Hacker News Reader

Comment by therealarmen | Hacker News Reader

Sigh. We're drawing conclusions from Compete again?

I've said this before and I'll say it again; using Compete for quantitative traffic comparisons is flawed, particularly with social sites that utilize embeddable content. Their data has been proven wrong countless times. Until recently, Compete claimed Reddit (http://siteanalytics.compete.com/reddit.com/) had less traffic than Digg (http://siteanalytics.compete.com/digg.com/).

I don't have anything against Google+. But lets hold off on the congratulatory praise until we're sure the growth is real.

View on HN

Seconded. Compete's numbers are just made up. I know for sure they are off by more than an order of magnitude for reddit (or at least were a year ago).

Most of the traffic tracking on the web involves plugins (generally in the form of a toolbar). I don't know about you guys, but I haven't had a 3rd party toolbar in my browser interface in a long time.

Compete, Alexa, etc. they're all tracking people who don't understand technology.

I generally agree with your sentiment, but let's not get carried away: using 3rd party toolbars does not automatically render you unaware of technology.

No, but it makes for a very non-representative sample. Consider the number of people who lack the administrative rights to install a toolbar on the computer they use most often.

> Compete, Alexa, etc. they're all tracking people who don't understand technology.

That would mean that G+ is even stronger as it seems mainly popular with the tech crowd.

But it IS fair to say that the demographics of the set of users who have toolbars installed into their browser is different than those who do not.

Specifically, those that have them installed are more likely to be non-tech users than those without.

That's my guess with 0 data to back it up (but plenty of useless anecdotes!)

Compete buys samples of traffic from ISPs as its primary datasource. Your toolbar theory is incorrect.

Compete's numbers are pretty good, you just have to understand the source. Digg showed more traffic, because Compete's datasource is US ISP logs, and any site with a digg button was added to digg's traffic.

The comparing reddit/digg on compete doesn't make much sense because of the digg button penetration. Just like comparing linkedin to google plus is stupid because many google properties make requests to plus.google.com.

Compete is a valuable tool, you just need to learn how to use it appropriately. Learn when it is reasonable to compare sites using it.

That's a reasonable explanation, but it's dishonest for Compete to put these numbers on a graph labelled "unique visitors". The Google+ poster didn't have any inkling about what the numbers represent.

Agreed.

According to one of Google's own services, google+ is pretty much dead:

http://www.google.com/trends/viz?q=google+plus&graph=wee...

compare with fb:

http://www.google.com/trends/viz?q=facebook&graph=weekly...

edit: thank you downvoters for illustrating the bias on HN. A fake stats piece promoting google as #1 on the frontpage? This place is turning into /. but with more google employees. I'm out.

Ugh. You can't divine traffic numbers from search volume and news mentions.

If that were the case, Mitt Romney should start a social network because he's crushing Facebook: http://www.google.com/trends/viz?q=romney&graph=weekly_i...

lol, romney doesn't even register when compared to fb:

http://www.google.com/trends/viz?q=romney,+facebook&grap...

red=fb, blue= romney

The reason this is relevant in the discussion is that when you type in the url in chrome (arguably the top browser) the word in the domain and subdomain count as search words that show up in trends. So when you type "plus" for the g+ url it's added to trends. You can check that little fact by looking at the trends for "plus".

If you've already been on Plus, and you're in chrome or (I think) Firefox, it won't do a search when you type "plus" as the top autocomplete suggestion will be "http://plus.google.com. To search, you'd have to key down the list of suggestions.

if you look the number on the axis they are pretty much the same, but facebook didnt have any spike...

Look at the scale on your two pictures.

you obviously aren't familiar with how google trends works. The y-axis is "based on the average worldwide traffic of <search term> in all years". If you search for multiple terms, the y axis is the average of the first term. So to make a comparison, you should have included both terms: http://www.google.com/trends/?q=google+plus,+reddit

According to compete Digg is bigger that Reddit and growing.

No, it's not bigger nor growing:

http://siteanalytics.compete.com/digg.com/

http://siteanalytics.compete.com/reddit.com/

My apologies, I can't seem to put them side by side, but I'm pretty sure when I checked some time ago that was the case. I clearly isn't any more.

Uh? It's actually saying reddit is bigger (1.8m) and growing, while digg is smaller (1.3m) and shrinking.

That's what "until recently" means.

Reply on news.ycombinator.com