Hacker News new | past | comments | ask | show | jobs | submit login
How much traffic being on a Google doodle gives (grok.se)
91 points by nhoss2 on Sept 16, 2011 | hide | past | favorite | 34 comments



I run a website that was the first result after clicking on a Google doodle. Here is our traffic spike:

http://i.imgur.com/HAyjo.png

It is hard to see from that screenshot, but on an average day we get 5-20k visitors. The day of the doodle we got 142k and sales/revenue spiked like crazy.


Buckyball? Did the doodle just link to a google search of the word buckyball?


Yep, exactly.


As a comparison, the logo for Freddie Mercury's birthday resulted in 1.2M Wikipedia pageviews on the day, and ~500k the day after: http://stats.grok.se/en/latest/Freddie_Mercury

Jorge Luis Borges? 2.0M -- but the day after (perhaps this is a time zone thing?) http://stats.grok.se/en/latest/Jorge_Luis_Borges

Those two were global doodles.

Goethe, who was given a doodle in Germany, received ~113k: http://stats.grok.se/de/201108/Johann_Wolfgang_von_Goethe

I don't have the time to do any more, but if anyone wants to (or can think of a way to automate it), the list of Google Doodles is here: http://www.google.com/logos/index.html

(I was assuming that Wikipedia was the first result for the person's name in these cases. I also only counted en.wikipedia.org for the global doodles.)

EDIT: extra contextual stats from grok.se


This will spark some controversy but I think if it was a wider appealing doodle then it would get more coverage. In my ignorance I would not click it because I do not know what it represents really.

I think the stats would have also been better a year ago, I now use the address bar in chrome to do all my searches so I do not see the doodle in it's full glory, I see it in the top left hand corner all small. I don't even look at it any more!

A year ago on Firefox I would have clicked it


> In my ignorance I would not click it because I do not know what it represents really.

I actually only click on ones I don't recognize.


Well my statement is not exactly true to be fair, I think it depends how much the doodle inspires me to click it, an orange didn't inspire me today and I didn't recognise the name

I generally read about the doodle on Mashable before I actually click on the doodle itself


You use Chrome's search but didn't use Firefox's? Why?

I've been using browser's search since Opera (Opera -> Firefox)


Well it was hard to break a habit of a lifetime

I used to have google bookmarked, I clicked on it then did a search

I was too lazy to go to the right and do a search from there. When I got Chrome earlier this year because Firefox was crashing all the time, the search box disappeared and I was left with an address bar and no bookmarks toolbar, so I did not click my google bookmark and do the search. I then realised you could search directly from the address bar.

That's the story and reason I started searching from the address bar!

Love Chrome, never looked back since ditching buggy, unstable Firefox!


I haven't ditched Firefox for the totally opposite reason you are giving us.

In Chrome, you type to search in the URL bar so it can go to Google's search results (and hence, giving them money).

Firefox has two bars: one for search (the one you didn't use altough it was way more straightforward than clicking a bookmark) and the URL bar that works as a "I'm feeling lucky button", to go directly to something you already know (you are not _searching_, you are _finding_) as in "wiki tesla" if you want to go to Nikola Tesla's Wikipedia article.

That's a dealbreaker for me and something I can't get on Chrome because of its limited API.


You can choose which search engine you use by going to preferences > search engine: http://sht.tl/3yo http://sht.tl/9erk


if you prefer to have no search provider in Chrome, go into Wrench -> Options -> Basics -> Manage search engines. add a new search engine with any name and any keyword, then set the URL to "%s". it assumes http://, because if you aren't explicitly entering the protocol as well then you definitely mean http://, and if you are including the protocol then that bypasses the search handling anyway.

after mousing over the URL and clicking "Make default" you're done. the default google search doesn't bother me, but some people prefer the explicit separation of URLs and search queries.

if you want to do a search or an I'm Feeling Lucky query, just go back to those same search options and add single character keywords for them.

Keyword: .

URL: http://www.google.com/search?hl=en&source=hp&btnI=I%...

then ". wiki tesla" does what you described without requiring a second input field. if you do more of those types of queries than regular searches, you could make it the default so it doesn't require the leading keyword.

once a few good keyword characters are setup to speed up your common searches like dictionary/wiki article lookups, hn, reddit, etc the lack of a dedicated search field becomes a feature.


Well I actually didn't use the built in search as I couldn't be bothered changing it.

As I was focusing a lot of work at the time on SEO, when doing a search in Firefox around the start of version 3, it used to do a search on the US google which would show me incorrect results. Therefore I used a bookmark which took me to google.co.uk. Firefox then localised the search but I still used the bookmark.

It's all about habit, I'm in a habit now with Chrome and I don't regret it. I might get back on Firefox on version 306, in about a years time.


This remind me about SEO and Spain. Several spanish SEOs thought about this and started building landing pages to target future doodles. And this eventually killed links to search results in doodles in google.es ;) I'm not sure if the limit still aplies after Panda and if other country specific googles have the same regulations.


How much traffic being on a Google doodle gives to a page that is the first result on the result page that appears if a user clicks on the doodle logo. I would not have clicked through to any result.


When Google featured Alexander Calder in their logo a page from my blog happened to be the 10th result on the first page of results for the term Alexander Calder.

I got over 6000 visits that day. Keep in mind that this was 6000 clicks on the 10th result on the search page. So I believe that the hits for the first two or three results must have been much exponentially higher.


Very interesting.

Also interesting is that the blurb on the search result page was truncated before the important information I was looking for could be revealed. I wouldn't have clicked through to the Wikipedia page if the excerpt had ended a few words after "He is credited with discovering ..."


That's all? Would have expected much more. Or is was this a "Sweden-only" doodle?


Ah, face palm: it's currently on the Google Homepage and the US hasn't really woken up yet...


But the UK's been awake all day (it's 3:12pm) - so very surprising that it's only generated 180k visits.

Maybe most people simply don't click the Google Doodle?

Seems strange.


Yes, there will be many more hits for the next days stats.

Random example: http://stats.grok.se/en/201108/Pierre_de_Fermat

That doodle was from Aug 17 2011 - 2.3M hits


Doodle doesn't point directly to the wikipedia article. It only does search for the name. Wikipedia being the first result gets a lot of traffic but there obviously isn't 100% CTR.


It only presents stats for entire days, so it'll show today's hits tomorrow (it updates an hour or so after midnight GMT).


most people go to google because they're looking for something. that something isn't typically the subject of the google doodle.


Hey maybe someone should do a graph over time on this


It's been on my todo-list for a long time to make stats.grok.se into something more useful and present more info there - for example graphs over (longer) times. Also the design is a bit dated. :)

I do have about three years worth of Wikipedia article traffics stored, so I could do long term tracking and some interesting data mining, but I keep getting side tracked into other hobby projects.


I am curious, how do you get access to Wikipedia statistics?


There's a lot of information made available at http://stats.wikimedia.org/


what a scary coincidence!! jusst TODAY i was asking myself how much traffic google doodles give. then i see this post!!


Genuine Question - Why is this topic of any interest at all for Hacker News?

This seems more like something SEO marketers or people with nothing to do would be interested in.

Q1 Does knowing the amount of traffic a Google Doodle gets in any way help someone become a better hacker?

Q2 Does this topic in any way lead to getting better at anything?

Or has Hacker News turned into some tech version of LOLCatz?


hackers are curious about the world around them. this gives a data point on how the world works with regards to the most popular search engine

hacking is not about getting better at something - it's a state of mind


It's related to the topic that is important to website creators, how to get viewers on our websites.


You have a point there. That someone could possibly anticipate what Doodles are going to be used and SEO map their site to that.

It still isn't the kind of thing I personally am hoping to get at Hacker News.

It seems to me that more and more people are interested in shortcuts to success and praying to the Tech Religions than actually doing something themselves or sharing stuff that is really worthwhile.


> Or has Hacker News turned into some tech version of LOLCatz?

You mean digg? Hopefully not, but this headline does tell a different story :( . Yet many of us still comment on this topic.




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: