Subscribe to Marketing Logic
There has been a stir in the SEO community; the buzz this past year has been all about social media. Getting your site listed on
Digg, StumbleUpon, Netscape, whatever the flavor, it’s all getting to be the same. Grumblings, faint at first, have been rising against the cross-current cheerleading of these social media properties.
Social Media Blues
There are a number of reasons for discontent. Being de-listed by a few small groups, spam, and ultimately the traffic that these social media sites provide is very much like a sugar high. There is a flurry of traffic that leaves as quickly as it starts, and sometimes there is less of a result than desired. As Kim Berg recently experienced in I Don’t Digg being Dugg, it also brings out the worst of what people have to offer;
“Update: At last count, this blog post was dugg 890 times. It was submitted by someone else and subsequently, the spam comments, both here and at Digg, have been incredible. I am no fan of Digg. Never have been. This experience and the comments left here just add to my contempt for a place where people act like wild animals instead of human beings.”
Apart from the sheer volume of traffic, the link benefit might be of temporary benefit for rankings, but the amount of visitors and the comments left were not helpful to her or any of her readers.
Under the Microscope of Analytics
This caused some curiosity in me, as I have a few client sites that have been the subject of social media traffic, so, what better thing to do than to analyze the benefit of social media traffic. I did, and the results were surprising.
First, I had data from two types of websites. Then, Kim Berg allowed me to analyze her blog stats, which reflected traffic from Digg and Del.icio.us. Including Kim’s blog, all three are content-based websites. Website #1 is an information-based website that drove people to subscriptions and memberships, website #2 is a technology-based blog with no direct conversions.
I compared the traffic to each of these sites by three methods:
Source
The source of the referral; search engines, links, and social media links. Typically, search engines provide the main source of traffic to these sites. Direct navigation is comparable, but search engines by and large provided the bulk of the traffic. Links accounted for a substantial amount of traffic to the site, and the social media site provided a significant part of the link traffic.
Engagement
Using time on site, page views per visitor, and conversions, engagement of visitors would determine the logical growth for improvement of websites. However it is also useful for determining the continued investment in links from other websites, marketing programs, and other initiatives. These statistics also show the difference in site attraction for different groups based on the source of the link.
Conversions
Visitors from external site links tend to convert at a higher rate than search engines. This has been determined from analyzing many websites, more than just these three. The reason is that visitors coming from another website are better pre-qualified than those coming from search engines. Other website visitors are coming to the subject site based on a recommendation (contextual link), whereas search engine visitors come to the subject site based on a ranking or comparison to other sites in the results pages. The difference is clear and is borne out by the engagement and conversion rates.
Here are the source, engagement, and conversion comparison of search engines, links and social network site links.
Site A
Search segment #1
1.6 minutes Avg. Time on Site
2.5 Page Views per Visitor
0.1% Convert at the Newsletter Subscription Page
Search segment #2
1.3 minutes Avg. Time on Site
2.5 Page Views per Visitor
0.24% Convert at the Newsletter Subscription Page
Search segment #3
0.9 minutes Avg. Time on Site
2.2 Page Views per Visitor
0.3% Convert at the Newsletter Subscription Page
Search segment #4
1.6 minutes Avg. Time on Site
2.7 Page Views per Visitor
0.28% Convert at the Newsletter Subscription Page
Link group #1
18 sec. Avg. Time on Site
2.1 Page Views per Visitor
0.9% Convert at the Newsletter Subscription Page
Link group #2
2.7 minutes. Avg. Time on Site
3.9 Page Views per Visitor
0.8% Convert at the Newsletter Subscription Page
Link group #3
4.18 minutes Avg. Time on Site
4.0 Page Views per Visitor
1.1% Convert at the Newsletter Subscription Page
Social Media Link Group
2 sec. Avg. Time on Site
1 page view per visitor
0.0% Convert at the Newsletter Subscription Page
Site B (Technical Blog)
Search segment #1
34 sec. Avg. Time on Site
1.2 Page Views per Visitor
Search segment #2
37 sec. Avg. Time on Site
1.3 Page Views per Visitor
Search segment #3
26 sec. Avg. Time on Site
1.5 Page Views per Visitor
Link site #1
1.2 minutes. Avg. Time on Site
1.7 Page Views per Visitor
Link site #2
1.4 minutes. Avg. Time on Site
1.5 Page Views per Visitor
Link site #3
.9 minutes Avg. Time on Site
2 Page Views per Visitor
Link site #4
54 seconds Avg. Time on Site
2.2 Page Views per Visitor
Social Media Link
1.3 sec. Avg. Time on Site
1 Page View per Visitor
0 comments on socially linked article. (Stumbleupon)
Site 3: Kim Kraus Berg’s Cre8pc Blog
Search segment #1
0.2 minutes Avg. Time on Site
1.2 Page Views per Visitor
Search segment #2
0.28 minutes Avg. Time on Site
1.1 Page Views per Visitor
Search segment #3
0.21 minutes Avg. Time on Site
1.2 Page Views per Visitor
Search segment #4
0.13 minutes Avg. Time on Site
1.3 Page Views per Visitor
Link group #1
0.53 Avg. Time on Site
1.8 Page Views per Visitor
Link group #2
0.96 minutes. Avg. Time on Site
1.8 Page Views per Visitor
Link group #3
0.41 minutes Avg. Time on Site
1.2 Page Views per Visitor
Link group #4
0.68 minutes Avg. Time on Site
1.5 Page Views per Visitor
Social Media Link - Digg
0.06 Avg. Time on Site
1.1Page View per Visitor
Social Media Link – Del.icio.us
0.21 Avg. Time on Site
1.2 Page View per Visitor
This is a most derogatory piece of evidence, in my opinion.
Despite the higher rates of conversion and engagement by visitors who are referred by external site links, social media site links consistently yielded the lowest rates of engagement and no conversions. On the technology blog, thousands of Stumblupon users left no comments. On the opinion blog, the Digg users that did comment left negative replies or spam.
No Killer App
I have cautioned against declaring social media as the end-all marketing force for SEO’s. This research, though limited, confirms my suspicions. Social Media provides a “sugar-high” approach to building links, much less an online business. It provides a lot of traffic, very fast. However the vast majority of that traffic is not engaged, rarely stays for more than a few seconds and can sometimes be rude. If page views are the goal for a site, social media will provide a lot of one-off page views, but rarely more than that. Comparatively, good external links provide traffic that will view multiple pages – typically many more than social media traffic.
For a search engine marketer, social media traffic can be a proof that they know how to build quick attention for a site. However, beyond that shot of traffic, what is there to show? For those in the SEO business, there is not much else, traffic sells. For businesses that make their living on and off the web, traffic like this is not helpful. It is not direct, and the numbers show that there is even less engagement and branding impact than any other source of traffic. So besides a quick influx of visitors who don’t stick around or even read the full page, what is the benefit?
Investment v Return
I recently saw a headline “Social Media Matures,” which made me laugh. (Sorry, but it did) Sure it’s maturing; it’s in its terrible two’s. Social networks have a long way to go before they provide substantial return for the promises being made. Currently, while still in its infancy, social media is the domain of technophiles and marketers, all of which have an agenda. Among themselves, they accuse each other of developing linkbait and articles solely designed to make the front page of Digg, yet they themselves seem to be the main consumers of their own products. See Gord Hotchkiss’ clarification on this subject, research seems to indicate that the average web user does not rely on social networks for news or research more than once a month.
Building traffic to a site has always been a longer, invested approach. Marketing has always taken more than a single channel to build a good brand, customer base, site links, content, and a business. Online marketing is no different. As a point of clarification, this article is directly related to using the social bookmark sites as a method of building links and traffic, not to sites that are using social media to create engagement for their own user base. That is a different matter entirely.
Your Feedback
I will be continuing this research, as any client I have receives this type of traffic analysis. If you have any questions about the methodology, which I realize is very basic in this article, I would be very happy to answer. I would be very happy to expand my basis of evaluation.
If you have client stats (or yours) that you would like to offer as an example, I would be very happy to evaluate those as part of this activity. I would like to analyze an ecommerce site that has been the subject of any social media attention, in order to see results across a variety of website types. I would be more than willing to give someone credit for allowing me to evaluate their sites as part of a larger base of data.
If you think I’m way off base, then send your log files (I’ll be discreet), and we’ll see.
Related Posts:
Three Downsides to Social Media
Coca-Cola watches the World Pass By, Decides to Join
Overlooked Indicators
Matt Bailey is the owner and founder of SiteLogic and has over a decade in the web marketing industry. He focuses on consulting and training to help companies take control of their websites and marketing strategies.
Nice post, Matt.
I think you’ve pretty much called it - Social Media is good for driving traffic, but not much else. It is kind of fun to go in and vote for your favorites, etc., and the communal bookmarking sites such as del.icio.us and ma.gnol.ia offer a great way to share resources…
But the marketing aspect seems pretty limited. When 99% of that traffic is non-converting and/or just making rude comments, you’re really just running up your bandwidth for no great purpose.
Comment by Joe Dolson | January 24, 2007 @ 8:53 am
[...] SiteLogic - Marketing Logic » Social Media – Under the Microscope by Matt Bailey [...]
Pingback by Who wants to be a social media whore? at Baron VC | January 24, 2007 @ 2:45 pm
Great reading, and, as far as I can tell, spot on. I have been following (researching) this line of thinking for a while. I finally got tired of digg, for the reasons you mentioned, and un-dugg my rss reader yesterday. The resulting digg-free day was a welcome change. Keep up the good work, your insites have helped me out.
Comment by Rick H | January 24, 2007 @ 5:28 pm
[...] As a general advocate of all things social media, linkbait and Digg (at least from a marketing perspective), I’m inclined to stand up for the value driven by these tools. I’m also a huge fan of Kim herself, and of Matt Bailey, who also wrote about the issue on his post - Social Media Under the Microscope: [...]
Pingback by Social Media Traffic and SEO | January 25, 2007 @ 3:53 am
This is a much needed analysis Matt, thanks. What would be even more interesting is what the numbers would reveal from a range of data across 3-6 months of this kind of activity.
Comment by Lee Odden | January 25, 2007 @ 4:25 am
I knew I would have to explain this sooner or later, so here it is.
I do not think social media is worthless.
What I am advocating:
1. As a responsible marketer, I have to be honest with my clients when presenting marketing options to them. I cannot contend that social media should play a large part in the marketing mix when it obviously does not provide the same levels of engagement as search or traditional link traffic.
2. Social Media is not for everyone, all the time. For SEO’s marketing their abilities to create links and to those companies targeting early adopters, it is a perfect vehicle to reach those markets. You cannot claim that what works in marketing to early adopters will work just as well for other online business models.
3. Sheer numbers of links over a few weeks are not proof of success. For traditional businesses, leads and sales are the goal. The bottom line is the bottom line.
Sure, there are success stories, there are with every market segment and marketing style. What I found interesting is that some of those success stories end with the site being banned from Digg by the very same users that made it a success. Unfortunately, that’s too short term for myself or for my clients.
Ultimately, I am railing against those who advocate social media marketing for everyone, all the time. It is not responsible, nor is it accurate.
Big thanks to those that have left positive comments – I appreciate your feedback and I will continue to make analysis as the data becomes available. For those that think I am way off base, I’m willing to change my opinions. If you are willing to submit your log files after a social media storm, I’ll be more than happy to compare the rates of engagement and conversion with the others.
Comment by Matt Bailey | January 25, 2007 @ 8:02 am
No one from the Web is famous…
25 Web celebs? - Has there ever been a true Web celebrity? Someone who everyone knows, and they’re……
Trackback by Clicked | January 25, 2007 @ 12:29 pm
[...] SiteLogic - Marketing Logic » Social Media – Under the Microscope by Matt Bailey Getting closer to good thinking on social media measurement (tags: socialmediaanalysis analytics) [...]
Pingback by Ketcheson.net :: links for 2007-01-25 | January 25, 2007 @ 1:33 pm
[...] For hype-balance also see: Don’t Digg Being Dugg Social Media – Under the Microscope These icons link to social bookmarking sites where readers can share and discover new web pages. [...]
Pingback by Branded Content and Targeting Your Social Media Marketing Initiatives | January 25, 2007 @ 8:30 pm
[...] But mounting evidence suggests that Digg traffic in particular is less like networking with like-minded individuals at a social event and more like getting attacked by a pack of wild dogs, who leave nothing of value in their wake, other than lessons learned on closing comments and crashed servers. [...]
Pingback by Not All Traffic Is Created Equal » Publishing 2.0 | January 25, 2007 @ 9:36 pm
Great research, but I think there may be an element missing….
As you point out, external links work the best because they include context. However, this immediately leads me to 2 thoughts:
1.) The writers of the sites that included the external links weren’t made aware of the links via telepathy… Some of them surely were made aware of the content via Digg, del.icio.us, et. al. and found the content compelling enough to write about it. So Digg and del.icio.us serve a purpose of feeding editors good content ideas.
2.) You don’t say what types of sites provided the external links, but I’m guessing since they’re not search engines, many of them are likely other forms of social media (read: blogs). Yet you created a category called social media, which I’m assuming is made up of Digg, stumbleupon, del.icio.us et. al.
Common definitions of social media usually include blogs, don’t they?
Let me know if I’m reading this wrong - if blog postings were included as your “social media,” the results are particularly depressing. But if they’re not included, I’d suggest a more appropriate label than “social media.”
Comment by Marcus | January 25, 2007 @ 10:58 pm
Marcus, thanks for the feedback, I really appreciate it. The information you are asking about will be the subject of an upcoming article. To answer your question, here’s part of what I found;
When I analyzed link traffic, it included blogs, “static” site links, and social bookmarking sites. I separated out the specific social communities of Digg, Stumbleupon, and Del.icio.us, as they provided the most link traffic and they were the subject of the study. However, when I compared blog links to standard website links, there was a difference. I am expanding my analysis to account for these trends and confirm my analysis.
You are correct, however, that there is another layer to this analysis – comparing blog link traffic to traditional website link traffic. Traditionally, link traffic reads more, stays longer and converts at a higher rate than search traffic. The saving grace of search traffic is volume of visitors. My initial impression is that the social media affect will carry over to the blogs, but that would be subject to the type of blog, the market that blog targets and the context of the link.
Comment by Matt Bailey | January 25, 2007 @ 11:20 pm
[...] Additional reading: Mathew Ingram, 10e20, SiteLogic, Chip Griffin, Small Business SEM [...]
Pingback by Site Traffic And The Steady Paycheck » Webomatica | January 26, 2007 @ 5:26 am
Very interesting article. I’ve never seen hard numbers like this before, but your data is consistent with my own intuition.
One statistic you don’t show, which would also be interesting, is the percent of visitors who visit the site for very short periods (e.g. under 30 seconds). While this data would be similar to the average “Time on Site” that you report, I suspect it would provide an even clearer indication of the amount of low quality traffic coming from various sources (since it can’t distorted by the occasional outlier).
Comment by Caribbean Guy | January 27, 2007 @ 4:25 pm
[...] Once again analytics proves it worth: take a look at these figures from an interesting analysis Matt did. [...]
Pingback by The Leading Edge » Blog Archive » Social Media Traffic Might Not Be What You Need | January 28, 2007 @ 9:13 pm
For B2B products and services, the social sites seem to be of very little or no value at all. It seems like they are mostly used by lower age and lower income individuals. I could see them being useful for Pepsi, snacks, that sort of thing. For the amount of effort it takes to get qualified adults to your site, you are probably better off sending them a letter.
On the other hand, if you are doing the work yourself and you have no budget, it’s hard to go too far wrong. I can’t see paying someone to do it though.
Comment by Patrick Kilhoffer | February 6, 2007 @ 12:58 am
I guess your view is a bit to negative and generalized. Whereas you probably are right related to ecommerce and conversions one may need to contemplate that social media is social and not hard selling and that social news is related to information. So if someone spends one minute at an information source this one has got the information. If this is the target of the resource, the supplier of the information could be satisfied with the results.
Comment by mediablogger | August 29, 2007 @ 6:51 am
Interesting comment. One one hand you say that my view is negative and generalized, but then on the other you remove ecommerce and business websites from the equation in order to make your point. I would think you are being too restrictive in only wanting to view social news sites inside of a bubble.
However, the article supports social media. Blogs, forums, review sites, and CGM are all fantastic ways of marketing a website, no matter what type of site it is. The poster child of Social Media has become Digg, which is restrictive and incorrect.
Blogs, Forums, review sites, etc. are social media and some of the best sources of engaged traffic. The article is meant to be a counterpoint to the advice of “just get on Digg” and other social news sites.
Considering the context of the visitor, I’ll agree. The social news visitor that spent less than 20 seconds on the page probably did get what they wanted. You have to ask yourself if you got what you wanted.
Comment by Matt Bailey | August 29, 2007 @ 8:54 am