I've been running it for over 10 years now, it's a database of Steam games, their updates, price history, charts, and a lot more.
In the early days we took monetary donations but stopped a few years in. It costs less than 100$ a month to run. Cloudflare reports 552.2M requests in the past 30 days, and 6.09M unique visitors.
You run SteamDB? Thanks for making this! I've used it for all those 10 years whenever I'm getting back into gaming and want to see what the popular games are! The player count charts are a really awesome feature. Thanks for keeping it running all these years.
You're passing on $50k-100k per month of ad revenue from just basic banner ads... at least based on how RPMs were a few years ago. I respect that, definitely not the decision I'd make.
Even assuming a _very_ low $0.01/session, he'd be making $60k/mo.
More realistically, if he decided to do this right, and sell direct ads using something like Kevel, with "promoted" game slots or something on the homepage and search, he could do $10-$20CPM direct since it's a large and well known site with high purchase intent.
Let's say a conservative 4 impressions per page and 3 pages per session.
(12 imps per session * 6m sessions)/1000 * $10CPM is $720k / mo
This is a _very_ achievable number, quickly, for a site that is this close to purchase intent.
I don't understand. Even putting a single adsense ad at a $1 CPM would net him close to $20k/mo. This might be one of the least monetized sites I've seen at this scale.
You know nothing about this guy or his values. It can be just as stupid to harm a great accomplishment in order to drive a fancy car, so maybe let people have their own values
I see your maths here but Cloudflare hits include any assets, images, and so on so one page views fans out potentially dozens of hits. Cloudflare also includes all bots and non-human hits.
Assuming $0.50 net CPM on one AdSense unit per page view, that would be $70k monthly/$840k yearly revenue. And since operating cost is $100, that's all profit. I'm no AdSense expert but I believe this would be super conservative, maybe an expert could chime in with higher realistic numbers. Even if you don't like advertising there are other ways you could earn money with a site like SteamDB.
Cloudflare stats are inflated. I have a site getting 30k visits a day according to Google Analytics but Cloudflare says it's close to 100k. So I think Cloudflare doesn't filter out bots in their analytics.
I'm not doubting the stats, just saying that Cloudflare is not an accurate gauge of website views and visits. Cloudflare shows 3x the pageviews on my site than Google Analytics.
I know nothing of this, but does .info reduce the numbers, compared to something like .com, all other things aside?
edit: err, it's a genuine question, from someone that doesn't work in/with advertising, and had a .info domain within the first month they were offered, in 2001. Here's a rephrasing: Do ad networks, or customers of those networks, treat TLDs differently?
Yeah I wish I wasn't anti-ad but it's so bad... Ads between every paragraph or ads that pretend to be a virus so some fake Microsoft/Amazon scammer can get you to call them.
It's funny I wonder about the influence of ads too like you by using ad block experience a better internet (no influence on behavior).
There's a hijacking thing too of intent, at which point I modify websites to hide these above the fold attention grabbers.
My kindle forces you to read an ad everytime you turn it on. I know you could root it, lost money on device, etc...
Agree with you. Ads make sites look bad and I say this as an owner of several websites that run ads lol.
SteamDB looks pretty amazing and sleek without ads. Putting ads would only degrade the user experience and force people to install adblockers or use adblocking browsers like Brave. Kudos to xPaw for keeping his site ad-free for all of us to enjoy.
Simply by the fact he runs SteamDB without monetization, I can safely assume he's much richer than me. I would'nt try to teaching someone who's richer than me how to make money.
Wow, you don't make any money on SteamDB? It seems like the sort of thing that should be able to make a lot of money. Even very minimal ads would probably bring in hundreds of thousands if not millions of dollars per year, and if you hate the idea of ads it seems like you could probably sell a premium subscription for some of the info you have that would be useful for devs, like imdb pro or crunchbase pro or something.
It's clear that he is a (very good!) full-stack developer. I am curious about where he works, or if he is retired or something, living off investments while working on SteamDB full time.
I think very few people would knowingly turn down income of a million dollars per year. In fact I'd say it's a privilege to be in a position to be able to turn that down. I even see some people here who are asking him to make the income even though he doesn't want it so he can give it to a charity.
I wouldnt be surprised if you could easily work out some sort of deal with Steam, if a game page from your site leads to a buy on their end they give you a small percentage.
I hate ads too. The thing I tried that felt ethical and useful was an affiliate code for related (to the content) products that would split the discount 50/50 between me and readers. I stated this clearly in the ad.
Avid user for years, so cool to see you on the social web! Thanks for your work! I'm annoyed for you at all these people trying to tell you what to do with your project, after you've been doing it successfully for 10 years and they have not. Y'all really think this person hasn't thought about monetizing? Take it easy startup-stans!
I can definitely see the frustration. Highly money-motivated people working their arse off to try and make even $1k MRR, and all this guy needs to do is push a button and he's got 100k. Must blow their minds.
I haven't read every comment, but I see a lot of surprise, a lot of respect and no annoyed/pissed off "startup-stans". Where are you getting this from?
You are an absolute chad.
I salute you.
Not to undermine your motive and intent, but people will gladly click on links to support your site.
Regardless, what a legend you are.
I want to say thank you for building this, but also there's no shame in putting a simple banner ad somewhere on the page. Most people who care have ad blockers, anyways.
I'm not an user myself but since many comments chastise you of not making money with ads, I would like to thank you for your work and for keeping it free from ads
Yes, thanks for reminding me and everyone else that you can just do something nice for others and pay for it.
Maybe it is worth paying $0.00000018109 to satisfy someones request? If they are running up the bill to $0.00001642036 per person (90.6751339113 req/user/month) there must be something there worth something for them?
I get that some people are to cheap to purchase fun for others but at these rates it is more like people are against fun in general. One can be cheap but one cant be that cheap?
I've had this discussion before... uhh... a few times.
Them: Why are you doing this????
me: Aren't you having fun?
Them: YEAH, but why????
me: You see little jimmy over there? You think he is enjoying himself too?
Bruh! I love this site. You absolutely could put ads here, in a way that isn't annoying.
I guarantee you that there are game shops out there that would pay you like $10k minimum to "Feature" their game for a month or something.
You feature 5 games a month for 10k, you suddenly are at 50k/mo.
Or, you could go the route of using traditional ads like google ads etc, but those can be kind of spammy and the ad targeting kind of sucks.
I'd really recommend exploring that, and don't feel bad about using ads. You deserve to get rewarded for the excellent site you've put together, a lot of people clearly get value out of it.
It really is impressive how versatile that stack is, both on the low end and high. Wikipedia is also basically that stack too (just with apache instead, and varnish/ATS playing an important role)
I'd love to learn how you managed to scale on a single server - other than the above, any tips or tricks? I'm also a PHP guy on Ngnix with MySQL but want to know what tips you might have for a dynamic (non-static) site to handle such scale on a single server
Equip your server with a decent CPU, enough RAM to allow MySQL to keep most of your database in it, and fast SSDs. Profile your code and optimize/cache anything that stands out as slow, aggressively cache data that is read often using memcached/redis/etc, use an http cache proxy like cloudflare in front to cache static assets (and maybe pages that aren't dynamic for every user)
The amount of value I get out of your service is nuts! No better place to figure out whether a multiplayer game is worth getting... Daily users is hard metric to game or lie about.
I actually wouldn't mind some ads on this. Not the generic Google banner ads, but some cutesy illustrated banner in the style of Darkest Dungeon and such. Open an ad slot for bidding and the most interesting banner wins.
Most people who do things "just because", don't monetize because either it isn't possible or it isn't worth the effort. For a site with as much traffic as GP's, it's probably worth it. Regardless, anyone turning down that much revenue out of principle deserves a lot of credit.
Sometimes making money off of something turns out to be incompatible with enjoyment, and you find out too late.
Depends on the person, how slippery the slippery slope is, and the motivation for a particular project.
I could ask myself "how much money would I be willing to pay to bring benefit to a large number of people?" It's definitely nonzero. But that's not a great question, because we all hate losing stuff we have (and paying is losing). It's a lot easier psychologically to forgo getting the money in the first place.
(Note that I wouldn't be able to forgo that much money. Though I might be able to procrastinate throwing the monetization switch for a while. How long is relative to my level of privilege in my own finances.)
Agreed. Depends on the personal point of view for every individual, and that's the reason I commented as my "2c".
Personally, if I were the guy running the steamdb website I would have brought in the monetization aspect as quickly as possible, once I knew there was a lot of traffic coming in.
I think Valve will shut it down once it learns about the monetization. AFAIK that is what happened to https://steamspy.com/ when they introduced paid subscription tier.
Valve didn't shut it down though. It just gave people privacy controls with sane defaults, cutting off data to scrapers. Not sure how much SteamDB depends on data like that but it seems to be doing fine.
Thanks for making this available. Has been a huge help over the years. Would make a good platform for other side projects (e.g. SteamDB YouTube) if you were to look for new challenges one day.
Had never seen your site before (stopped gaming before Steam became popular), but in visiting it now, very surprised to see that CS is still __by far__ the most popular online game.
Oh, cool, I used to run something very similar called steamwatch.com, back before they allowed you to have watch lists. It would monitor prices, let you know about sales, and I would occasionally run game give away raffles for free with whatever ad revenue I collected. It only ever made maybe $100 over its entire run. This would've been over 10 years ago though so I doubt mine and yours ever coexisted.
Used it some years ago and thought that it was a really good front-end to the Steam store since the search interface is a bit lacking. Am surprised to see that this has millions of visitors but has such low running cost. Must be really highly optimized SQL queries and code there!
I would suggest you to place a donation regressive counter with the price that takes to make the website run, let's say 100 USD, everytime someone make a donation the value updates until it reaches the donation goal of the month.
How would you suggest starting data-mining for a project like this?
Every time I think of a project like this, I convince myself that steam/amazon/twitter/Facebook/etc will rate limit any bot that tries to scrape data from their service.
Thak you (and Marlamin) for SteamDB and especially for providing Linux filters as well as the steamdb IRC bot that you ran. Sad that you moved your channel off of IRC when Freenode died.
Hey there! Thanks for creating this site. Long time user here. It is an awesome product!
I was curious about the system design diagram of SteamDb. What type of DB do you use to record historical prices?
thanks a lot for making this happen, in this exceptionally altruistic way, learning that such as staple as steamdb is a passion/hobby project really makes one believe in the software engineering mission
Just wanted to say big big fan. Have extensively used it during sales to see the lowest record prices. I absolutely love everything about the site, the design, the layout. Thank you for making it.
i am a bit late to comment but thanks for creating and maintaining the service for all this time! it really came in clutch when deciding on buying new games.
i am curious to hear from you about your experiences with people trying to scrape data off the site. i noticed that we cannot go back beyond a couple years for price history without logging in for one. is that a decision made because of people scraping data?
Thank you! SteamDB has been super helpful for me. I like to be able to pull posters and banner art for games, and it's also super helpful for untangling bundles of Rocksmith DLC.
Thank you very much! Your killer feature for me is being able to sort my library by fewest hours played and highest review rating so I get the best value out of my library.
The number of users maybe, with millions of users making requests to get the data, a $100 single server won't be able to handle the load. that's why i asked, anyway great achievement
I've been running it for over 10 years now, it's a database of Steam games, their updates, price history, charts, and a lot more.
In the early days we took monetary donations but stopped a few years in. It costs less than 100$ a month to run. Cloudflare reports 552.2M requests in the past 30 days, and 6.09M unique visitors.