FA+
Shop
Art Posting Analysis of 2020 and Site update
4 years ago
I think that statistics are neat, so I thought that I'd make a write up of what I observed from all of the data I've hoarded through the year, it even might help some people to even defeat the algorithm and figure out when to post and when not to post. Some people might think this is revealing stuff that's behind the curtain that will lead into losing money and customers, but hey, it's not like I'm earning any money from this.
But first, site updates!
Site updates:
-Site now has a new coat of paint, going from black, white and grey color scheme to dark blue across the board.
-NSFW Notification! You shouldn't be really going to sites like these if you're underage.
-Since midnight 31th of December 2020, the server is now using content distribution network to deal with the massive load of traffic saturating the 100mbps link the server uses. Over 8 gigs of traffic served from the CDN in the first 12 hours! Unfortunately this breaks the automation that posts new content on Discord every time I upload something, but it stops getting the server to its knees every time I upload something and the traffic floods in.
The Anal-ytics:
This is not meant to look like boasting about "haha look at how much traffic I have", I just want to share some of the data that I hoard. No I'm not going to show off the Grafana dashboard full of graphs for my power usage (unless you like that kind of content?). I'm consolidating data from only 2 sources as FA doesn't have any native way of viewing or exporting the amount of views, favorites or comments and you can't easily punch in the numbers into an InfluxDB database and have pretty Grafana graphs to look at. The data from the site are skewed to the low side in terms of user visits, as the data is scraped off webserver access logs that have been anonymized to comply with GDPR.
I think you'll start to see a pattern with some months and why I was forced to either upgrade to a better hosting or start utilizing a CDN. You will also see that the growth on twitter is just basically you being at the mercy of the algorithm, while on FA it is the timing when you upload. I found out that uploading around midnight GMT+0 on Fridays onto FA is the idea sweet-spot, as Europeans are still awake because they don't have work tomorrow and people from the west coast are just returning home from work, making it ideal for most people to view the published content.
I don't really bother with data from e621, 'tis a silly place. Every time I upload something there its like "why is my post sitting in red for over a day now???" with no explanation and its only use is to just dump images on it in case my server catches on fire (AGAIN) or I die and nobody pays for the server bills or domain renewals. Bit morbid but once you spend a month in hospital dying you start to think about this stuff. One interesting thing I discovered is, that if you upload stuff onto e621, it will appear on rule34 in a few days, as rule34 scrapes the content off e621. Seems a bit unfriendly but it solves the issue of having to upload stuff onto ANOTHER place.
Now... The DATA!
January:
569 number of visits to the site for which 302 were unique and 3.30GB bandwidth was used. 6,902 tweet impressions and 17 new followers on Twitter. Most popular piece was "Commission: Lyndrael"
February:
448 number of visits to the site for which 253 were unique and 2.70GB bandwidth was used. 9,170 tweet impressions and 9 new followers on Twitter. Most popular piece was "Commission: DegenerateBael"
March:
827 number of visits to the site for which 421 were unique and 8.00GB bandwidth was used. 13.7K tweet impressions and 46 new followers on Twitter. Most popular piece was "Pirate Trouble"
April:
744 number of visits to the site for which 376 were unique and 7.31GB bandwidth was used. 19.1K tweet impressions and 47 new followers on Twitter. Most popular piece was "A new foe has appeared"
May:
511 number of visits to the site for which 304 were unique and 6.99GB (nice) bandwidth was used. 13.5K tweet impressions and 22 new followers on Twitter. Most popular piece was "Sticky Defeat" with "Choose your words wisely" almost beating it to the punch
June:
578 number of visits to the site for which 315 were unique and 7.91GB bandwidth was used. 12K tweet impressions and 11 new followers on Twitter. Most popular piece was "Midnight spear polishing services"
July:
656 number of visits to the site for which 364 were unique and 12.34GB bandwidth was used. 21.5K tweet impressions and 38 new followers on Twitter. Most popular piece was "Onlyworgs"
August:
785 number of visits to the site for which 367 were unique and 14.25GB bandwidth was used. 23.8K tweet impressions and 37 new followers on Twitter. Most popular piece was "Warrior's backside"
September:
768 number of visits to the site for which 362 were unique and 18.78GB bandwidth was used. 16.7K tweet impressions and 22 new followers on Twitter. Most popular piece was "Rough Filling"
October:
774 number of visits to the site for which 378 were unique and 18.72GB bandwidth was used. 22.4K tweet impressions and 28 new followers on Twitter. Most popular piece was "Pumpkin Spice Worgen - Female version"
November:
855 number of visits to the site for which 474 were unique and 20.90GB bandwidth was used. 17.6K tweet impressions and 19 new followers on Twitter. Most popular piece was "Easy Theft - Alt Version"
December:
929 number of visits to the site for which 511 were unique and 33.14GB bandwidth (oh my GOD) was used. 19.5K tweet impressions and 18 new followers on Twitter. Most popular piece was "Petrifying Surprise - Female version"
The large dip in May is mainly because I started a new job back then and had to focus on it more, followed by sickness that got me in the ER again and then I moved to a new place. Then from July to October, I thought that I have fully growth into the kind of niche stuff I make and carried on as usual, although saddened by the lack of new watches or unique visitors, until november and then december. I'm going to be honest, I have no idea what happened there.
Well hopefully you enjoyed this wall of text about the numbers that my content generates.
But first, site updates!
Site updates:
-Site now has a new coat of paint, going from black, white and grey color scheme to dark blue across the board.
-NSFW Notification! You shouldn't be really going to sites like these if you're underage.
-Since midnight 31th of December 2020, the server is now using content distribution network to deal with the massive load of traffic saturating the 100mbps link the server uses. Over 8 gigs of traffic served from the CDN in the first 12 hours! Unfortunately this breaks the automation that posts new content on Discord every time I upload something, but it stops getting the server to its knees every time I upload something and the traffic floods in.
The Anal-ytics:
This is not meant to look like boasting about "haha look at how much traffic I have", I just want to share some of the data that I hoard. No I'm not going to show off the Grafana dashboard full of graphs for my power usage (unless you like that kind of content?). I'm consolidating data from only 2 sources as FA doesn't have any native way of viewing or exporting the amount of views, favorites or comments and you can't easily punch in the numbers into an InfluxDB database and have pretty Grafana graphs to look at. The data from the site are skewed to the low side in terms of user visits, as the data is scraped off webserver access logs that have been anonymized to comply with GDPR.
I think you'll start to see a pattern with some months and why I was forced to either upgrade to a better hosting or start utilizing a CDN. You will also see that the growth on twitter is just basically you being at the mercy of the algorithm, while on FA it is the timing when you upload. I found out that uploading around midnight GMT+0 on Fridays onto FA is the idea sweet-spot, as Europeans are still awake because they don't have work tomorrow and people from the west coast are just returning home from work, making it ideal for most people to view the published content.
I don't really bother with data from e621, 'tis a silly place. Every time I upload something there its like "why is my post sitting in red for over a day now???" with no explanation and its only use is to just dump images on it in case my server catches on fire (AGAIN) or I die and nobody pays for the server bills or domain renewals. Bit morbid but once you spend a month in hospital dying you start to think about this stuff. One interesting thing I discovered is, that if you upload stuff onto e621, it will appear on rule34 in a few days, as rule34 scrapes the content off e621. Seems a bit unfriendly but it solves the issue of having to upload stuff onto ANOTHER place.
Now... The DATA!
January:
569 number of visits to the site for which 302 were unique and 3.30GB bandwidth was used. 6,902 tweet impressions and 17 new followers on Twitter. Most popular piece was "Commission: Lyndrael"
February:
448 number of visits to the site for which 253 were unique and 2.70GB bandwidth was used. 9,170 tweet impressions and 9 new followers on Twitter. Most popular piece was "Commission: DegenerateBael"
March:
827 number of visits to the site for which 421 were unique and 8.00GB bandwidth was used. 13.7K tweet impressions and 46 new followers on Twitter. Most popular piece was "Pirate Trouble"
April:
744 number of visits to the site for which 376 were unique and 7.31GB bandwidth was used. 19.1K tweet impressions and 47 new followers on Twitter. Most popular piece was "A new foe has appeared"
May:
511 number of visits to the site for which 304 were unique and 6.99GB (nice) bandwidth was used. 13.5K tweet impressions and 22 new followers on Twitter. Most popular piece was "Sticky Defeat" with "Choose your words wisely" almost beating it to the punch
June:
578 number of visits to the site for which 315 were unique and 7.91GB bandwidth was used. 12K tweet impressions and 11 new followers on Twitter. Most popular piece was "Midnight spear polishing services"
July:
656 number of visits to the site for which 364 were unique and 12.34GB bandwidth was used. 21.5K tweet impressions and 38 new followers on Twitter. Most popular piece was "Onlyworgs"
August:
785 number of visits to the site for which 367 were unique and 14.25GB bandwidth was used. 23.8K tweet impressions and 37 new followers on Twitter. Most popular piece was "Warrior's backside"
September:
768 number of visits to the site for which 362 were unique and 18.78GB bandwidth was used. 16.7K tweet impressions and 22 new followers on Twitter. Most popular piece was "Rough Filling"
October:
774 number of visits to the site for which 378 were unique and 18.72GB bandwidth was used. 22.4K tweet impressions and 28 new followers on Twitter. Most popular piece was "Pumpkin Spice Worgen - Female version"
November:
855 number of visits to the site for which 474 were unique and 20.90GB bandwidth was used. 17.6K tweet impressions and 19 new followers on Twitter. Most popular piece was "Easy Theft - Alt Version"
December:
929 number of visits to the site for which 511 were unique and 33.14GB bandwidth (oh my GOD) was used. 19.5K tweet impressions and 18 new followers on Twitter. Most popular piece was "Petrifying Surprise - Female version"
The large dip in May is mainly because I started a new job back then and had to focus on it more, followed by sickness that got me in the ER again and then I moved to a new place. Then from July to October, I thought that I have fully growth into the kind of niche stuff I make and carried on as usual, although saddened by the lack of new watches or unique visitors, until november and then december. I'm going to be honest, I have no idea what happened there.
Well hopefully you enjoyed this wall of text about the numbers that my content generates.
