Bounce rate from Yahoo is 40%, from Google is 70% - is this realistic?

Hi everyone,

According to Google Analytics when Google is the referrer (organic traffic) then the bounce rate has now reached 70% (used to be 42%). Meanwhile Yahoo traffic (about 25% of visitors of what Google brings in) has a bounce rate of 40%.

How come the large discrepancy and is this even realistic? How often is this statistics truly updated in GA?


I sometimes wonder if people are more purposeful when they search with Yahoo as opposed to searching for almost anything on google.

some time it happen with my website also i also wonder about this but i think this should be near about 45%-50% is fine

hehe…in other words you reckon people using Yahoo are usually more intelligent and know what they’re looking for rather than just clicking through whatever’s thrown at them?

Did you notice whether both tools were referring traffic from the same sources or from different? This should clarify your issue.

is the traffic is noted individually by these two search engine?

I’m not sure what you mean. I’m using Google Analytics. There at I look at traffic courses and amongst all the other referrers seem Google and Yahoo. Yahoo provides about 20-25% of the traffic that Google does, but traffic coming from Yahoo has a bounce rate of circa 40% while the same organic traffic is coming in via Google has a bounce rate of circa 70%.

I’m on page 1 of SERP’s in both these search engines, in Yahoo at position 2 last time I checked for a main theme keyword (which brings in the traffic) while in Google (for the same exact keyword) position 4.

As Yahoo and Google rank sites differently, maybe visitors from Google and Yahoo are coming via different search phrases (did you check?). Maybe the Google search phrases are less relevant to your site than Yahoo’s, leading to a higher bounce rate.

Or maybe the average Google user is more likely to suffer from ADHD than the average Yahoo user.

Over what time period do you see this discrepency? If it’s just a one off, it could just be a random happening.

Are you using Adword campaigns? If you are, perhaps people are clicking through via adwords then leaving. Don’t get too concerned about bounce rate. Some very profitable sites have a bounce rate in the 90%. The way the site is set up, means 6-10% buy!

Sorry if this sounds obvious to you but where exactly in Google Analytics can I deduct both the keywords that are driving traffic AND their referrer? So far I can only find a list of keywords that drives traffic, there isn’t a way as far as I see to add a filter that filters out what the referrer is each time.

Speaking of keywords. There’s a lot of long-tail keywords I’ve not even specifically ranked for (never wrote any content both on and off-site to target these) that have very often produced 1 visit that’s 100% bounce rate. I don’t even think this was a human visit since the visit time on these is usually 0 seconds let alone more. Instinct tells me these are web-bots, but if so why on earth are they searching for keywords that have virtually non-existent traffic?

If I drill down into top exit pages then I can see that both \ and \index (the main page) yield 31.95% and 25.42% respectively.

Just by flicking through the list of top exit pages in GA I would take a guesstimate to be an overall average bounce rate of say 40-50%. Unfortunately some of these pages don’t have enough traffic yet to say my estimate can be normalised.

Maybe I’m barking up the wrong tree here but what I think is happening is that certain pages are ranked higher in Google but not necessarily representative what people are looking for. As most people are lazy and nervous instead of clicking through the website (where they will probably find what they need anyway) they just click the browser’s “back” button or type in a different url. As a result the bounce rate in Google is artificially high even though, as stated above, GA reports a good if not potentially healthy low bounce rate for the main page.

That’s my theory anyway, seems logical but it would seem more time is of the essence before I can be sure (at least in respect to the above theory).

Now, if I think about the “how can I overcome this obstacle” my own thinking is to go for some more volume keywords, both on and off-site via articles, get some rich backlinks thus link juice, increase SERP rankings thus funnel more traffic and then see what happens. I can optimise SEO for new content pages but as for the lots of long-tail keywords Google is sending me micro traffic for, these are such adhoc terms that the traffic volume must be very low to not warrant further effort (even though the content is helpful to a specific problem, just people call it by different names and so it’s a game of cat and mouse).

It’s all organic traffic, no PPC campaigns running currently.

You guys probably think I’ve answered my own question :slight_smile: Possibly, as I say though it’s only my swing on things and I could be right, wrong, very wrong or just darn silly arriving at such conclusions :slight_smile: Hence with second opinions are always welcome.


It’s not obvious :slight_smile:

You’d need to create a custom report (Custom Reporting > Manage Custom Reports > Create New Custom Report). I’d drag metrics like Unique Visitors and Bounce Rate (both in the Site Usage section) onto the graph and Dimensions of Source and Keyword (in that order). Then click Create Report. The unique visitor count will give you an idea of whether the sample size is big enough to be significant. If UVs are only say 2 or 3 you can ignore those as the sample size is too small to draw any meaningful conclusions.

You can chop the custom report around to try and get more perspectives on your trffic, the above is just a starter for 10.

When you click on google, for example, to see all the keywords that brought googlers, you can export the entire list to Excel. If you do the same with Yahoo, you can then work with those 2 sets of data in the same spreadsheet if that makes it any easier to compare results.

They’re probably people, not bots, I would imagine.

If a visitor visits page 1 and then page 2, the GA code is executed for each, and 2 timestamps are recorded. The time spent on page 1 is calculated by subtracting the first timestamp from the second. If, however, the visitor leaves the site after visiting only page 1 - i.e. they “bounce” - no second timestamp is recorded and GA sets the time spent on page to be 0. Bounces have a time spent of 0.

Just to give you a positive solution; I’ll post. Tom Sant is a professional proposal writer. He has written some excellent books on communication. He has helped write proposals equaling 11 billion in sales. He teaches corporations how to write. Seeing that a web site is a proposal then you really need to know the common factors that cause people to bounce from your content. It could be that your keywords are attractive enough but your content is not deep enough to solve their problems. It sounds to me like you have succeed in communicating through key words but somewhere you have dropped the ball on followup with good content. Numbers are also the essence. You may preach up a storm but only a few repent even something as good as eternal life. Eternal life is a good key word. Thousands look it up. But if they expect me to tell them about Budah - they’ll bounce like bunnies.

Fare well to you brother!

Thanks for that, actually quite intuitive once one knows where to look.

Overall average bounce rate to date for the major search engines is:
Google - 70.14%
Yahoo - 32.53%
Bing - 40.00%
AOL - 83.33% (not really enough traffic to judge)
Ask - 50.00% (not really enough traffic to judge)

Overall bounce rate over time via looking at the chart seems to be on a downward trend.

Traffic came in through 332 sources + keywords, the vast majority have very little traffic to pull out any honest verdict. Where there has been enough traffic to judge, those keywords have a healthy bounce rate of circa 25%.

Looks like more traffic has to surface before making a judgement call. Back to building link juice and both on/off-site content.

Thanks for that. I’ll take that onboard, attraction marketing I believe it’s called right? Anyhow, thanks again. Part of the problem is that some of these problems for which people search for answers are somewhat complex. Unless you give them an answer in three short sentences they get fed up and bounce. People crave simple solutions to built-up problems, if only there was a magic wand to explain that. How to build an atomic bomb in three easy steps :slight_smile: