Something Changed with the SERPS Sunday 8/14

[Bewenched]

Our site got struck down hard in the first Panda and no matter what little tweaks we did, nothing seemed to make any difference good or bad.

With that being said, yesterdays traffic spiked significantly according to analytics, not quite to pre-panda, but getting closer to it.

[thaidomain]

The same happened to our site. But also Google Analytics shows quite different results : bounce rate, time on site, keywords. All have dramatically changed, so maybe it has more to do with the Google Analytics set up that with real more visitors.

[freejung]

[analytics.blogspot.com...]

I saw the same thing last week. Check your server-side analytics/logs.

[SEOTranslator]

There was a new Panda update by Google on the 12th, when Panda was deployed to the rest of the world - perhaps they have re-tweaked the algorithm to get back unduly penalized sites. The new Panda is supposed to impact 6-9% of the sites (the first one was 12%). Perhaps you're on the lucky side... ;)

[gotmetoo]

The same happened to our site. But also Google Analytics shows quite different results : bounce rate, time on site, keywords. All have dramatically changed, so maybe it has more to do with the Google Analytics set up that with real more visitors.

Basically, in light and along the lines of the recent "improvements" Google screwed up another good thing by rendering Google Analytics unusable.

Whatever session ID "improvement" they were after failed miserably.

It is so much fun watching analytics now with a single session being chopped up into multiple visits. It was truly entertaining to find out that our site had 17 visits from google organic visitors today using a fairly common (sarcasm) keyword "one sort of widgets in one specific area stores -competitor widget term" (visitor punctuation retained).

Our saite-wide bounce went up from 24% to almost 80% in the recent few days.

New vs. Returning visitors is proportionately affected as in the example above GA reports 1 new visitor and 16 returning visitors each looking at 1 page per visit, all coming through the same source, having the same browser and resolution coming through the same network and all located in one small town with a population of under 5000, with a total time on site 00:00:00 (funny... according to internal logs this particular visitor looked though 17 pages within a total session length of just under 8 minutes).

This is not a single example. Most of our visitors are affected in the same fashion.

For those of you who is interested here is the cheerful announcement by Google staff regarding this issue. Comments are hilarious. [analytics.blogspot.com ]

Anyway. Closing my GA account today and moving to another online service.

I've had it. If they use GA stats for SERP (and thy probably do) GA is simply dangerous now. :)

[Karma]

They can still pick up all the info they need from Chrome/Google Toolbar anyway.

[gotmetoo]

They can still pick up all the info they need from Chrome/Google Toolbar anyway.

True. But we only have less than 9% Chrome users on our site for the past year. Growing at 1% a year.

Still better for us that 100% data given up to google.

[Karma]

How about Google toolbar users?

I agree on the principle, I just don't think it'll make any difference.

Edit: If like you say, GA metrics do affect serps, removing GA could have a negative impact.

[gotmetoo]

Edit: If like you say, GA metrics do affect serps, removing GA could have a negative impact.

Negative how? According to GA our site bounce rate and pegeviews-per-visitor are 90% and 1.00 respectively. Actually, if you do the math looking at our GA stats it comes out to less than one pageview per visit in most cases now. It is just seems that GA interface can not go any lower than 1:00. :-)

I am sure they will still track my stats through toolbars or Chrome browsers... my point is that I need a reliable metrics tool and GA is not it.

I also do not believe that they will fix it, as the mantra lately is that "according to our research our new improvements only negatively affect about 1% of the users".

[Lapizuli]

There seem to be two different complaints about the changes to Google Analytics:

1) They are not useful right now because of the lack of ability to compare with historical data
2) They are buggy.

The first doesn't have a big impact on me, but I'm definitely wondering about the second.

I am truly at a loss as to how to understand the new metrics. My bounce rate is up slightly, my time on page and time on site change their proportion depending on the page, and my visits are up slightly - and all that makes sense if some sessions were broken into several.

But my return visitors are way, way up. WAY up. As in, over 90% of visits are return now. And the visits are from all over, not just one area. This cannot be accurate. Or if it is, that means...no, it can't be accurate. Yikes!

[Bewenched]

I would say it was just analytics, but quantcast is showing the same results as are our own server logs,

traffic went up
bounce rate went down
sales went up

I think they tweaked something.

[rlange]

Lapizuli wrote:
But my return visitors are way, way up. WAY up. As in, over 90% of visits are return now. And the visits are from all over, not just one area. This cannot be accurate. Or if it is, that means...no, it can't be accurate. Yikes!

It's not accurate. It's another side effect of their change to session tracking in Google Analytics.

If what used to be a single session from a new visitor is now being divided into, say, three sessions, then you now have one new visit and two returning visits.

Maybe I'm not understanding the change, but what was described in the blog post doesn't sound like it should be having the effect that's being witnessed; at least, not to the degree that's being witnessed.

Google Analytics Blog
Overall, this change may slightly increase the number of visits. Based on our research, most users will see less than a 1% change.

Yet plenty of people are seeing increases of 10% or more. Are their visitors really clicking through from so many different tagged links? I find that hard to believe.

Also, the people witnessing these significant changes are seeing them from Google organic only. I'm not sure what that indicates. Do their visitors keep clicking on the same SER and Google considers each click to be from a different source, even if the search query hasn't changed?

--
Ryan

[ffctas]

Are absolute unique visitors affected by the analytics change?
Thanks

[Bewenched]

Well.. just an update, had three great days of traffic then back to the way it was.