Low Quality Traffic After The October 14 Panda Tweak?

[Uctqnde]

Hi,

I work on a website that lost a little over half of it's traffic with the April 11 Panda update.

Nothing changed until the Panda 2.5 update, when bounce rate dropped back down to Pre-Panda levels and we recovered some traffic. After a day or two it dissipated and we were back to the post-panda baseline.

October 13 we saw a massive recovery in traffic - back to, if not higher than Pre-Panda levels and it's holding.

However, the traffic is different. Our bounce rate is almost double what is was Pre-Panda and our time on site, and all engagement stats have decreased.

Is anyone else experiencing anything like this? A return of traffic, but clearly low-quality? I don't want to assume it's low-converting, because it's too early to tell, but something is definitely off and I take the approach that we have to work with what we've got (for better or worse) and I try to focus on the user above all, but I am a little befuddled with what's going on.

Any insight would be great.

[hispdcha]

I use GA and review server logs daily. It's consistent on both sides. Traffic has remained steady since the 14th, it's just after the 20th, it's been a bunch of junk traffic.

[chalkywhite]

I think something has change in the last few hours, since the 13th ive had nothing but rubbish traffic for almost non-related keywords.

Ive just check GA and i seem to be getting traffic to what "were" popular pages with the keywords that also where used before the 13th. Hopefully its and end of month rollout.

[brinked]

What I meant to say was (and sorry for not being more clear) that when my client was hit, it was the exact same day when panda was rolled out and everyone went into a frenzy. It was only natural to assume the panda update caused the 1 ranking drop. This drop made up over 40% of his traffic though so it can still be associated with panda.

To be honest, moving into a new house, dealing with other projects, I have not had much time to really study panda so I am out of the loop right now for the most part regarding panda. It would make sense though if low traffic keywords were not effected much since there just might not be anything better to offer searchers, even if it means ranking high a pandalized site. Am I making sense? Just woke up, I need some coffee.

[hispdcha]

Well, on the 13th I moved into two #1 spots for descent keywords, I've had my friends across the country check it on their end and they confirmed the move to #1. My conversions jumped and then they dropped on the 20th. The rankings have stuck, nothing that I can see has changed on our end. Same traffic levels, the traffic is just junk. :(

[Uctqnde]

@hispdcha

Interesting that your conversions jumped - I am not seeing that on my end, if anything the opposite. The traffic seems to have been junk from the start of the Oct 13/14th tweak on my end.

@brinked

Thanks for clarifying, I understand better now!

[Hissingsid]

Just reviewing my analytics data and the low quality traffic started before 14th October but since that date I have had 411 single visits each for different multi word searches where they spent zero time on site and the bounce rate was da daa! 100%.

I've looked at a few of these and the vast majority of the terms are suggestions by Google but they are things that you wouldn't make up yourself. Its as though Google has tried to read the whole of the web into a huge pyramid and has put at its foundation millions of multi word search terms that only a computer program would think of. Then if you have a page that is the source of this random selection of words you rank at #1 for that term. Now Google goes out to try and snag you a visitor by slipping that exact term into the suggestion list. I can only think that the folks who click those things suddenly gain an additional 100 on their IQ because they realise their mistake before the page headers get loaded.

Something very strange going on here.

[falsepositive]

Google gives us junk traffic in order to drive visitors to their ads and their own top ranking properties. That way, they can really control their revenue... not that hard to make record revenue each quarter now, don't you think? Just a thought.

[tedster]

If that was really what's going on, wouldn't Adwords advertisers be talking about a jump in clicks? I'm in touch with some pretty big Adwords advertisers and that's not what they're seeing.

There does seem to be something wrong with Google traffic (especially for Pandalyzed sites) right now - but it's not clear so far what's going on.

[falsepositive]

Thanks tedster, for the empirical info. Given the mess Google is in at this time, it's not surprising how even G supporters such as myself can lose their trust in search. If it's not confusing serps, it's poor traffic, no traffic or simply badly converting traffic. Throw in unstable rankings and well, the time is ripe for theories.

[Web_speed]

I've looked at a few of these and the vast majority of the terms are suggestions by Google but they are things that you wouldn't make up yourself. Its as though Google has tried to read the whole of the web into a huge pyramid and has put at its foundation millions of multi word search terms that only a computer program would think of.

Eureka, i think i got it!

I think i know now what might be causing the junk traffic with high bounce rate.
KEYWORDS F$^%^&G SUGGESTIONS gone high-wire, together with "instant" = perfect storm.

Your post made me recall that it actually happened to me more then once since bloody "instant" became mainstream. Instant is also available on the spy-bar and is on by default. Often does not function properly, eg. occasionally i am unable to go BACK one page to the SERPs after hitting enter, and often then not i find myself hitting or jumping into the wrong keywords suggestions or accidentally clicking the wrong page, very annoying really.

Now take the above and multiple by millions upon millions of daily searches and you have one hell of huge misguided pack out there all hitting "accidentally" the wrong pages = JUNK LOW QUALITY TRAFFIC.

Lets face it, things went down-hill FAST for many websites since "instant" became main stream... panda did not help much either but i have no doubt in my mind now that "instant" + "keywords suggestions" are the "real" problem causing low quality traffic many are experiencing recently. They must have changed something in that area or at least made it available on many more screens world wide at around the dates mentioned in this thread.

[Web_speed]

try this

1) Go to google.com
2) Once on google.com start typing your keywords in the search box, some suggestions will start to appear.
3) use your "down arrow" to go down the keyword suggestion list and then press the "right arrow" as if like you want to move the cursor in the search box so you can fix something you've typed....you cant, google will throw you automatically to the "i'm feeling lucky page" as soon as you touch the keyboard "right arrow" key.

And this my friends is a huge contributor to the spike in untargeted traffic on so many web sites.

They keep pumping out the idiotic features lately....its simply mind boggling.

[Hissingsid]

I think that someone at Google needs to know this and do something about it. Plus they should open a department for unexpected consequences because they are all far to enthusiastic about introducing new stuff without considering the evil that might happen by accident.

[theseanstewart]

One of my clients that wasn't affected by Panda is experiencing abnormally low conversion rates and it has been like that since Monday. The organic traffic has been consistent but the conversion rate went from around 1.90% to .6-.85%. It's very strange that something like this would suddenly happen with the same traffic.

[danijelzi]

There's no way that Google is not broken:

- for query 'widget LATE 2011 review', they deliver on the first page 7 results about widget EARLY 2011 review, with word early bolded, like it was intended search term. In one result MID-2011 was in bold and in another one (my favorite) bolded was LAST (Thursday).

- another one, for query 'widget specs' almost all first-page results are widget reveiws, with reviews in bold.

[Hissingsid]

Looks to me like they have made some mistakes in developing their thesaurus. Antonyms are the diametric opposite of what people are searching for so why have them as semantically related when trying to return useful SERPS. It beggars belief. LSI gone mad!

[storeowner]

This is the start of the third week of unusually low conversion for me. Also traffic have dipped a bit in the last couple of days.

I'm in Canada and my site is hosted in the States.

Anyone bounced back from this yet?

[Web_speed]

This is the start of the third week of unusually low conversion for me

I'd use the words "almost dead". With a minuscule number of sales trickling around Sunday night (every week) and then close to nothing the rest, week after week now. Conversions just getting worse, not better... Adwords included.

[Hissingsid]

If you take a close look at your analytics data you may be able to corroborate what I am seeing. Unusual search terms with 100% bounce rate and zero time on site. If this is the case we have found a bug in Google caused by two newish features of Google search that combine to cause mayhem.

See above.

[zdgn]

A particular conversion metric on one of my sites has always ticked like clockwork ever since the site's birth 7 years ago (site has not been pandalized - yet - knockwood.) This little conversion measure is specifically designed for human user and reflects useful, real interaction with the site (fully js based, so less bot access too.)

Around mid October, that metric has almost DIED for all Google organic traffic! I just can't figure out why and still investigating (all things apparently remain equal, traffic, pageviews, bounce, etc.)

It sure looks like it HAS TO BE poor (mis-targeted) traffic...

[Uctqnde]

@Hissingsid

I am definitely seeing what you are...very unusual search terms that we've never had traffic on before, zero time on site, etc.

It's one of the first things we noticed after the 10/14 tweak. Even though the terms are unusual and extremely long-tail, many are relevant to very specific pieces of our content, it's very bizarre.

@tedster

We are also seeing very strange behavior with our PPC...unusually high number of IMP's despite a drastic cut in spend (for various reasons, some related to the season), etc. We were seeing unusually high CTR's, but that seems to have subsided for the most part.

[Bewenched]

Still getting what I consider to be Very low quality traffic to our site and a lot of international to our US based business.

Bounce rate now is around 48% where before panda our bounce rate was around 13-24%

Now if our site was dinged by Panda for being somehow low quality wouldnt our initial bounce rate be much higher.

I should also note for anyone not familiar with me our site is nearly 14 years old and is ecommerce.

[ohno]

Snowed under at the moment but this is probably the most interesting thread I have ever read on here.

[Hissingsid]

Bounce rate for Google organic is now so high that the share of useful traffic provided by Bing and Yahoo has almost doubled.

[crobb305]

horrible bounce rate! Up from 29% to 40% since Saturday(ish)

[wingslevel]

joining this one late, but, me too. large ecommerce site conversion rate is down 15% since 10/13. my conversion rate is established over 13 years and that size move is unprecedented. virtually all of the loss in conversions comes from google organic traffic. obviously a traffic quality problem. i haven't made any usability related changes to the site during this period. overall traffic is up, so sales $ are almost flat.

[Whitey]

@wingslevel - it would be good to have a greater insight into the keyword data from folks on the threads.

Has there been a change to your core keywords specifically that lead to this?

Does anyone want to challenge or support the theory that brands are now associated with a specific bias towards high value keywords at the expense of lesser quality sites?

[snickles121]

I have to disagree about bounce rate. One page we have has the highest bounce rate of our entire site at 90%. But, it still maintains its keywords from before Panda to now. Bounce rates can be high if the user finds all the information on the page they sought and just clicked out. However bounce rate may play a factor if the user clicked back to search results and then clicked on another site because they did not find the info they were seeking.

Make sense...

[Uctqnde]

@ Whitey

We have definitely seen a difference in our keywords, especially long-tail.

We are getting traffic on keywords that we didn't prior to Panda, and not getting traffic no a large bucket of keywords we did prior to Panda too - very interesting.

I am in the process of some analysis now, so will know more in the next day or two.

[crobb305]

Bounce rates can be high if the user finds all the information on the page they sought and just clicked out. However bounce rate may play a factor if the user clicked back to search results and then clicked on another site because they did not find the info they were seeking

Yeah, we understand bounce rate. We also understand that subtle changes in index quality can affect bounce rate. For a week now, I've been observing highly irrelevant pages ranking for high-profile phrases. I can't get into specifics here, but I have hinted at it in other threads. Clearly, Google is not monitoring the index above page 4, and it is driving very poor traffic. I've also noticed a reduction in my longtail traffic. The longtail converts better for me, so this is also affecting my bounce rate. The longtail is the best traffic out there.

C

[ohno]

Very quiet today, lots of single page views. Anyone seeing similar?