Earlier quoted context omitted.
You ought to include those in the data set.
You should include those in the data set.
Is it "should"? I don't like that word. Are there other options?
You should follow me on Twitter
51–60 of 75 posts
Re: You should follow me on Twitter
#52I like his articles, but this one is silly. 4 Messages on 4 blogposts and he's drawing conclusions? That little data doesn't warrant any conclusions: too many uncontrolled variables. Results might as well be random. Audience, and expected clickthrough rate, per post topic may differ, the effect of repeating the implorement several times kicks in and if I bothered to give it some more thought, I could probably come up…
i wasn't under the impression that it was 4 messages on 4 blogposts? i thought it was a universal include? but then it would only additionally account for archived posts..
Re: You should follow me on Twitter
#53This is fantastic. Just curious - did you run any tests for statistical significance?
Assuming a truly random sample of 5000 unique, non-overlapping viewers for each experiment, the standard error for the experiments runs from .3% to .5%. The true value for any given sample is pretty likely (> 95%) to be within +/-2 standard errors. So in this case, the difference needs to be more than 0.6%~1.0%, depending on the experiments you're comparing. In other words, these look like they're significant differe…
Do you exclude traffic and clicks received from HN ?
Re: You should follow me on Twitter
#54Re: You should follow me on Twitter
#55I like his articles, but this one is silly. 4 Messages on 4 blogposts and he's drawing conclusions? That little data doesn't warrant any conclusions: too many uncontrolled variables. Results might as well be random. Audience, and expected clickthrough rate, per post topic may differ, the effect of repeating the implorement several times kicks in and if I bothered to give it some more thought, I could probably come up…
i wasn't under the impression that it was 4 messages on 4 blogposts? i thought it was a universal include? but then it would only additionally account for archived posts..
I only count 14 articles on http://dustincurtis.com/index.html, of which the first bunch doesn't contain any of these Twitter 'invites'. Neither does his blog. Am I overlooking something?
Re: You should follow me on Twitter
#56His blog has appealing typography and layout.
Re: You should follow me on Twitter
#57One thing this doesn't measure is quality. Only quantity. Don't quite know how you can measure that though.
Right. I'd like to believe the people you really want reading your stuff aren't so suggestible. But I think is research for commercial sites.
Re: You should follow me on Twitter
#58I cannot stop myself from doing a plug for my startup Wingify (http://www.wingify.com/) here :)
We have an A/B, split and multivariate testing platform ready to use. With features such as running tests specific to a visitor segment (say organic) and having multiple goals (say CTR and time spent), the platform is quite powerful. So, if any of the fellow HNers wants to setup a similar experiment, I would be really glad to create an account right away.
By the way, it would also be interesting to see how would call to actions rank if you segment traffic into these buckets: regular blog readers, new visitors, visitors from twitter. My hunch is that traffic from twitter would have behaved quite differently from regular blog readers.
Re: You should follow me on Twitter
#59I've read in the past that telling people to do something isn't as effective as telling them that everyone else is doing it. I wonder how a phrase like "most people who like my blog follow me on twitter" would fare?
how about "Follow me on twitter here. I have x followers."?
But some people are going to respond negatively to large numbers (or at least very large). e.g.: if I see "I have 28,437 followers" I'm likely to think: okay, this is probably some Perez Hilton type person.
But if the info could be tailored more (e.g., "2,437 HN subscribers that follow me") I would be more interested. In other words, not everyone finds following the most massive crowd a sign of value.
Re: You should follow me on Twitter
#60Dustin continues to impress me. Wondering had he try more 'counter-intuitive' suggestions like:"You must not follow me at Twitter"?