5 ms·

Fascinating. I thought that HN was more like a 100K community (of monthly active users). I guess my ballpark is correct? I now think it's slightly less.

drclau 5y ago
You know what would be nice? HN stats! Active users (daily, monthly, etc.), how many posts a user checks on avg., how many comments per user, how many submissions per user, top commenters, top etc. etc. etc., breakdown per OS more etc. etc. etc. :) @HN, what do you think?
Lapsa 5y ago
I don't click on every #1 link.
furyofantares 5y ago
I probably click the #1 link less than 10% of the time and idunno there's only like a 1 in 3 chance I check HN when any given story is on the front page, so from their 29k views now I'm thinking HN is 1M active users. Which of us is way off?
wongarsu 5y ago
Some guestimating: I clicked on 3 of the current top 10 posts, so if that's a representative click-through rate then that would put the HN community at ~100k daily active users (which probably means more like 1M monthly). That's assuming the post stayed in the top 10 for 24 hours, or at least long enough for visitors from every time zone to notice it.
onion2k 5y ago
A 6 hour window is less than 1% of a month. If you think HN has 100K MAUs then you should expect way less than 28,000 people to hit HN during that 6 hour window to see the link in the #1 spot. Even if the article was on the front page for a full day you wouldn't expect 28,000 clicks because fewer than that would be hitting HN on that particular day. Traffic isn't distributed linearly across the month but still. Even then, I don't think this is a representative snapshot of HN users anyway. The average time on the site was 37 seconds. That's enough to work out what it's about and decide it's not interesting (to that person). As Codeamigo is a coding school, and HN has a lot of users who can already code, that's not surprising. It's likely that far more people read the comments first and didn't bother clicking after understanding what it is. More importantly than all of that though, is that the link wasn't about Codeamigo. It was about starting it after being rejected by Codacademy. The interesting part of the narrative was about coming back from rejection, not the product itself. The fact 28,000 people clicked at all is amazing.
- traek 5y ago
  This would only be correct if stickiness (the DAU/MAU ratio) were around 3%, i.e. if each MAU only visited once per month.
mceachen 5y ago
I think it'd be hard to back into that number only based on the analytics of one top-page link. I had almost a quarter million hits from a #7 post in January. Here are more details: https://forum.photostructure.com/t/front-page-of-hacker-news/227 https://forum.photostructure.com/t/front-page-of-hacker-news...
ksec 5y ago
From Dang [1] in 2020. >These days around 5.5M page views daily and something like 5M unique readers a month, depending on how you try to count them. And that is excluding API access. Generally speaking I think it depends on the topic and time of day it is on HN front page. Since HN is mostly US specific so it starts at around GMT-4 during working hours. Yes. People surf HN more during works, that is why you see lower submission on weekends. I am also suspicious of decline in page view on HN. Compared to pre 2020. At least the upvote number on front page and comments seems to be trending downwards. And yes, I do wish HN could have more stats available [1] https://news.ycombinator.com/item?id=23808787 https://news.ycombinator.com/item?id=23808787
- dang 5y ago
  HN has been growing at the same roughly-linear rate for many years, including this year and last year. I haven't checked, but I'd be surprised if upvote numbers are going down.
vodkapump 5y ago
I personally very rarely actually go check out what is linked on HN, due to mostly using it on internet connections with very restricted speeds. I also know a fair few people in my social circle that do the same, HN is the goto "I need good content that doesn't zap 100mb of data in 3 minutes" place

drclau 5y ago

You know what would be nice? HN stats! Active users (daily, monthly, etc.), how many posts a user checks on avg., how many comments per user, how many submissions per user, top commenters, top etc. etc. etc., breakdown per OS more etc. etc. etc. :) @HN, what do you think?

Lapsa 5y ago

I don't click on every #1 link.

furyofantares 5y ago

I probably click the #1 link less than 10% of the time and idunno there's only like a 1 in 3 chance I check HN when any given story is on the front page, so from their 29k views now I'm thinking HN is 1M active users. Which of us is way off?

wongarsu 5y ago

Some guestimating: I clicked on 3 of the current top 10 posts, so if that's a representative click-through rate then that would put the HN community at ~100k daily active users (which probably means more like 1M monthly). That's assuming the post stayed in the top 10 for 24 hours, or at least long enough for visitors from every time zone to notice it.

onion2k 5y ago

A 6 hour window is less than 1% of a month. If you think HN has 100K MAUs then you should expect way less than 28,000 people to hit HN during that 6 hour window to see the link in the #1 spot. Even if the article was on the front page for a full day you wouldn't expect 28,000 clicks because fewer than that would be hitting HN on that particular day. Traffic isn't distributed linearly across the month but still. Even then, I don't think this is a representative snapshot of HN users anyway. The average time on the site was 37 seconds. That's enough to work out what it's about and decide it's not interesting (to that person). As Codeamigo is a coding school, and HN has a lot of users who can already code, that's not surprising. It's likely that far more people read the comments first and didn't bother clicking after understanding what it is. More importantly than all of that though, is that the link wasn't about Codeamigo. It was about starting it after being rejected by Codacademy. The interesting part of the narrative was about coming back from rejection, not the product itself. The fact 28,000 people clicked at all is amazing.

traek 5y ago
This would only be correct if stickiness (the DAU/MAU ratio) were around 3%, i.e. if each MAU only visited once per month.

mceachen 5y ago

I think it'd be hard to back into that number only based on the analytics of one top-page link. I had almost a quarter million hits from a #7 post in January. Here are more details: https://forum.photostructure.com/t/front-page-of-hacker-news/227 https://forum.photostructure.com/t/front-page-of-hacker-news...

ksec 5y ago

From Dang [1] in 2020. >These days around 5.5M page views daily and something like 5M unique readers a month, depending on how you try to count them. And that is excluding API access. Generally speaking I think it depends on the topic and time of day it is on HN front page. Since HN is mostly US specific so it starts at around GMT-4 during working hours. Yes. People surf HN more during works, that is why you see lower submission on weekends. I am also suspicious of decline in page view on HN. Compared to pre 2020. At least the upvote number on front page and comments seems to be trending downwards. And yes, I do wish HN could have more stats available [1] https://news.ycombinator.com/item?id=23808787 https://news.ycombinator.com/item?id=23808787

dang 5y ago
HN has been growing at the same roughly-linear rate for many years, including this year and last year. I haven't checked, but I'd be surprised if upvote numbers are going down.

vodkapump 5y ago

I personally very rarely actually go check out what is linked on HN, due to mostly using it on internet connections with very restricted speeds. I also know a fair few people in my social circle that do the same, HN is the goto "I need good content that doesn't zap 100mb of data in 3 minutes" place