Yesterday (February 18) marked another important milestone for my blog - for the very first time my blog traffic crossed 10,000 page views in one day. Now, don't get me wrong - I would love to report 10,000 daily page views instead of 10,000 in one day, which happened thanks to the huge traffic spike made by a dozen of referring links.
By now my blog gets about 2,000 page views per day, which of course is already a very great result, but having that traffic bump made me smile this morning. OK, I was actually closely monitoring stats already since yesterday and hoped I will manage to crack 10,000 unique users per day. I was short for about 1,300 users at the end.
10,000 page views in a day
When speaking about traffic spikes - in most cases (at least for me) - it doesn't mean my earnings are up as well. As you can see from te above screen - pages / session ratio is superb low, barely reaching 1.1. Normally my pages / session ratio is between 1.3-1.5. Although I believe I should aim something closer to 2 pages per session. The same applies for average session duration - 28 seconds instead of about one minute and 30 seconds on regular days. Bounce rate of 87.16% is very poor performing results. On regular days my bounce rate is just above 60%.
Now answering on the question, mentioned in the title of this article
How to get 10,000 page views in a day
There is no magic or secrets, for me it took more than two years of everyday blogging, but the real reason is not even everyday blogging but a research on a highly social sharable and debatable topics, yesterday my article about Average salary in the EU got on the front page at Reddit, basically I would call it more a random luck, instead of a focused efforts of getting there. Reddit has been involved in previous two of three traffic spikes to my blog.
Speaking of me, I'm not a Reddit user and I don't know for sure how things are working there, but I believe users there are savvy tech people, being ad blind and not engaging with my content anyhow on my blog (I got just one comment from about 6,000 users yesterday) instead they are continuing conversations on Reddit. My blog was just a source (a tool) in the hands of the mighty Reddit users.
On the other hand - if people are not engaging, revenue is low - this probably should be called a "wrong traffic" - it depends, basically yes, there is no direct benefits from such traffics spikes, but I believe they help in a long term. Yesterday I got a few additional referrals from some other forum sites as well - which is a direct benefit from Reddit and confirms my theory about tech people there - instead of commenting or clicking on ads on my site, folks posted more links to my blog in other resources, which might help in future to bolster my SEO (I'm not doing any offline SEO, most of the backlinks are earned in natural way)
Can your hosting handle 10,000 page views per day?
The short answer is - it should. if your website is a simple, not involving complex database queries and most of the users visiting your site are anonymous with cache enabled you should be able to handle 10,000 page views in a day with ease. You can get started with a cheap hosting from Bluehost or you can choose more expensive but more reliable VPS from sources like Linode.
Speaking of me - I ran a $40 box on Linode + Nginx and of course Drupal, yesterday's traffic spike was barely noticeable on my Linode performance reports. Just recently I upgraded my blog to secure HTTPS pages and implemented KeyCDN. The only thing I noticed here was increase in bandwidth:
KeyCDN traffic report
Traffic increased to about 4.7 GB from around 1GB bandwidth on regular days.This led me to further investigate is there a way I could save some bandwidth and deliver even faster page views for less money.
Lessons Learned
This was already fourth traffic spike to my blog, by now the largest one, there is a simple lesson I have learned by now - traffic spikes come and go, I'm not too much worried about them, my biggest concern is just can my server handle them - for now it seems - I could handle 50,000 page views per day (well it depends of course, but I'm not worried of my server infrastructure)
Thus there is one additional lesson I learned - I should further fine tune my page load times. I taught that there is still room to improve my monetization strategy from Google AdSense. Currently I'm serving 3 AdSense ad units, they are generating me some decent (not much) amount of money, but CTR are well bellow 1% (there are two ad units bellow the fold and one sidebar unit). I decided to experiment in next 10 days (till the end of February, by removing header image for each article, which looks great, but takes 80% of above the fold placement) and move up one of the AdSense ad units to see if there will be an increase in CTR.
That said, it's time to wait for another traffic spike, probably I will cover some article on 10,000 unique daily users (hopefully anytime soon)