16 replies to this topic

[AbleReach]

Are bounce rates for social media different from bounce rates of other traffic sources?

I was thinking that for new visitors the bounce rate might be lower, especially lower than from search, because someone browsing a social media site is probably not in search-and-find mode as much as they are already in a mood to look around.

Also, the link might be seen as more likely to be worthwhile - bookmarkworthy. Depending on the way the site is identified via the social media page, the linked content is sort of pre-identified as targeted and interesting information on a topic, possibly on the say-so of an authority.

This leads me to wonder if wanderers from social media are more likely to subscribe to a site's feed, for even more wandering later.

First time visitor > follows social media link > wanders or subscribes or both > wanders again later?

Edited by AbleReach, 03 November 2007 - 05:37 PM.

[A.N.Onym]

Depends on what the content is and the site (the social audience) is.

If it is a well researched article in an academic circle, then perhaps, you may expect some subscribers.

But if its an image, then expect 99.9% drop off within 10-15 seconds.

Also, the more traffic you get, the higher the bounce rate. Few ppl stop by and read. Even less subscribe.

Social media traffic is infamous for its bounce rates and lack of interest in sites. Esp Diggers.

[bwelford]

I agree. I'm intrigued by Stumbleupon. I had over 130 visitors today when one of my blog posts got in there and was presumably somewhat visible, but they seem to have left without trace. Perhaps they're trampoline experts.

[AbleReach]

Interesting. In October one of my sites got its first Stumble. That day they got about 67 visits from it. There have been a handful since then - under five. 98% of Stumble traffic was new users. Pages per visit was 1.96 - not huge but not a bounce, either.

The bounce rate for those Stumblers was 22%, which I thought was pretty good. Site average bounce rate for SE traffic is about 50%, higher when a term isn't especially targeted, lower when it is.

[iamlost]

The bounce rate for those Stumblers was 22%, which I thought was pretty good. Site average bounce rate for SE traffic is about 50%

Remember that a SE is delivering by query term(s), which may have multiple meanings, plus page title and description snippet, which may be vague or misleading, while SU is delivering by personal recommendation of actual content, supposedly. That the SE is only twice as 'bouncy' could be seen as a rather good effort.

The biggest problem with most SM 'referral' site traffic is that it is 'skim' traffic not quality traffic. They behave more like people at the market checkout reading tabloid headlines and subheads while waiting their turn. A few, if the line is long, i.e. they have the time, will open one to read a bit but rarely buy.

In that context, congratulations on getting an almost 2-page-view average. Your two-headed alien-abducted Elvis-Britney love-child articles must be really something.

AbleReach: seriously though, a 50% SE traffic bounce rate average is screaming for SEO triage attention.

...it is really hard to get a bounce rate under 20%, anything over 35% is cause for concern, 50% (above) is worrying.

By SEO triage attention, I mean that for some reason(s) those pages are not being well SE categorised vis a vis visitor expectations. This to my mind is as important as actually ranking well - there is little point in ranking well but bouncing the traffic.

[SEOigloo]

Barry -
Stumbleupon is really interesting. We had 400 visitors from one stumble awhile back. I've been trying to learn more about it since.

Elizabeth-
Congratulations. This thread got Sphunn by Marty Weintraub (a fellow whose writing I'm becoming awfully fond of, as it happens).

Iamlost-

I think your comment about ranking well but bouncing traffic is such an important one. It would be nice to see some in depth posting on bringing those bounce rates down.

Miriam

[AbleReach]

In that context, congratulations on getting an almost 2-page-view average. Your two-headed alien-abducted Elvis-Britney love-child articles must be really something.

Actually, it was an educational resource. Maybe two-headed babies are more likely to bounce?

[EGOL]

I get a steady flow of traffic from Stumbleupon. Last month we received over 4000 visitors and the number seems to be growing. Many come in through the homepage but most come in through specific content pages. The number of stumblers seems to increase as our content collection grows.

The average time on site form them is only about 20 seconds - which is much lower than our average visitor - but some stumblers do stay much longer and look at a lot of pages. I believe that it is valuable traffic because these are people who have declared an interest in our theme.

[cre8pc]

HUGE bounce rates...it amazes me how Reddit/Delicious/Stumbleupon/Sphinn drive in traffic in sometimes pleasant numbers but of this crowd, scant few become regular readers.

It's gratifying to see things like my blog being added to "Favorites" somewhere, but the time it takes and effort to make something stick is enormous.

[victor363]

One is push-based marketing, the other is pull-based marketing.


Sometimes I get suspicious that some copywriter's are good at one but bad at the other.... what do you think?

[AbleReach]

Does anyone want to guesstimate what might have a lower bounce rate?

My intuition tells me that link bait would be the worst, because people would go there for the buzz on that particular page.

[iamlost]

1. Every source of traffic has a different bounce rate. That is why to get a true understanding every page must not only be judged by it's overall bounce rate but also by the individual sources driving the traffic, i.e. various SEs (Google, Yahoo...), various SE components (G-Search, G-Images, G-Maps...), type of Social Media (blogs, IM, Fora, Bookmarking...), individual SM site (FaceBook, Flickr, Digg...).

It is quite possible to have a 66% page overall bounce rate that looks much much different on a granular pattern level, such as, 90% Digg, 50% Facebook, 30% G-search, 60% G-image, 20% Cre8asiteforums, etc.

2. Every SE query for which a page is returned will have a different bounce rate. The more irrelavant the connection between the query term and the page content the higher the bounce rate. If you have a page on 'the value of white canes for blind people' which is returned for 'the value of white cane sugar' the bounce rate likely approaches unity. Some things are not wholly within your control. Always nice to know who to blame. And if such a disconnect is widespread some rewording might be justified.

3. The more a page is designed to provide an immediate answer, including immediate conversion, i.e. ad or affiliate link, the higher the bounce rate. The more a page is designed to 'pass on' traffic to (an)other page(s) for additional information or conversion opportunities the lower the bounce rate. Finally, something within our control! And knowing the difference between good bounce (conversion), so-so bounce (external link click, no conversion) and bad bounce (no external link click, no conversion) are important statistics.

4. What you consider 'conversion' also really really matters. If for instance you consider a new feed subscriber to your blog a conversion 'your niche' Social Media will do reasonably well: i.e. Sphinn sending SEO/M types to a SEO/M blog. Digg traffic not so well. If you consider conversion an ad click SM is usually not so good. SM traffic is largely skim reading info searching traffic which usually must be transformed into some other mode before conversion is possible. Two-step traffic is a difficult ROI.

A mindset difficulty I see (frequently) is Social Media being considered as one huge amorpheous blob. Even within individual SM sites, such as FaceBook, there are differing groups each with unique responses. The traffic from each individual group within each SM site within each SM type will bounce somewhat differently on every referral to every site of every niche.

When hunting (some call it gaming) SM it is critically important to know which to let pass and which to focus on tight. A good hunter will have a significantly lower miss (bounce) percentage. He will also have a plan to that transform that SM target audience into (edible) converting traffic.

[storyspinner]

I am a big believer in knowing your target audience, and that applies to social media as well. Thus that leads to bounce rates either being extremely high or relatively low.

Most people see high bounce rates from the social news sites, because by nature, they promote "read the article and run" kind of situations.

If you or or clients are looking for success from Social Media, then target places where your audience congregates. Say for example your client sells cookware, well then write articles for the social news sites dedicated to cooking, participate in cooking forums, talk to cooking bloggers and build a profile in the cooking social networks out there (there are some!) LOL By utilizing social media sites that are more relevant to your niche, you'll lower your bounce rate.

I think gauging whether or not social media produces high click rates or not, is relative - and highly dependent on which social media sites you target.

I really believe that you need to strategize when it comes to social media, have a plan and choose the social media site you want to participate in, and have a plan for each and every one. Do the research, and know each sites pros and cons, and you're likely to have lower bounce rates.

[A.N.Onym]

That's a very good point: high bounce rates show up when the content is untargeted. I'd add that it can be of low quality and badly formatted.

The thing is, people seem to expect that social media success comes fast. It doesn't, just as with SEO or anything at all.

Obviously, you need time, effort and patience to:
- research your industry online, learn the places where your customers gather and the sources they are influenced by
- build relationships with the users on those sites, with bloggers and the readers
- understand what really interests your audience and grabs them by their hearts (or tongues, if we talk about cooking)
- learn how to format the post in the most social friendly way (Steve has a nice post about it formatting)
- actually research the post topic and create it

Then again, you can make many attempts during your progress, but ultimate success will only come after you reach the critical mass within your community.

Edited by A.N.Onym, 20 November 2007 - 01:49 AM.

[rcjordan]

High bounce rates also show up when reading interest is extremely targeted, i.e., the readers are using bump pages, feeds and widgets to cherry-pick their daily reading list.

Yesterday, Shoe published his stats on some blog widgets, I found the bounce rates interesting. No, more puzzling than interesting.

http://www.shoemoney...idget-showdown/

[storyspinner]

High bounce rates also show up when reading interest is extremely targeted, i.e., the readers are using bump pages, feeds and widgets to cherry-pick their daily reading list.

But those really aren't social media - I think (and AbleReach can correct me if i'm wrong) she was asking about the bounce rates from social media sites, i.e. like Digg, Reddit, Stumbleupon, etc. What you're pointing out are more tools - RSS Feeds, widgets, etc.

Those tools will naturally have a higher bounce rate, because those users are "tuned into" the blog/site and go and look immediately as soon as they are notified of something new - even though they are highly targeted, and then immediately leave after they are done reading the "new" page. They know the site, and likely have read a lot of other things prior, so they would be the creators of higher bounce rates.

Social media sites, if you strategize correctly and don't always count on the likes of "digg" can actually prove to be a "bouncy" (i.e. visitors view other pages).

[AbleReach]

They know the site, and likely have read a lot of other things prior, so they would be the creators of higher bounce rates.

Good point about feeds, etc,

Yes, I was talking about social media, specifically.

There is a paradox here that I think is set up by an emphasis on scoring. Where attempts to score create volume for the social media site, trying to stand out in the crowd means link bait titles will stand out more -- that, and users who are authorities.

Something "hot" or "new" with a top amount of sphinns or diggs or whatnot may get there fast by having a bait-like title. Bait encourages curiosity, which may bounce more than.... than what? Interest?

To attract fish you need something that looks like bait. Social media can do that quite nicely.
The paradox is that to catch fish you need something that acts like a hook.

Edited by AbleReach, 26 November 2007 - 06:20 PM.