HN Explorer

Ask PG: What's the traffic effect here from the TechCrunch mention?

49 points by robg ↗ HN
Would you mind quantifying some numbers based on the TC front page treatment - before and after?

Pageviews? New users? Submissions? Comments? Anything else interesting?

22 comments

[ 3.0 ms ] story [ 57.3 ms ] thread
pg
We got 14k unique ips yesterday, which without TC would have been 10k. The effect of the spike was diminished by being spread over 2 days, because we use GMT days for stats.

The most dramatic change was the number of new accounts. There were 258 new accounts in the last 24 hours. On a typical day there are 50-60.

The most interesting thing to me was that the TC traffic exposed a design flaw in the account creation code. That's why the site got so slow yesterday.

http://news.ycombinator.com/item?id=133804

Xichekolas
Huh, not near as bad as I expected. For some reason I pictured a hoard of 10k new users posting nothing but xkcd and penny arcade.
engtech
would that be so wrong?

:)

harbinjer
Well, 1 or 2 isn't bad, but eventually it would become noise, not signal.

I don't think that that goal of this site is quantity of users or stories, but quality.

Xichekolas
10k new users wouldn't be so wrong...

10k new users posting xkcd and penny arcade comics would be. I can go to the xkcd/PA site to view the comics when I feel like it. I can go to fark/reddit/digg for humor. There is no need to make this site equal in content to them. I'm not saying they are bad, but I don't see the point in duplication.

Zak
10k new users might be a problem, in the same way that the flood of new Internet users from AOL was referred to as the "September that never ended". Any time users flow in to a community too quickly, it is difficult to bring them all in to compliance with the social norms of the community, resulting in a change in those social norms which the existing community members usually do not like. It's one of the reasons people get upset about immigration.
dcurtis
Oh. Well that's far less damaging that I was expecting. So far, no influx of techcrunch commenters, that I can see.

Thanks for sharing.

jotto
I think this is an opportunity for us to inspire an even bigger community of the benefits of grounded, esoteric conversation. Let's not worry about short term damage, but try to teach and motivate the newcomers and avoid creating an unwelcoming community.
webwright
Hopefully this will help dispel the myth that TechCrunch coverage is a "get-coverage-or-die" issue for startups. Getting on it is great, and the splash coverage from it is important.

But 4,000 uniques just isn't going to do it unless your site/service is TRULY viral. If it is, then you should be able to take off with 100 uniques just as well (if a little more slowly).

pg
Being on TC isn't just important for the numbers. It's who those people are. I suspect practically every startup investor reads TC, for example. That's not a factor for News.YC, but for the average startup those are good people to have know about you.

Also, the traffic boost is probably ordinarily more than 4k uniques. We probably have a fairly big overlap with TC readers.

webwright
Traffic boost is not normally much more than 4k in my experience. I've had two startups on TC (multiple times each) and worked at a third startup that also got coverage. Ranged between 3.5k and 7k uniques (so you're right-- it's on the low side).

Regarding investors, I think you might be right there-- TC coverage probably helped our YC app, for example (you'd know more than I!). But, in conversations with investors, I've NEVER had one say "Oh yeah-- I saw you guys on TechCrunch." It's a nice bullet point in an email/execsum that lends credibility.

Arrington literally has talked about founders calling him crying and begging for coverage and busting into his house to plead with him. That just seems wrong to me.

thorax
To provide an opposing anecdote, our startup literally got direct contacts/calls from VCs specifically due to our coverage on TechCrunch.

We had been front page on Digg and that got us a crazy number of uniques, but the initial VC interest did not start until we were on TechCrunch and Lifehacker a week or two later.

konsl
We've been covered on TC twice (without any other significant coverage on the same day):

We saw an additional ~2200 UVs for the site on the first post; ~12000 UVs for the widget (incl in post) and ~1000 UVs for the site on the second post.

To this day, we still have people e-mailing us (as a result of TC coverage) -- usually they want you use their service.

mriley
I suspect there was a decent amount of overlap between techcrunch readers and news.yc readers prior to the article. I wonder what the spike numbers are for the yc startups that have articles written about them when they first release.
wschroter
I've had a few stories I've tracked on the TC effect.

Recently they wrote about my company, GotCast.com, and we had about the same reaction as Paul saw.

Before that I wrote a story for TC about a company in the photo sharing space (Jan of last year) and they had significantly more traffic from it.

I think anymore the spike is a lot less noticeable than people give it credit for.

wumi
and does anybody have numbers on coverage in news.yc?
alex_c
Last October, about 1400 visits spread over 3 days (half of it on the first day).

Obviously it varies based on a number of factors (I'm sure an YC start-up launch announcement gets a lot more clicks, for example).

whacked_new
"On a typical day there are 50-60."

Which means there are tens of thousands of accounts? I'm amazed that the community still feels small here. That's some crazy lurker:commenter ratio.

pg
There are

  > (len hpasswords*) 
  14357
accounts. Not all active, obviously.
engtech
I just took a look at my referrers for the past year on my blog. Without a doubt 90% of referrers are social bookmarking sites like this and reddit.

It might be the psychology of it all. You visit a social bookmarking site to find interesting links; you visit a blog to see what they have to say more than click on the links.

michaelneale
I hope its not the beginning of the enditt.