What happens when you're the #1 story on HN | Hacker News Reader

What happens when you're the #1 story on HN

jimplush.com

72 pointsbeagledude39 comments

So how much cold hard Adsense cash did I get yesterday? I must have cleaned up for that giant traffic spike! Alas, 4 clicks out of 30,000 page views for a total of $2.76 :) So don't expect a goldmine from Hacker News traffic.

I wish information like this was more forthcoming in other blogs like this.

When I hit HN I got almost $40 from 10,000 hits, but that was mostly because I accidentally had the link back to the homepage right above an ad unit.

"Accidentally"

It's funny when unintended changes and mistakes lead to better conversions. An example that comes to mind was PlentyOfFish's poorly resized images leading to higher profile clickthroughs since people wanted to see the images with correct aspect ratios.

The reason for low clicks could be due to the ad-blockers, which most of the users (hackers) would have installed.

that's a good point

I'm not sure, but I think that saying stuff like that can be interpreted as indirectly encouraging users to click on ads.

Interesting article and discussion here: http://news.ycombinator.com/item?id=2049105

One point of correction: It is definitely possible to see current, instant results in Google Analytics. Just click the drop down menu in the upper right and change the time period to include the present day. By default it shows the last month, not including the present day, but you can adjust the time period to show just results from today, for example.

So many people do not realize this and then say that you have to wait a day before seeing any stats in Google Analytics. On the contrary you definitely can see instant results, the interface just doesn't make this as easy as it should be.

Not the case, stats for the current day generally won't be properly up to date. See my other comment on this thread (http://news.ycombinator.com/item?id=2187209)

That really depends on the traffic your site is getting. If it's a decently large site (> 50k PVs a day), then yes, data for the current day will be lagged. But for most sites smaller than that, I've found the data to be only lagged by 15 minutes or so. It's definitely improved over the past year.

Can't speak to that, I don't pay close attention to daily stats for my personal site, and all the sites I work with are big enough to have a heavy delay.

I find there's a 20 min refresh.

I was a #1 story on HN this week too. I wrote up a postmortem here - http://natgordon.posterous.com/hackernews-post-mortem

40k pageviews to a blog entry should never crash your server.

And yet it does all the time. It looks like this guy has a typical LAMP stack running on a Linode slice, presumably running a single query to return a single record and render it as HTML.

Even assuming that all 40k pageviews came in one single hour, that's still eleven requests per second, which gives you 100ms to serve each one. That's like years in CPU time. There's no reason that should even spike the CPU off of zero. There's definitely no reason you'd need to worry about caching for a load that small.

I just don't get it.

Is LAMP really that slow out of the box?

With Google anayltics, you can see the current day's traffic. You just need to manually adjust the date range to include it.

Depends on your traffic volume, we drive enough that it takes at least 24 hours before it charts traffic, to see traffic from Monday I have to wait until early Wednesday :(

Even if you manually change the date range to today? Most people don't realize that Google Analytics defaults to "yesterday" but can be manually changed to today.

(sorry if you already know this & answered it in your previous comment. I just don't want others to miss out on GA's real time capability)

Yes, you can change the date range to include the current day, however it won't show the correct amount of traffic.

Officially they say that data won't appear until after 24 hours [1], but generally some of it will, but usually not all.

If you know the Analytics patterns of a site reasonably well (as I do my company's sites) then you can often look at the day's traffic and estimate how it's doing, but never expect it to be the final number.

(In some cases, I've actually seen traffic for the day go down after twenty four hours, once Analytics has finished working it out. I'm not sure why..)

[1] http://www.google.com/support/analytics/bin/answer.py?hl=en-...

Hmm, I have never noticed that big of a discrepancy. It seems fairly accurate for my sites. Even when my web app got 8000 hits in a few hours I was able to see them in near realtime. In my experience if the count is off, it can't be off by much.

I moniter a few sites that between them hit 20+million page views a month (~30k/hour), and can say from personal experience that, for any real idea of a day's stats from Google Analytics, you really do need to wait 24 hours.

I've noticed at least a 2 hour lag with Google Analytics.

The post mentions disappointment in viewer conversion. I think it's fairly obvious that this kind of referral is awful for conversion. Viewers are loyal to Hacker News, not to the site referred. If the post showed a promise for more interesting content in the future, then you may gain some subscribers. But when it's merely a link to something else cool found on the internet, there's little incentive to subscribe.

FWIW, I had a #1 HN story that generated maybe 100 new subscribers out of 100,000 pageviews. I was impressed, however, by how many checked out the homepage to see other posts.

PS: I find putting ads on a personal blog to be a bad idea. Unless you blog for a living, the tiny revenue you might make from it is not worth the unprofessional appearance of displaying ads.

Google Analytics might indeed be the old man in the room, but it's free. Real time analytics do matter when things get viral, but that kind of service comes at a cost. AFAIK there is no free product (as in online service) that provides such functionality with no price tag.

ObserverApp.com is free, and it has real time push notifications of every single visitor (try it, open observer in one tab and your site in another - it will increment in a matter of seconds).

The only problem is that nearly every second ajax link in the admin times out. That's been broken for a long time.

It is indeed free, and when you bother to set up the conversion goals and link to Adsense then it's as good as many paid solutions in many situations. The people ready to dismiss Google Analytics are the ones who haven't been shown how to get the most from it.

When I try to view the site, a blog comment field popups and takes over the entire page, with no method of dismissing it: http://screencast.com/t/IwfGfxcSsZG4

I'm hoping open sourcing these stats from yesterday's article will help someone out there. I've learned a ton from reading HN so here's something back.

Thanks for this. I'd always been curious what sort of traffic one could expect to get from sites like Reddit and Hacker News. Actual numbers are so rarely published.

Mind me asking what size Linode you are running EE on?

I have an EE install on Linode with nginx and MySQL, so I was wondering what I could expect?

Looks like I have the Linode 512 plan, so far so good :)

I've got a small blog network running on a 512 also -- Wordpress with supercache installed. During the Queensland floods one of the blogs got 50k visits in a day ... and I didn't notice.

Caching. Rules.

I don't think Google Adsense is good for such kind of traffic. Google Adsense works better with targeted traffic (Search Engine traffic mostly). You may want to write a blog post and to advertise a product (affiliate program) that might interest the HN crowd. If you sell 15 copies of it, with $10 commission, that's $150.

I've hit #1 a few times (or, more accurately, links to my site have - I didn't submit them) and this roughly mirrors what I encountered. Between 20-35k extra pageviews spread over a couple of days.

I've hit #2 and #5 and not had to cache my site. I was running Rails 2.3.8, thin and nginx on the cheapest slice that Slicehost sells.

Is it normal not to have to cache for HN + Reddit?

depends on concurrent connections vs spread out, cpu, traffic of other customers on your box, apache configurations, etc...

It's just good practice to cache when you can. There is no reason to ask the database for the same article 1000 times a second if it's not changing. I didn't feel the server was going to crash at that moment but I did notice MySQL was doing plenty of work it didn't need to.

The concern was if it got picked up by another site like digg or reddit then it could have been more of an issue.

Yes, I've hit the top 5 several times, and never had db load issues before, even without caching.

Has HN readership passed that of say Slashdot? HN is now my daily read but I stuck with SL for 10 yrs or so. Does HN have the equivalent of getting slashdotted...?

No.

http://www.alexa.com/siteinfo/ycombinator.com

http://www.alexa.com/siteinfo/slashdot.org

I don't think I've hit the top, but I've been on the front page of Hacker News 3 times in the past 4 days. I really should start writing about these things...

And now he is, once again, on the front page of Hacker News. The résumé you posted earlier was indeed good-looking yet very good at conveying the important information.

Reply on news.ycombinator.com