January 25th, 2007

Not All Traffic Is Created Equal

by

We all bow down to the gods of traffic, but the reality is that not all traffic is created equal. Perhaps if you’re selling junk page views to an ad network for rock-bottom prices, it’s all the same, but for most pages on most “quality” content sites, some kinds of traffic are definitely better than others.

Digg and other social news sites have captured the imagination of publishers because of the massive amounts of referral traffic that Digg in particular can drive, leading to the obvious comparisons to Google and search engine traffic. Danny Sullivan points out that, in terms of raw traffic, sites like Digg seem to beat out non-Google search engines. (TechCrunch’s referral sources was an important reference point.)

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.

SiteLogic has a great, detailed analysis of how traffic from different sources behaves on three different sites. The third analysis, of Kim Kraus Berg’s CRE8PC, has specific data on Digg vs. Del.icio.us vs. link traffic vs. search traffic. The difference is striking.

Digg referrals spent an average of 3.6 seconds on the site, compared to link referrals that spend an average of one minute or more. Search and Del.icio.us traffic don’t spend much time, but Digg’s average is only 25-50% of the other averages.

Kim herself wrote a post titled “I Don’t Digg Being Dugg”, where she observes what many others have –Digg users are abusive and generally uncivilized:

Another way of looking at it is this. You take a walk through a park and quietly enjoy it and the experience. Perhaps you will recommend it to someone else. Or, you can visit the park and leave graffiti all over the benches, paths, and toss toilet paper into the tree branches.

Many sites, including the New York Times, have put “submit to Digg” buttons on their content pages, figuring that they need to bow down to the traffic gods like everyone else. But it appears that the Digg traffic god is not a benevolent god.

At some point sites are going to start to discriminate among traffic sources in terms of quality rather than quantity. Sure, if you have advertisers who are willing to blindly monetize any page view, it might not seem to matter on the surface. But I wouldn’t count on that lack of discrimination to last.

Comments (55 Responses so far)

  1. [IMG] Scott Karp / Publishing 2.0: Not All Traffic Is Created Equal

  2. Perhaps if you’re selling junk page views to an ad network for rock-bottom prices, it’s all the same, but for most pages on most “quality” content sites … Source: Publishing 2.0 Author: Scott Karp Link: http://publishing2.com/2007/01/25/not-all-traffic-is… Techmeme permalink

  3. diggel kapcsolatos elemzések

  4. Bloggers like readers. Politicians like press. Non-profits like donors. Businesses like prospects and love customers. So any traffic you get for your blog or web site is good, right? Not according to Scott Karp

  5. what can you promise your clients that will pay off for them in the long run? Counter point – Social Media Traffic Isn’t Useless by Rand Fishkin. More perspective – Techmeme vs. Digg – Knowing Your Audience & Client by Liana Evans More: Not All Traffic Is Created Equal by Scott Karp Beyond Google: Social Media Engines First, Other Search Engines Second by Danny Sullivan The Digging Debate: Is Social Media Worthless? by Tamar Weinberg Social Media Too Sweet For Websites

  6. I’m looking forward to those quality links many of my friends have promised will certainly come, because there’s a tropical island I fully intend to move to with all the millions I make from the conversions. Sadly, Tamar, the hint you dropped about my not being male didn’t get past their slick radar. Do you think it matters that all this marketing and attention about me or this blog is accurate? Yep. That

  7. Target YOUR Audience: Pick your audience and appeal to their interests. And while not all traffic is created equal you do not have to target Digg if you are looking for clients, it is easy to bias your demand toward the target market. If you can come up for reasons people would want to talk about you then you will get more exposure than you can handle. There is

  8. dream that they constantly work toward. If they do so, massive amounts of traffic will come their way. From this, the theory goes, readership, brand awareness, ad revenue, etc., should all shoot through the roof. But will it really? Scott Karp states: Not All Traffic Is Created Equal. He looks at metrics from around the web, with some interesting tid-bits of information: “Digg referrals spent an average of 3.6 seconds on the site, compared to link referrals that spend an average of one minute or more. Search and Del.icio.us

  9. My friend and occasional sparring partner Scott Karp over at Publishing 2.0 has a good post today about the transitory nature of the Digg “flash crowd” phenomenon — the point being, as he puts it, that “not all traffic is created equal.” Scott points to a study at SiteLogic that looked at traffic flows to several sites, including Kim Krause Berg’s Cre8pc, before and after they had been “dugg.” The conclusion, not surprisingly, confirms what many people have been saying since

  10. assumption is correct, that the ratio is indeed 1:10 then the raw numbers would lead you to assume that 5000 unique visitors are better than 500 unique visitors. Using ‘time spent viewing the content’ as a metric leads us to a different conclusion. Not all traffic is the same

  11. ads. Visitors who found your site via search are looking for something and, heck, it just might come up in a contextual ad. Visitors who came via social bookmarking sites (on average) are a different class of user – they chew up bandwidth, spend a few seconds (3.6 seconds on average, to be precise) on your page, post an inane comment, and move on. They can deal with some ads to offset the trouble they’re causing. Sure, Digg users aren’t going to click on ads, but at least they’ll contribute to the CPM ad

  12. Attention Metrics: Going Beyond Numbers A little over two months ago, I asked the question: should we measure traffic in minutes? The question was prompted by Scott Karp’s observation that not all traffic is created equal, and my own conclusion that consequently, 5000 quick visitors from one source are not as useful as 500 visitors from another. A few days ago, Compete answered my question. Towards the end of my previous post I asked,

  13. new subscribers. However, you better have a beefy server or Digg could shut you down. Darren Rowse wrote a great post with tips on building a digg culture on your website, so I won’t repeat all of that here, but I do recommend that you read it. Scott Karp has a differing view saying: “…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

  14. Which brings me full circle. The link to the story about the iPod and its Digg effect couldn’t be accessed last night; the server has been brought down by all the Digg users trying to link to it. On this issue, Scott Karp points out that Digg can do more harm than good: 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,

  15. off from most of the possibilities inherent in the Internets. See, I’m becoming a web2topian too! Although I still think that digg, while the idea behind it is cool (people vote on the news!), is almost entirely useless. Even worse than useless, wrote the estimable Scott Karp a while back: … 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

  16. off from most of the possibilities inherent in the Internets. See, I’m becoming a web2topian too! Although I still think that digg, while the idea behind it is cool (people vote on the news!), is almost entirely useless. Even worse than useless, wrote the estimable Scott Karp a while back: … 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

  17. Not all web referrals are equal

  18. Scott,

    You make a great point. Advertisers are learning to discriminate more and more about the quality of audiences that they receive from publishers, and they are already adjusting the rates that they will pay for those audiences accordingly. This will certainly trickle down to publishers’ allocation of resources to various link referrers and tagging sources. They will discriminate more and focus more on those that send them the best people. Certainly quality will matter just as much as quantity. For example, if you are a US-based site with US-based ad sales, why would you spend time generating links that bring you traffic from Brazil?

    Dave

  19. I’d noticed this back when I used to run AdSense. Can’t say my site was every a heavily trafficked one, but every once in a while I’d get linked to by a big site (never Digg though) and witness up to 4000 visitors, up from my usual 100-200 or so. The funny thing was that despite a 20 fold increase in traffic, I wouldn’t get any more ad clicks than on any other day. The only thing that my number of clicks really correlated with was search engine traffic.

  20. [...] of the Digg “flash crowd” phenomenon — the point being, as he puts it, that “not all traffic is created equal.” Scott points to a study at SiteLogic that looked at traffic flows to several sites, including Kim [...]

  21. [...] MrBabyMan Not All Traffic Is Created Equal [...]

  22. If your advertising model is based on a per-impression model (CPM), Digg is good for you. If your advertising model is based on a per action or per click model (CPA/CPC) such as AdSense or afilliate links than Digg traffic is bad for you.

    Search traffic does tend to run high in regards to conversions (people buying or leaving via an advertising link). Digg can sometimes give you a few extra advertising clicks but it tends to be less than 0.001%.

  23. Great point. Completely agree that “At some point sites are going to start discriminate among traffic sources in terms of quality rather than quantity”. I suspect the availability of consistent, reliable, granular metrics of user engagement is one of the variables that will drive this transition. What are the other critical variables?

  24. HMTKSteve,

    If your advertising model is based on a per-impression model (CPM), Digg is good for you.

    Only if your advetisers want to reach young, male (94%), geeks.

    gz,

    Just basic demographics is a big variable.

  25. Wow…94% male? There is something a bit sad about that.

  26. [...] a bit of discussion concerning the traffic from Digg versus from other social sites. I thought I’d toss in two cents as I’ve had experience with traffic spikes from [...]

  27. Rand Fishkin gave a great presentation about the dynamics of getting dugg at the last Pubcon. His findings – Digg users don’t ever click on ads, and they rarely come back. So why would anybody want to get Dugg? Because Digg users will link to you in droves.

  28. Scott,

    I said “good for you” but that does not necessarily mean it is also good for your advertisers!

  29. Good point you have raised there. Getting Dugg may not actually be all that beneficial as we expect it to be. Will those visitors return, is a question that needs to be asked.

  30. [...] of the comments — and I have a fucking thick skin. Nice to know by reading Publishing 2.0’s Not All Traffic Is Created Equal that it’s not just me having a thin-skinned moment. (Thanks [...]

  31. Diggbait!

    It’s nice to have a rigorous analysis, but I think this is cast against something of a strawman. Of course advertisers know all about raw audience numbers vs. target demographics, and one is not the same as the other.

    The issue is that most of us get neither :-(.

  32. [...] a page, however, doesn’t mean that Digg users won’t make their presence felt. Karp continues in his piece: [M]ounting evidence suggests that Digg traffic in particular is less like networking with [...]

  33. [...] Publishing 2.0: Not All Traffic Is Created Equal “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 (tags: digg comments community metrics traffic del.icio.us) [...]

  34. Lawrence, do you have examples of Dugg articles that have received lots of links from Digg users?

    Prof. Daga, you can see Digg’s demographics here.

    HMTKSteve, good for you but not good for your advertisers is a market imperfection and it won’t last.

  35. [...] digg it Not All Traffic Is Created Equal [...]

  36. [...] Flutterby? It looks like other sites are learning the lesson all over again: Violet blue pointed to Not All Traffic Is Created Equal, which gives even more reasons that you might not want to be [...]

  37. [...] Not All Traffic Is Created Equal » Publishing 2.0 (tags: advertising CPA CPM contextula audience traffic) [...]

  38. That’s about all I needed to hear about it. Who needs readers like that. Or are they just trolls?

  39. [...] bring server-crushing loads of traffic to a site, but readers coming from Digg spend, on average, 3.6 seconds on a site before they bounce off to the next flavor of the moment. Either they’re really fast [...]

  40. [...] YOUR Audience: Pick your audience and appeal to their interests. And while not all traffic is created equal you do not have to target Digg if you are looking for clients, it is easy to bias your demand [...]

  41. Interesting post, useful information. Funny the post still has “Submit to Digg” link below it. :)

  42. [...] * Update: if you’ve read this far you’ll probably like a post called “Not all traffic is created equal” from publishing 2.0. Technorati tags: news, newspapers, aggregators, digg, del.ico.us, reddit, [...]

  43. regarding the “submit to digg” button still on the post:

    I don’t think the problem here is with Digg, or the traffic from Digg, or even the whole of the Digg community-I’m all but sure that the majority of the Digg reading community don’t comment on the posts (I can vouch for this at least personally-most Digg users go for news and interesting content, not for the “discussion”) but instead look for interesting media and topics and then head over to those links that interest them.

    I think that the traffic isn’t the bad part, it’s the segment of the Digg community that tends to go with the traffic that’s problematic. By and large the folks who comment on Digg are..well..everything that’s already been said and is widely understood. The rest of the folks, who don’t generally go to Digg for anything but news and linkworthy content, are just fine.

  44. [...] “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,” says Scott Karp at Publishing 2.0. [...]

  45. [...] Bad traffic – the sudden spikes of traffic that occur when a small blog is mentioned at one of the big sites may look impressive, but it’s not as useful as a smaller quantity of regular readers. [...]

  46. Good post, but it’s not clear to me why it’s bad to get an occasional surge of readers from a source like Digg. Unless it brings your servers to a crashing halt, you’ll almost always find a few new marginal readers among the hordes from this source, and at no cost. It’s no different than circulation-building techniques: Your most desirable subscribers are the ones who renew time and again with little or no prompting (or marketing expense). Next are those originally referred by complementary, desirable sources. But to meet your numbers, you may have to resort to “cheap” sources occasionally. They give you a rush of trials, and you accept the fact that they won’t convert as well. It costs you little or nothing and you won’t spend anything trying to convert them later. “Bad” traffic will go away on its own, so no need to take action against it now.

  47. [...] Karp states: Not All Traffic Is Created Equal. He looks at metrics from around the web, with some interesting tid-bits of [...]

  48. [...] Not All Traffic is Created Equal – Publishing 2.0 [...]

  49. [...] “Not All Traffic Is Created Equal [...]

  50. [...] Not All Traffic Is Created Equal Publishing 2.0 Why getting dugg isn’t all it’s cracked up to be [...]

  51. [...] bringing down your servers, isn’t especially good for having people spend time looking at your site.http://publishing2.com/2007/01/25/not-all-traffic-is-created-equal/ Mike Gunderloy Tuesday, February 13, 2007 Deleting… Approving… Other recent [...]

  52. [...] It was also interesting to note some other stats on Digg traffic from Scott Karp. [...]

  53. [...] with out knowing who you are measuring first?Also, not all traffic is created equal….. http://publishing2.com/2007/01/25/not-all-traffic-is-created-equal/ Todd: Sure. There should also be a Unified Theory of Everything but they haven’t cracked it [...]

  54. [...] to a minute spent on the site, the traffic numbers don’t mean much (i.e. not all traffic is created equal). Furthermore, the rise of new web technologies such as AJAX which don’t require page reloads [...]

  55. [...] to a minute spent on the site, the traffic numbers don’t mean much (i.e. not all traffic is created equal). Furthermore, the rise of new web technologies such as AJAX which don’t require page reloads [...]

Add Your Comment

Subscribe

Receive new posts by email