Google Analytics Data Sharing Options

[Randy]

And it finally comes.

Not sure how new this is but it's the first time I've seen it. This morning when I logged into GoAn I got the following statement:

In order to improve your experience with Google products, Google Analytics is updating its data sharing policy. You now have the ability to share your Analytics data with other Google services. This will improve integration, enable additional features in Google's advertising services (including Google Analytics, AdWords and AdSense) and improve your experience with these products.

Press "Accept" to enable data sharing between Google Analytics and other Google services or for additional options, choose "More data sharing options".

If you click the More Options it gives you radio buttons where you can choose to Share Data with Google Products Only, Anonymously Share Data with Google Products and the Benchmarking Service (no clear definition of what the benchmarking service is or who runs it) or Do Not Share GoAn Data.

So the day has finally come when Google wants to start using the massive amounts of data they have via Google Analytics. It's nice they're at least giving everyone a heads up and a choice in the matter, but it does leave one to wonder if they're going to start using GoAn data as part of the SERP algorithm. That specific part isn't mentioned anywhere I saw, but it does leave one to wonder.

[MaKa]

I noticed the message in GoAn too. Wonder whether I should start running a test with one of my not performing websites and some automated visitor robots...

[Randy]

That's exactly what I'm going to do with my personal site MaKa. (IMG:style_emoticons/default/wink1.gif) The benchmarking thing, which sounds like may be sending info to a 3rd party by the way the choices/exmaples are worded, sounds quite interesting to me actually. I just don't know how much info I want to give up for free. I wonder, and it didn't say, if this info is only going to be available to those who opt in or if Google is going to make it widely available to anybody via another tool.

I've been waiting for this to happen. I'll give them credit for at least telling people about it and giving the the option of opting in or out. I'm not sure they'd have to do that legally since all of the data is stored on their servers.

FWIW I went looking around a bit after I saw it this morning and apparently it's brand new. The first mention I can find of it is from yesterday in a post on the Google Analytics Blog. I expect we'll hear a lot about in in the coming days and weeks though. Both from the crowd that will like it and from the chicken littles. Fortunately I'm not in either of those camps, yet.

[MaKa]

Unfortunately I don't think I'm going to have the spare time to setup such a test at the moment. Let me know if you need a website to verify any results you are seeing.

Could be an interesting feature to see the verticals. It's a bit of a shame you have to do the sharing on account level as I've got a couple of website I would be happy to share data for, but a couple of other website where I would be a bit more hesitant.

[Randy]

I think your setup may be different than mine MaKa, if I understand you correctly.

I have different domains, each with its own account id set up under a single logon. Meaning I log on once and I can flip through them all. From what I can see from my GoAn setup it looks like I can opt each site in or out, even though they're all under the same master account.

Unless I'm missing something that is. It's all so new it's impossible to tell. Anyway, it sounds as if it might take a couple of weeks to start seeing some data. I'll try to make sure to re-visit it as soon as some info starts coming in on my test/personal site. If nothing else it'll be interesting to see what vertical market they think that rambling fits into. (IMG:style_emoticons/default/giggle.gif)

[MaKa]

Yes, I've got two accounts, one with 10+ sites and the other with 3 or so. It kind of grew that way from having to be invited to join and the limited number of sites when GoAn first launched. I'll have to remember to create a separate account for any additional websites I add.

[Randy]

Or maybe drop a note in the GoAn suggestion box that they allow a way to separate those out into accounts in a way that'll allow you to opt some into their new program, without losing all of your tracking history of course. It should be easy for them to do and may be something they never even considered since it stemmed from the GoAn limits early on.

[oneofthe3lions]

Ive had that info after log in for a week or two now.

I dont really see where the positives are against the negatives. Ok it gives you a way to see how similarly themed sites are in your industry. That seems one of the main things that it does. But surely we know our competitor sites etc in our own niches and we have tools to look at those type of things anyway?

Also these 'verticals' from previous experience are going to be so generic again, not location specific etc and therefore kinda irrelevant. I know if a site has more links than me, more traffic than me etc.

What I dont know of course is thier closing rates on enquiries for example. Or what thier bounce rate is or what thier conversion rates are. That would be the only reason that one would like to use data sharing surely? Then if you did want to do that you wouldnt want your competitors knowing those type of things about your site would you?

Im very reluctant to share any business data that is private and confidential and that is what it is. There are certain ways over recent time that Google has used peoples information in an unsatisfactory manner.

I have read with amusement many people already saying "i dont actually know what this is, but im accepting it anyway" and doing so. If you dont know the pros and cons fully then why on earth impliment it? I think a big factor in this is the 'fear factor' "Google wants it I best give it to them, send them a sitemap, my analytics, share my data else my traffic may suffer if Im not towing the line somewhere."

To much big brother for me, maybe im paranoid, maybe the USA Governent and its corporations as a whole scares me as I dont want a microchip planted in my head just yet. haha

Overall tho. I just think I will wait a few months, try to see how people are using the information and if there are actual potential benefits to using the data myself before giving my consent.

[Randy]

Waiting probably isn't a bad idea OOT3L's. At this point there is no information available as to what it'll show. I'm still waiting for the first data to show up on the two (non critical) test sites I enrolled in the program.

But that's the thing, until someone enrolls nobody else will know what information is shows and what it doesn't. If all they provide is incredibly generic information it would be a safe bet on your part that I'll be jumping ship in a heartbeat. The same goes if it's too specific. Or if they make the data free and open for everyone to see, regardless of whether each person has agreed to the data share. (IMG:style_emoticons/default/lol.gif) After all, if one can get the data without any risk, why would one agree to risk?

Being a guinea pig is something I don't mind as long as I get to choose which sites are being the guineas. On the other hand, if there aren't some significant details available, or if those details are made available regardless of one's data sharing position, I'm not dumb either. (IMG:style_emoticons/default/wink1.gif)

[Randy]

To follow up with this...

I don't see much value at all in the way they have Benchmarking set up currently. For the two sites I enrolled the only items they provide benchmarking data on are:

Visits, Bounce Rate, Pageviews, Average Time on Site, Pages per Visit and New Visits. All in the Vistors area of the dashboard.

While all of those are data you want to see about your site it doesn't really show you anyting important. As a for instance, I'd much rather see Conversion Rate comparisons so that I could get a feel for how my conversion rates stack up against the competition. FWIW I do understand this is an Optional feature in GoAn, which may be why it's not included. That said they could always exclude sites that don't have any goals set up from this section of the benchmarking.

I'd also like to see a Channel comparison. Or in their vernacular Traffic Sources. It might be good to see what percentage of visitors come from which sources and compare this to my sites. The same could be said of Visitor Trends, Browser Capabilities and Visitor Location information. Those are all stock in GoAn so should be easy for them to provide benchmarking.

Overall the info they're providing at this point just don't give me any additional insight.

The only somewhat useful data I'm getting for the two sites I enrolled in the beta is that I get a lot more traffic than my compatriots who enrolled their sites. On my personal blog site (non-ecomm) I get roughly 3-4 times more traffic than other sites Google thinks is in the same market. For the one e-comm site I get literally 10-12 times more traffic. This was to be expected because I don't rely that heavily on free search engine traffic as so many others do. However I can't confirm the Channels/Traffic Sources is the cause because that information isn't part of the Benchmark. So no insight really, just a guess on my part. Though it could be just as likely that the other sites who opted into the benchmarking beta test simply aren't any of the Top 10 sites in the field.

Following along with the above my Pageviews are exponentially higher than the competition, as one would expect. My Bounce Rate, Pages per Visit and Average Time on Site are right in line with everyone else. A little better, but nothing too far out of line one way or another. New Visits percentage was slightly lower than the competition, which is just fine by me because it means my traffic is more loyal and comes back more often. Even though I'm getting dramatically more traffic, and thus possibly less focused traffic, than they are.

My final assessment at this time is the Benchmarking data isn't worth much, if anything at all. I don't see anything in the data I'm getting that would cause me to take action in changing something. Interesting perhaps, but not actionable. So if you're on the fence about giving up your data to Google I'd say don't do it. There's not anything earth shattering to be gained right now.

I'm going to leave my couple of sites in the test and try to find a place to make some recommendations to whatever team is working on the Benchmarking project. If they can get more useful data in there it might be worthwhile. However at this point in time there's nothing I see to get excited over.

[Karri]

I saw this new feature a few days ago when checking some analytics for a client.

How are the competitive categories (or "verticals" per GoAn) set up? Even though it's aggregate data, what if the category is super niche (as in the case of my client using GoAn)? Wouldn't this info be a tad 'too' revealing for most business' comfort level?

I haven't seen how it all looks in action yet but really curious ...

Karri

[Randy]

I assume they do the verticals by the keyword phrases they see the sites getting hits through Karri. But they don't say that outright, so who knows.

I'm not that concerned about Google getting their hands on the data, because they already have it. They could certainly use it in many, many ways and we'd never be able to tell. If this was a concern of mine (and it is, which is why only two test sites are set up with GoAn) I simply wouldn't use GoAn at all.

On the benchmarking stuff the worry for me would be how the data is shared with others. And from everything I can see in my test sites there's nothing in there that remotely identifies what other sites are signed up.

[MaKa]

I decided to add my sites to the Benchmarking anyways, and I found pretty much the same as Randy. Its interesting to see, but you won't get any actionable results from it.

PS. I saw that there finally is an option to see your stats grouped weekly or monthly rather than only daily. Yeah!

[Randy]

I'm still holding out some hope MaKa.

Mainly because benchmarking is so new that I doubt many GoAn subscribers are using it, and because I'm hopeful they'll eventually add some more useful information. Say like conversion rates.

It's not a bad start and I've seen nothing like it elsewhere that relies on real data, but it still has a long way to go. I have a feeling they may have left conversion rate data out of the mix early on because too many people would have balked at allowing anyone else to see that sort of info. Well, that and not many seem to understand how to set up real goals, and because goals can vary sooooooo much.

As an example of that last one one of the sites I've included has 1 Goal that is basically sales driven. Buy something and the goal conversion is recorded. However I also have a section of the site that offers some stuff for free. Doesn't even require registration to get the free stuff. I have that set up as another Goal so that I can tell how many are using my free stuff.

The difference between the two on the conversion rate side of things is pretty dramatic. Literally ~70% of all new visitors go to the free area, so count as a conversion of that goal. A smaller, though still significant percentage convert on the Buy side of things. Many of those convert for both of my Goals. FTR, there are actually other goals I have set up, including a breakdown of how many subscribe to month-to-month as compared to annual or lifteime pre-pay and so on. This just confuses the numbers more since nobody looking at the benchmarks would know the difference in my goals setup.

This type of difference between goals could throw a significant amount of skew into the picture, so they'd have to come up with some way of controlling this to provide reasonably accurate data. I'm not even sure how they could do this in an automatic fashion.

[Karri]

I assume they do the verticals by the keyword phrases they see the sites getting hits through Karri. But they don't say that outright, so who knows.

I guess that is what I'm worried about ... the keyword phrases are pretty niche, searched on, but niche. Then again, if someone wanted to know what phrases a page is targeting it's not that hard to figure out anyhoo ...