Performancing Services

July 2007 Performancing Experiment: A Statistical Analysis of Cost/Benefit for StumbleUpon Submissions

Experiment: Submit every new Performancing article to StumbleUpon by at least 5 people during a 2 week period and evaluate the traffic benefit (will you help?).

You can track the experiment by monitoring pMetrics statistics for Performancing.

Having run Nudger.org in Alpha for a few months, it has become increasingly clear to me that human tendency is to pursue the simplest solution with the biggest perceived payoff. My goal here is to change your mind about the biggest perceived payoff in social voting, and to change your promotion behaviors (just like I’m starting to change my own).

When bloggers are choosing whether to submit to Digg, Reddit, StumbleUpon or Netscape, and they only have one choice, they almost universally go for Reddit.

I want to change this tendency because I am confident that it is the wrong approach. Normal Reddit traffic is borderline useless, except to artificially boost up your stats and maybe get an adrenaline rush. Reddit traffic doesn’t convert nearly as well as search traffic, and hardly any better than Digg (which is notorious for having poor ad conversion).

One hypothesis is that, all things being equal, there is a measurably better cost/benefit reward ratio for a steady dose of Stumble/Netscape then a steady dose of Reddit/Digg. Yet, I also believe that SEOs everywhere, including myself, are inclined to go with the Reddit/Digg combo or maybe just the Reddit solitaire. Why? Well, in the case of Digg it is so damn popular. In the case of Reddit, it is so damn easy (and I mean easy…)

The problem, as I see it, is that the simplicity of Reddit can become addictive and often provides no lasting effect. If you have an average blog you are probably getting somewhere between 75 and 400 visitors per day. Reddit can singlehandedly double or triple your traffic in any given day. But this traffic is largely junk traffic.

Netscape has the benefit of building strong, deep links over time. Some of my blogs have long-term PR 4 or 5 links straight from the Netscape domain. Many articles get short-term PR 6 and 7 link love. While you won’t drive much traffic from Netscape, you will help drive SERPs over the long haul. Netscape should be part of anybody’s long term link building strategy

Stumble traffic, in contrast to Reddit, is very focused and converts quite well. Sure, the initial submission is more time consuming, but then it’s all “thumbs-up” from there. Plus, I’ve found that a single Stumble can send thousands of visitors to your site. No need to get on the front page or anything. Just get a few people to “thumbs up” your article, and you’ll have sticky readers with an interest in your site topic flow right for high conversion rates.

So my hypothesis is this: Reddit has the least payoff of all the major social voting sites, but that it’s also the one solution that most people use. Instead of using Reddit, I think that a wiser use of your time is to go after topic-targeted traffic with StumbleUpon. It doesn’t hurt to do both, but avoid the temptation of using Reddit for all your social voting needs.

Author: ryancaldwell

8 thoughts on “July 2007 Performancing Experiment: A Statistical Analysis of Cost/Benefit for StumbleUpon Submissions

  1. Ryan, just remember that StumbleUpon nofollows all links from their site. There’s no link love benefit in that respect.

    In terms of traffic, I’ve watched StumbleUpon for a while now, and I believe that the most traffic is gained by good tagging of submissions and also the popularity of the topic. I think there’s a tag cloud on the site that shows you popular topics and of course there’s Stumble Buzz, which shows you what’s hot at the moment on StumbleUpon.

    I wrote an article a few weeks ago that gave me about 18,000 pageviews over about 7 days. I believe most of the success of that was down to extensive – and appropriate – tagging. I think the tags are the way StumbleUpon decides what’s relevant for you.

  2. That’s because Stumble readers are *focused* – Stumble sends people that it thinks will be interested in your topic.

  3. I’ve submitted stories to Reddit, Digg and Stumble but Stumble worked the best.
    Reddit sent me all of one visitor. Digg send a few dozen but non stayed longer then a few seconds. Visitors from Stumble read at least two articles.

  4. Brett,
    That’s a great idea. First five volunteers get free stumbles at their sites.

  5. So once you get 5 volunteers, will you let us know when the experiment begins?

    Might I suggest that you use sites provided by each of the volunteers also. Say 5 other sites to submit as well and then the comparison can be made with a basket of sites in addition to Performancing.

    Oh, btw count me in.

  6. I’ve been using StumbleUpon and Netscape for about 2 or three months and get some good traffic from the two. Digg and Reddit are pretty much useless for my website.

  7. I’ll need five volunteers to commit to Stumble submissions and/or votes over the next 2 weeks.

Comments are closed.