Now that my player has been on TechCrunch and Valleywag I’ve been able to measure some new things about each of their audiences:
1. Audience. Just how many people visit their page (Kyte.tv shows me how many people are online concurrently. Valleywag has been averaging about 200 to 300 people, TechCrunch averaged around 1000).
2. Engagement. How many people click on links, or comment on items. TechCrunch regularly gets more than 100 comments. Valleywag rarely gets more than 10. When TechCrunch linked to me I got 1,000 visits. When Valleywag links to me it’s rare I get more than 100.
3. Loyalty. How many subscriptions do each site have on Google Reader and other feed readers. I use the example of Gizmodo vs. Engadget. Gizmodo has about 44,000 subscribers while Engadget has 350,000, on Google Reader.
4. Influence. % of posts that show up on Techmeme, Digg, my Link Blog, Slashdot, StumbleUpon, etc.
Anyone building a new metric based on these four things? If so, we could REALLY understand a LOT more about our audiences and advertisers would have a lot better information to choose from.
I’d probably add a fifth metric:
5. Concentration of people with intent. Does your site attract a lot of people who buy digital cameras, for instance? Then it’ll make a LOT more on Google advertising. That’s one huge reason why DPReview sold for a good sum to Amazon.
Anyway, this gave me a chance to dust off my old whiteboard. Oh, on my whiteboard is the Social Media Starfish. Yes, that’s a tease. I’m writing about that for Fast Company Magazine.
[kyte.tv appKey=MarbachViewerEmbedded&uri=channels/6118/63106&embedId=10003038]
Scobleizer 
Yes, the whiteboard is back! Because online friends don’t let online friends read Valleywag.
LikeLike
Yes, the whiteboard is back! Because online friends don’t let online friends read Valleywag.
LikeLike
Hi,
I don’t understand the Kyte dynamic. I don’t see your player on those web sites.
what is the purpose of this metric, advertising, sponsorship? I agree that comments are very important. Valleywag makes it more difficult to comment which I think is because they might provoke quite a few not so nice ones.
It seems like subscribers to feeds is almost parallel to newsletters of old. I think that knowing ones ranking or activity level on other sites is important but I don’t know what’s wrong with visitors and pageviews?
I encourage you to go quantify yourself at quantcast, they offer a bunch of stats for sites that may or may not be useful / accurate but I’ve heard from site owners that once they quantify themselves and quantcast measures the numbers are surprisingly accurate.
I also don’t understand how techmeme works — i think it might be very subjective as opposed to automated. Which is fine if it is, but I think they should be honest about it or at least transparent about it.
LikeLike
Hi,
I don’t understand the Kyte dynamic. I don’t see your player on those web sites.
what is the purpose of this metric, advertising, sponsorship? I agree that comments are very important. Valleywag makes it more difficult to comment which I think is because they might provoke quite a few not so nice ones.
It seems like subscribers to feeds is almost parallel to newsletters of old. I think that knowing ones ranking or activity level on other sites is important but I don’t know what’s wrong with visitors and pageviews?
I encourage you to go quantify yourself at quantcast, they offer a bunch of stats for sites that may or may not be useful / accurate but I’ve heard from site owners that once they quantify themselves and quantcast measures the numbers are surprisingly accurate.
I also don’t understand how techmeme works — i think it might be very subjective as opposed to automated. Which is fine if it is, but I think they should be honest about it or at least transparent about it.
LikeLike
I think the whiteboard is your best prop, even when you aren’t using it much. I really liked this piece/video. Here are some observations observations.
On loyalty, all Google Reader is really measuring is loyalty among early adopters (anyone using Google Reader today is *still* an early adopter!). That said, I read Engadget regularly, and almost never read Gizmodo(and don’t subscribe to its feed), but I am an early adopter. That said, I think Google Reader is a fair argument, but it’s possible Gizmodo’s loyalty metric is understated (actual traffic analysis would probably clear that up).
#4, influence — it should be included and *is* important. But I believe with the exception of Slashdot, all those service can and are regularly “gamed”. There are sites that have “less” true influence (based on overall traffic, engagement and loyalty metrics) but appear on those services regularly. For now, I’d “underweight” this component in your overall analysis.
LikeLike
I think the whiteboard is your best prop, even when you aren’t using it much. I really liked this piece/video. Here are some observations observations.
On loyalty, all Google Reader is really measuring is loyalty among early adopters (anyone using Google Reader today is *still* an early adopter!). That said, I read Engadget regularly, and almost never read Gizmodo(and don’t subscribe to its feed), but I am an early adopter. That said, I think Google Reader is a fair argument, but it’s possible Gizmodo’s loyalty metric is understated (actual traffic analysis would probably clear that up).
#4, influence — it should be included and *is* important. But I believe with the exception of Slashdot, all those service can and are regularly “gamed”. There are sites that have “less” true influence (based on overall traffic, engagement and loyalty metrics) but appear on those services regularly. For now, I’d “underweight” this component in your overall analysis.
LikeLike
techmeme is totally automated by subjective algorithyms. 🙂
LikeLike
techmeme is totally automated by subjective algorithyms. 🙂
LikeLike
maybe the algorithms should be objective rather than subjective 😉 Or is it really true that Allen Stern and CenterNetworks (#25) is more influential than Kara Swisher on AllthingsD (#36)?
From this perspective, I’d love to hear your thoughts: which is probably more flawed for judging actual influence, Techmeme’s algorithm or my perception of true influence? I can take it if you think its my thinking that’s flawed but would love to know what you think.
LikeLike
maybe the algorithms should be objective rather than subjective 😉 Or is it really true that Allen Stern and CenterNetworks (#25) is more influential than Kara Swisher on AllthingsD (#36)?
From this perspective, I’d love to hear your thoughts: which is probably more flawed for judging actual influence, Techmeme’s algorithm or my perception of true influence? I can take it if you think its my thinking that’s flawed but would love to know what you think.
LikeLike
center networks writes a lot more often than kara does. That explains their rankings.
LikeLike
center networks writes a lot more often than kara does. That explains their rankings.
LikeLike
that makes total sense, at least as far as explaining how TechMeme works. But it doesn’t make me feel any better about the influence component of the Scoble Uber-Metric (SUM!) 😉
LikeLike
that makes total sense, at least as far as explaining how TechMeme works. But it doesn’t make me feel any better about the influence component of the Scoble Uber-Metric (SUM!) 😉
LikeLike
Robert, I would trust more Alexa.com than Google Reader stats because many people indeed never use RSS readers…
LikeLike
Robert, I would trust more Alexa.com than Google Reader stats because many people indeed never use RSS readers…
LikeLike
Well, as a shameless plug (sorry about that) I would say I wrote a rather “attractive” metric system to solve the current metric issues. It’s called Newton’s Universal Law of BLOG Attraction, and it’s all about the blogs that are attractive to me, and those that find me attractive:
http://vanelsas.wordpress.com/2007/10/09/newtons-universal-law-of-blog-attraction-better-than-a-techmeme-leaderboard/
LikeLike
Well, as a shameless plug (sorry about that) I would say I wrote a rather “attractive” metric system to solve the current metric issues. It’s called Newton’s Universal Law of BLOG Attraction, and it’s all about the blogs that are attractive to me, and those that find me attractive:
http://vanelsas.wordpress.com/2007/10/09/newtons-universal-law-of-blog-attraction-better-than-a-techmeme-leaderboard/
LikeLike
While I like TechMeme as a news resource, it’s a fairly self-fulfilling system, and I’d really be cautious before using stating that techmeme ranking and true “influence” are related. A better statement is that having a techmeme rank means one is probably influential within the technology blogging community. It doesn’t necessarily mean one’s influence has a wider span than that…
LikeLike
While I like TechMeme as a news resource, it’s a fairly self-fulfilling system, and I’d really be cautious before using stating that techmeme ranking and true “influence” are related. A better statement is that having a techmeme rank means one is probably influential within the technology blogging community. It doesn’t necessarily mean one’s influence has a wider span than that…
LikeLike
Robert Seidman – you sure do use CN a lot in your examples with regards to comparisons. Last week it was “CN can’t possibly have as much traffic as x and y” – today it’s “CN can’t be as influential as Kara” – what’s next week’s comparison?
Scoble – Writing more does not equal better results on TechMeme.
Scoble – Comparing TC to VW is like comparing a banana to a rubber tire – they are completely different products. A better comparison would be TC to RWW or GO. In addition, just because a site sends more traffic, does not make the receiving site more influential in of itself.
LikeLike
Robert Seidman – you sure do use CN a lot in your examples with regards to comparisons. Last week it was “CN can’t possibly have as much traffic as x and y” – today it’s “CN can’t be as influential as Kara” – what’s next week’s comparison?
Scoble – Writing more does not equal better results on TechMeme.
Scoble – Comparing TC to VW is like comparing a banana to a rubber tire – they are completely different products. A better comparison would be TC to RWW or GO. In addition, just because a site sends more traffic, does not make the receiving site more influential in of itself.
LikeLike
Allen Stern, all the examples were *this* week, all *two* of them. Next week I go back to not caring about web metrics. It seems like Scoble has it under control. 🙂
LikeLike
Allen Stern, all the examples were *this* week, all *two* of them. Next week I go back to not caring about web metrics. It seems like Scoble has it under control. 🙂
LikeLike
Here is my analysis of why your analysis is incorrect:
http://www.centernetworks.com/scoble-rubel-a-list-incorrect-analysis
LikeLike
Here is my analysis of why your analysis is incorrect:
http://www.centernetworks.com/scoble-rubel-a-list-incorrect-analysis
LikeLike
concentration of people with intent could be easy to measure. Just take the average click-through rate of Google Adsense units, or similar contextual units over time. The higher the CTR, the higher your “concentration of intent.”
Of course this would be relatively easy to game on low-traffic sites. You’re relying on Google’s bogus-click detection technology – a % of completed transaction would be more accurate, although more difficult to measure.
LikeLike
concentration of people with intent could be easy to measure. Just take the average click-through rate of Google Adsense units, or similar contextual units over time. The higher the CTR, the higher your “concentration of intent.”
Of course this would be relatively easy to game on low-traffic sites. You’re relying on Google’s bogus-click detection technology – a % of completed transaction would be more accurate, although more difficult to measure.
LikeLike