be a big help to me. The only downside with things like this are that it moves responsibility away from the development team and across to marketers, but also moves all of the risk. Fantastic potential, but an area where people have to be a bit more careful.
- BigQuery integration to allow you to export hit-level raw data and query it directly. Funnily enough this was accidentally leaked by an Econsultancy blogger more than two years ago & I was delighted by the idea then.
- Doubleclick display information import, allowing you to view 'display' data in multi-channel reports and attribution models.
- Play Store integration. Very, very good for Android app owners, allowing them to optimise their app campaigns & optimise their Play Store presence!
API to programmatically set up/manage analytics accounts and users. This is fantastic for companies with a large range of sites (let's say an international retailer with a site focusing on each country). This will allow them to manage all accounts centrally if they choose.
Also good for consultants like me, or analytics agencies, who have their own Google Analytics templates that they can apply to new clients a little more easily.
Peter O'Neill, Founder at L3 Analytics:
There is no single killer feature, different announcements really benefit different people. Premium became more appealing with the big query integration of raw unsampled data and Google Tag Manager SLA, marketers will love the view through reporting and new ABO reports while agencies will find their lives made easier with configuration work via API.
I feel the academy could evolve into the most significant feature if it greatly expands and includes a demo account to learn with.
Damion Brown, Econsultancy trainer and web analytics consultant at Datarunsdeep:
Historically, if you wanted something tracked as an event you'd need to manually code it into the html of the page. That's fine in smaller scenarios, but in an enterprise environment it's usually going to means bugging a developer to make a change, and with staging and testing it can be a long and uncomfortable process.
This announcement suggests that Google Tag Manager will make all that a thing of the past, which will really open things up. You can track a lot more with Google Analytics if you make use of Events.
How big a deal is the introduction of demographic reports? How will you be using this?
This is the big one really from most users' point of view. Something that really gives utterly new data to every website owner. The demographic reports themselves will be useful and interesting.
On top of those though, they're adding demographic segmentation. You will be able to view virtually any report in Google Analytics, segmented down by a particular demographic criteria. The possibilities there are almost endless. There are very simple uses, but also quite sophisticated ones.
For example, this along with some other Google Analytics features would allow you to ask: "what was the rough demographic makeup of my top 10% of customers?" and then to set up a remarketing list to reach other visitors that fit that profile but did not convert, or even to slice that down further, for example limiting it to (say) visitors who viewed 5 or 6 products and the cart page but did not convert.
The demographic data integration is interesting. The value of it depends on the business questions being asked.
I'm definitely going to see if these segments display different behaviour and what insights that provides.
For my money, audience demographics will have the biggest impact for marketers. Being able to see reports in Google Analytics showing, for example, which products are selling most among 25-35 year-olds will be huge, and if you segment that further by traffic sources and social touchpoints leading to conversion, it has a lot of potential.
I also think it will give content sites a new way of looking at how male and female audience members are engaging. Before too long, the idea of looking at analytics data without a demographic filter is going to seem archaic.
Where does this demographic data come from? Given the (not provided) issue, are there privacy implications?
Google holds profile data on most users. You can see what it thinks about you, and where that data comes from here.
From what I understand you need to be using the DoubleClick version of Google Analytics to make use of this within GA. That means essentially it's passing anonymous data derived either from your Google profile or from your browsing habits across to Google Analytics from Doubeclick.
Google's story around '(not provided)' is very much around privacy, but the common perception is that there are other agendas driving it.
This makes the company look marginally more hypocritical, but they are a company of many tens of thousands of people, and already do things that step further over the line than this.
For example, I often take a look in Google Analytics' social reports, where you can often see personally identifiable data, let alone anonymous demographic information.
I believe it is based on DoubleClick cookies in which case it's OK, although I am concerned about growing reliance on these cookies. If linked to Google+ cookies, I would be really worried, as I think that would cross the line.
Hoping this isn't the thin edge of the wedge situation but I don't think it crosses any privacy lines as yet. Not provided is definitely not about privacy. And I don't believe it is about pushing people to paid search.
The demographic data comes from Doubleclick, the ad network owned by Google. This demographic data has always been collected by Doubleclick and has been available to anyone using the Google Display Network, so in a sense there's nothing strictly 'new' about the data itself.
What's new and properly exciting it its integration with all the other reports in Google Analytics.
I don't really see much of a connection to (not provided), which is a Google Search decision and not a Google Analytics decision. Privacy implications will probably be the same as for regular display advertising, but it's probably wise to check guidelines in your local jurisdiction.
How, if at all, do these changes help the average online business?
The 'classic' to 'universal' tools will affect everyone over the next few years, the improved educational content will be useful for anyone who wants to push further with analytics but is limited by the time to get over the learning curve.
Some of the changes are limited. Some of the others are limited to particular types of user.
The display attribution changes are a pilot I believe, the 'demographic' info is (I believe) only for those using the DoubleClick version of Google Analytics (this needs a very small technical change to turn on), BigQuery integration is only for Premium users, and the tag manager and API changes will only be useful to consultants like me, larger companies with dedicated developers working on analytics, or very forward thinking agencies.
If sites can get their heads around switching across to the DoubleClick version of Google Analytics, and they feel comfortable doing so, then demographic data will be the one that affects the average business most day-to-day.
And, once they've done that, they can also make use of the ridiculously powerful remarketing tools it also unlocks.
A business still needs to invest resources to get value from web analytics. With limited resources, switch to Google Tag Manager and take advantage of auto-tagging for events. Sign up for the academy and go through classes in the evenings.
You still need to define your objectives though, be asking the right questions, know how to get answers from the data and be prepared to act on these insights.
Demographics will be useful to all businesses and all websites. It's data that has always been a little out of reach for most organisations.
Aside from this though, it's probably fair to say that most of the changes are under-the-hood. They're still significant, but they relate to stuff happening behind the scenes.
The Management API for example. This lets you make bulk changes to give large groups of users access to large groups of Properties and Views without having to do them all manually.
That doesn't mean much to your average online business, but to someone that manages analytics for large organisations (yep, someone like me) it's a really significant change.