The Techmeme Myth

Is Techmeme influential, a driver of traffic and a unbiased source of the best tech news online? The tech meme aggregator founded by Gabe Rivera in 2005 has long been held up as one of the best sites in the tech blogosphere. As a business, Rivera has done extremely well from it, charging a rumored $3000-$4000 per ad on the site, while remaining a one man setup.

But is it really that great based on evidence? BusinessWeek named Rivera in its list of The 25 Most Influential People on the Web in September with the headline The Traffic Driver: Gabe Rivera. As best we can on available information, we would suggest that the influence, traffic and impartial nature of Techmeme is a myth.

Foreword

The moment I so much as mention Techmeme in a post I will be accused of having an agenda, so lets start with the key points. I was a long term reader of Techmeme, and my first blog was indexed by the site. Over the years I spent a lot of time on Techmeme, and even had the pleasure of meeting Gabe in person last year for first time. Then something strange happened; I stopped getting headlines on Techmeme. More details can be found in Life after Techmeme, but what I missed in that post was noting (along with my falling out with Michael Arrington) that at the same time I stopped getting headlines on Techmeme, I wrote a post (which actually made Techmeme) titled Techmeme and the Noise Problem. I don’t believe in coincidences.

We’ve had a few stray headlines since that time, but at no where near the rate before. Rivera claimed publicly that nothing had changed, and that The Inquisitr had stopped writing posts that were worthy of inclusion, a claim backed by others (one Canadian writer with a long term agenda against me came in hard). As it turned out, our last full month with Techmeme headlines was our worst month on record. The Inquisitr has actually averaged 300-400% more traffic then we did that month (sometimes higher) and has had more links than it ever has. If anything, we are better off without the Techmeme links.

Rivera has continued to accuse me of being “a liar” both publicly and privately ever since, but has never once provided proof that the weight given to Inquisitr headlines wasn’t changed. Around 1-2 weeks ago (I can’t be sure of the timing as it was pointed out to me by more than one person) The Inquisitr was completely excluded from Techmeme, not even appearing as a link-in source on stories. Gabe may argue that this is untrue, and I only have my own observations from this week to go on; he may also reinclude The Inquisitr to prove this to be untrue, so it comes with that proviso.

My simple agenda is that I dislike Rivera’s statements that Techmeme is unbiased and is highly influential, when neither is true. My personal issues aside, this post will consider facts, with some commentary where required. I cannot stop some people saying that this is some sort of personal vendetta (I’d note that if it was, it was started by Rivera), but I can present facts and lets others draw conclusions from them.

Myth: Techmeme is Influential

The idea of Techmeme being influential comes from its use by some high end blogs and VCs. That some people of influence read Techmeme cannot be disputed, but extending that out to influential isn’t supported by numbers.

Here’s Techmeme’s traffic stats from a range of site tracking tools:

Quantcast: 7,188 visits per month/ site rank of 91,617
Compete: 291,621 people
comScore: no record of traffic (to small to measure)
Alexa: rank 119,812 with 1.2 pages per visitor

Compete is the odd one out (high use of the Compete toolbar or plugin perhaps), but even if we take a middle line, the influence is small. Techmeme doesn’t do a lot of page views. Consider that 1.2 pages a visitor on Alexa would also indicate search engine traffic as opposed to regular engagement; if it were influential more people would be refreshing Techmeme regularly because they are on the site checking for the latest headlines.

Publicly available statistics aren’t always brilliant, but a site that is said to be so highly influential should on these measures be ranked far higher than it is, by most objective measures of a websites worth. The numbers don’t support the myth.

Myth: Techmeme drives traffic (and how YHacker News beats it)

BusinessWeek called Rivera “the traffic driver” based on the myth of influence, but the numbers don’t stack up. I’m yet to talk to a person who can claim to have gained huge amounts of traffic from Techmeme, but that’s anecdotal of course and not based on facts. So lets look at some facts from this site.

The biggest amount of traffic ever delivered to The Inquisitr from Techmeme before we were excluded was 1,112 page views on a single post. In that time we had a reasonable list of headlines, some of them top of the page, some of them further down. The total amount of traffic from Techmeme during that time into the leading 10 posts (by traffic from Techmeme) is 5122 page views, or 512 page views per post. The distribution of numbers varies. There was one post (mentioned) with more than 1000 page views, three posts between 736-766 page views, and the remaining posts ranged from 158-400 pages views.

Sound like a big traffic driver? Digg and Reddit have peaked in 5 figures for this site, although Reddit does smaller numbers on average (but regularly low to mid 4 figures), but that’s probably not a fair comparison. What is a better comparison is the very niche YHacker News site.

On every publicly available measurement, YHacker News is smaller in traffic than Techmeme, and yet you wouldn’t be able to tell from the numbers they deliver

Our top ten posts by traffic delivered from YHacker News totalled 7073 page views. The top post came in at 2639 page views, more than double the highest figure from Techmeme. The second and third posts fell just short of 1000 page views (998 and 995) and the remaining posts ranged from 242 to 469.

So is Techmeme really a traffic driver compared to YHacker News? On the evidence we would suggest no. YHacker News may have a smaller readership, but their influence is extended through the engagement its readers have with content on the site, something Techmeme doesn’t match.

Other sites may have different numbers from Techmeme, some higher, some lower, but even taking that into account, a niche site with significantly less traffic statistically delivers more traffic than the “traffic driver” that is Techmeme. On the evidence at hand, the numbers don’t support the myth.

Myth: Techmeme is not biased

Rivera has long claimed that Techmeme is an unbiased tracker of leading tech news. I’ve already mentioned our improved traffic and link spread post Techmeme earlier in this post, so I won’t revisit our situation, but we can look at other sites for evidence of bias.

The concept around Crunchmeme, or that TechCrunch gets unfavorable treatment in Techmeme isn’t new, and the number of headlines TechCrunch gets on Techmeme (backed by the Techmeme Leaderboard) anecdotally supports the idea. The real question is that do numbers support the notion that there a bias towards TechCrunch; and likewise is it harder for other sites to get a headline on Techmeme.

The obvious comparison is Mashable, TechCrunch’s largest competitor, and a site (and team) hated by the crew at TechCrunch, people Rivera has always been very close to (the two times I’ve met Rivera were in Michael Arrington’s house). As of today, Mashable does not feature in the Top 100 blogs on the Techmeme Leaderboard, but this wasn’t always the case. Historically, at least according to TechCrunch’s Blogger Board, Mashable ranks at 33. So what happened to Mashable?

According to public measures of traffic, Mashable is competitive with TechCrunch on traffic. Quantcast places Mashable ahead on traffic, but Alexa, Compete and comScore has TechCrunch in the lead, but only by a small margin.

Mashable has had 9 headlines on Techmeme over September and October (at the time of writing) vs TechCrunch’s 245 headlines.

In terms of incoming links, Mashable ranks 14th on Technorati to TechCrunch’s 3rd. Technorati’s authority score which counts incoming links over the last 6 months ranks TechCrunch at just over double of Mashable’s score. From this we can establish that TechCrunch gets more links, but at a ratio of around 2.3:1 vs Mashable. The headline split for Techmeme is 27.2:1.

But does TechCrunch have an easier path to Techmeme than Mashable, even allowing for the link profiles?

We checked 20 random TechCrunch headlines from September and October (random as we could be, but I didn’t close my eyes and click). Of 20 tested headlines, 2 had no visible incoming links shown on Techmeme, 15 had 1 incoming link, and 3 had more than one link showing.

Here’s an example of a story on Techmeme today with no visible links. Rivera likes to claim that other links are considered but not shown, and that makes some sense, given according to Technorati, the only site linking to this post that isn’t a spam blog (as the time of writing) is The Inquisitr through our best of the web section on the front page πŸ™‚

But how many links do Mashable stories need to hit the front page when they do? Of the 9 posts Mashable had on Techmeme in September and October, the average number of links shown was 3. No posts had no shown links, 2 has 1 link (including one post where The Inquisitr was the only site linking in visible), 1 headline had 2, 2 headlines had 3, 3 headlines had 4, and one had 5.

For Mashable at least, getting a headline, if and when they are able to, involved more sites running the story as shown on Techmeme.

It should be noted that not all shown links link to the headline stories, Techmeme groups stories based on context as well, and doesn’t necessarily give the headline to the first person who posts, or the site with the most incoming links. So Techmeme has to pick who gets the headline when multiple sites are writing about the same story, but not necessarily linking to each other.

Here’s the split again: TechCrunch 245 posts vs 9 from Mashable. That doesn’t allow for topics that both would have covered, but there is sufficient overlap on big stories to suggest that the overlap would be around 50%, so lets try that again:

Where there is an estimated 50% overlap of coverage, the split goes TechCrunch 123 vs Mashable 5 (both rounded up).

Now lets take TechCrunch’s treatment on Techmeme out of the consideration, and compare Mashable with another site featured on Techmeme regularly: Silicon Alley Insider (SAI). I’m not suggesting that SAI doesn’t deserve to be there, Henry Blogdet and his team are doing some great work and should be up there. But we can use the site in comparison to Mashable because it’s currently in third spot on the Techmeme Leaderboard (CNet news is 2nd, but it’s not strictly a blog, so it’s not as well compared).

At the time of writing, SAI has had 118 headlines on Techmeme for September/ October 2008 compared to Mashable’s 9. On links, SAI has a Technorati authority score of 4556 (it’s ranked 41st overall) and in traffic, the various services place SAI at one third to one half of Mashable’s traffic; Quantcast has the split SAI 650,000 to Mashable 1.4 million, and both are direct measures from each site (although Quantcast does under-report numbers a little bit in my experience.)

Even if we exclude the idea that Techmeme is biased towards TechCrunch, the stats prove that Techmeme is biased against Mashable, even if we allow for a wide ranging margin of error in the stats shown here.

Rivera’s line that Techmeme is not biased doesn’t stack up. He weighs the importance of sites, punishing sites and people he doesn’t like, while favoring those sites and people he does like.

Conclusion

It saddens me that Techmeme has come to this point. I’ve spent more hours than I care to mention on the site over the last 4 years, and it will always hold a small amount of affection in my memories. But enough is enough. This idea that Techmeme is the great influencer, the bringer of traffic, and an unbiased source of leading tech news isn’t supported by facts. The great Techmeme myth is a fairytale of spin; you don’t have to take my word for it, look at the numbers.