A look at short-term traffic

Posted on February 19, 2007.

If you didn't already know, one of my recent articles made it to the front page of Reddit and del.icio.us. People are always saying that short-term "social bookmarking" traffic is worthless, but those people are usually lying. If your aim in web publishing is anything other than the philosophical goal of pursuing inner peace, a short boost in traffic is bound to be a good thing. The real challenge is to write articles that are actually useful and draw in traffic boosts; that exercise is left up to the reader.

I thought it might be worthwhile to take a look at the effects of this short-term traffic boost, especially for anyone out there who has never had such a thing and would like to know if it is a worthwhile goal. I'm including some screen shots from the TanTan Wordpress Reports plugin, which collects data from my Google Analytics and Feedburner accounts, as well as from Feedburner's traffic analyzer.

Unique Visits

As you can see here, my site made it to the front page of Reddit on Thursday. My unique traffic jumped from 260 visitors on Wednesday to 4,740 visitors, which is an enormous increase. The following day it jumped even higher to 5,830 visitors, the result of my article making it to the front page of del.icio.us combined with the second-day traffic from Reddit. You can also see that this boost is short-lived; in the few days after that my traffic hovered around 1,500, before dropping down to 690 on Tuesday.

Average Pageviews per Visit

This graph has a lot of significance when compared to the previous one. This is the average number of pages that each visitor actually viewed before leaving my site. If this number is very close to 1, it means that a lot of visitors are simply viewing one page and leaving. For a site with hundreds of pages, it is not ideal to have visitors looking at just one of those; it makes the rest of the content seem superfluous.

Unfortunately, that's the behavior of this short-term influx of traffic from bookmarking sites like Reddit and del.icio.us; people visit your link only to read that link and then head back to the site that they came from to look for other links. As you can see in this graph, my average pageviews per visitor took a plummet when the new traffic started coming in, and went back up as the traffic went down. The challenge for publishers attracting this short-term traffic is to encourage their low-attention-span visitors to stick around, or at least get attached to the site in some way. Offering an RSS feed (and making it easy to find) is a start, but it's also useful to offer related or top articles. This happens to be something that I haven't been very good at myself, and I'm going to have to work on that in the future.

Feed Subscribers

Something you will always see with this quick traffic is an initial boost in your subscriber count that levels off after a couple days. For an explanation of why this happens, you have to look at the different clients:

The first graph is for Wednesday Jan. 31, while the second is for the following day, when the boost occurred. As you can see, the culprit here is Firefox Live Bookmarks, which Firefox users use to bookmark feeds in the browser. It acts as a feed-fetcher and reports that count, so a sudden wave of bookmarks from Firefox users will always inflate the subscriber statistics. In the following graph from Saturday, you can see that the stats have leveled off:

As for the long term effects of this traffic boost, that approximate increase of 100 subscribers due to my one popular article has remained consistent up until today, and it might just stay that way. For a sudden influx of approximately 7,000 visitors, 100 happened to stick around; anyone who says that results like that are worthless just isn't looking at the data.

Referrers, Popular Content, Entry Pages

And just so you can see that I'm not making this stuff up, here are the graphs of top referrers, popular content, and entry pages:

Here you can see the top referrers over the course of the week; Reddit, Stumbleupon, and del.icio.us, as well as quite a few accesses directly to my site.

Here you can see the most viewed pages at my site; one thing you can assume is that a good portion of that fresh inbound traffic decided to visit the "About Me" page after reading my article, which I would probably do too if I was in their shoes. Every blogger should have a good about page; it's a great way to connect with your readers and make yourself more memorable.

Finally, here are the actual entry pages (or landing pages) that people came to before browsing around. Clearly my one popular article dominated all others in the course of that week. When it comes to blogging, this is almost always the case. You are going to have that handful of articles that ring a good chord with a lot of people and become the "pillars" as Yaro so often puts it, drawing in the majority of visitors throughout the life of your site and making you semi-famous online.

So what now?

Well, I hope that gave you some insight into the world of blog traffic and visitor loyalty. Regardless of the numbers, the same rules remain true: write quality content and do what you love. Maybe I'll be seeing you on the front page of Reddit soon.

1 Trackbacks/Pingbacks

Pingback: Social Bookmark Secrets » Social Bookmarking Traffic Proof on February 24, 2007

5 Comments

Tony on February 19, 2007

I haven't made it to any social "front page"s yet, though I've seen similar results on the smaller scale when making it into the "upcoming" lists.

Also of benefit are of course the secondary backlinks that pop up like mushrooms for anything that makes it to the "front page".
Christian Montoya on February 19, 2007

Definitely; those backlinks might even be the most important benefit of all.
Yaro on February 20, 2007

Nice write up Christian and I especially like your point about a good "About Page" - so few bloggers really invest enough time in that section of the site.

As you said, the backlinks will eventually be the best benefit of this traffic event because your SEO is going to improve. You may not feel it for a few months though, but it's worth celebrating now.

Cheers,

Yaro
Jem on February 21, 2007

Although I've never managed to get "frontpaged" (I don't write for this anyway) I do find that short bursts of traffic caused by things like CSS galleries/StumbleUpon are more beneficial than people give credit for, because almost always the hits don't go down to the exact number that they were before. I certainly find myself gaining new daily/loyal visitors which causes long term (slow) traffic growth.
sigurd at silverstripe dot com on February 22, 2007

Nice stats, and correlate with an experience of ours, too. I agree that short term popularity is great; you do bulid up a percentage of long term loyal visitors. Although you do need to continue it to have a substantial long term result.

Last week our open source cms, SilverStripe was featured on Ajaxian.com's homepage (as a showcase of appropriate but innovative AJAX use), providing a ten fold increase of visitors, along with a drop in pages/visitor. We served about 45,000 pages in a day.

This quickly led to going to the front page of Delicious, then got lots of Stumble Upon visits. Then, as others blogged and posted to forums, we found ourselves with a multiple day burst of traffic, and thousands of downloads of the product.

I wrote a similar artictle, with graphs and numbers at "Surfing a big wave" on our website,
which was then noted on Oreily's Radar

In terms of long term results, Alexa.com now ranks SilverStripe far higher, and we went from being page three to the top of page 2 (i..e. result #11) for the google.com search "open source cms" (just one more higher and we'll be front page, which will be awesome!)