hi folks! jae here. welcome to the first financial update of 2023! this is a long one with a lot to discuss so strap in. as always, I recommend taking a look at past financial updates [https://cohost.org/staff/tagged/financials] for context on terminology and concepts. first off: where have we been? why has this financial update taken so long to come out? 2023 has been kind of a mess for us thus far. I was unfortunately kind of a non-entity due to health issues (complications from long covid) during the entirety of Q1, meaning that all work I had planned (including revenue-generating work) was automatically delayed. I’m currently the only one who writes these, which means that if I forget or am unable, it doesn’t happen. (we’re working on fixing this part.) I realize this is kind of a non-answer but it’s the truth. I have a bad working memory so sometimes I forget things, like that I need to write the financial update. fellow ADHD-havers will probably understand. on to the meat! ---------------------------------------- THE “NUMBERS” SECTION EXPENSES to recap, we have three major expense categories: payroll, hosting/operations infrastructure, legal and professional services. our total expenses are roughly $46k per month. as we do every year, we increased our pay inline with an inflation rate of 8.4%, effective February 15. this means pay is now $94,616 for all employees. our total payroll expenses (including taxes, health insurance, etc) are approximately $36k per month. we also completed migrating our CDN and firewall services to fastly, which means we now have a fixed cost with them. previously, we were getting CDN and firewall services for free, due to cloudflare forgetting they had given us a complimentary enterprise contract. this means that regardless of how much we’re paying fastly, it’s more than we were paying before. our fastly contract is set at $4,850 per month (plus a 7% sales tax) for bandwidth/CDN and firewall services. we move approximately 10TB of data per month, 2/3 of which is images. on the firewall side, we blocked 2.9m requests in May, down from 5m in April. we consider the CDN and firewall services to be essential for our site to operate well; we’re able to spend less on other hosting expenses, as well as spend less of our time managing the site load that comes from (targeted and untargeted) “””cyberattacks”””1. the remaining $4k-ish of expenses comes from our other hosting expenses — the servers the site runs on, storage for all of our data, alerting and performance debugging so we know if the site goes down and how to make it faster — and occasional legal/professional services. REVENUE we made around $5300 in may, meaning we only lost $41k. ideally, we would like to lose less money. SUBSCRIPTIONS AND USER COUNTS putting them in bullet point form for legibility. I find paragraphs full of numbers difficult. all numbers approximate and accurate as of June 14, 2023. * total users: 130k * monthly active users: 12k * subscriptions: 1.9k * MRR: $9008 * this is a larger number than our revenue this month because it amortizes yearly subscriptions across the entire year, which isn’t how that works in real life but makes it a useful metric for Overall Business Health. * a lot of our annual subscribers subscribed in June and November, so revenue in those months will look better than average; the other 10 months will look worse. * active user conversion rate: 17.69% readers keeping score at home will note that these numbers are all down or unchanged from six months ago. site growth is effectively flat at this point, as the rush of people off of twitter has slowed down. this means it’s time for hard conversations. SUSTAINABILITY first off, the elephant in the room: at our current burn rate and bank balance, we will not make July 15 payroll. don’t panic, we’re not on the verge of collapse; we’re raising additional funding to help carry us through the year. our most recent funding round has lasted us as long as we expected. as I said above, our monthly expenses are around $46k per month. since we’d like to build a buffer and need to pay back the bonds used to fund the company, let’s set our “sustainable” revenue requirement at $50k per month. COHOST PLUS! our average per-subscriber cohost Plus! revenue is $4.72 per month. sustainability would require around 10.5k subscriptions; 5 times as many as we currently have. our current active user conversion rate of over 17% is unreasonably good and shouldn’t be counted on, but let’s do it anyway. if we maintain that rate, we would need 60k active users. with our current active/inactive user rate, we’d need around 666k2 total users. assuming a more industry standard conversion rate of 8%, we’d need 131k ACTIVE users. that’s more than we have total! given this, the only path to cohost Plus! sustainability is “more users”. however, as I’ll get into later, “more users” increases operation costs, which means you need more revenue to balance it, etc etc it’s a cycle, you get the idea. THAT ALL SAID: if every single active cohost user subscribed to cohost Plus!, we’d be sustainable! a 100% conversion rate is both normal to want and possible to achieve. (it’s not, that will never happen for a variety of reasons. many users can’t afford it, many users don’t want to pay for it, many users just don’t care. this is the reality of any subscription-supported business.) USER SUBSCRIPTIONS brief recap here: we are planning to launch a Patreon-style subscription management service within (hopefully) the next few months. we are planning to launch with a 5% platform fee that scales down once we’re sustainable. this section is about that unlaunched product and is thus largely hypothetical. let’s assume current cohost Plus! MRR of $9k; user subscriptions would need to account for an additional $41k. at a 5% rate, we would need to process around $820k per month of subscriptions on the platform. is this possible? maybe! given that this is an unlaunched product, the only data we have to go off of is how other platforms are doing. this data is effectively useless. Patreon, for example, moves (as of the most recent estimate we could find, which is a few years old) approximately $83m per month — 100x what we need. however, Patreon has a 10 year head start, brand recognition, roughly 100x as many employees, and over 400x more funding. despite all this they are, somehow, not profitable [https://blog.patreon.com/a-note-from-jack]. they are a bad data point for us to use, but there really aren’t a lot of options there. it’s impossible for us to know how many creators would use our subscription management service as well as their existing one. while we’ve talked to some creators, it’s also impossible to know how many of their existing subscribers would move. given that we know very little, it’s hard to bank on this as our Primary Revenue Strategy. we think it’ll help, though. TIPPING this will actually launch before user subscriptions. I’m working on this first — it’s my main priority once asks are live — since it contains a decent amount of the infrastructure work needed for user subscriptions but is, overall, way less complicated. we are planning to launch tipping with a 0% platform fee. users will have the option to give us an optional cut, but we don’t expect most users to change from the default of 0%. we do not consider tipping to be a revenue driver. personally, I think it will make us about as much as the merch store (i.e. almost nothing). THE INHERENT UNPROFITABILITY OF SOCIAL MEDIA Twitter is unprofitable [https://www.ft.com/content/ccb40e03-099f-4bfe-ad34-2e6c3ec88c81], Tumblr is (likely) unprofitable, Reddit is unprofitable [https://www.reddit.com/r/reddit/comments/145bram/addressing_the_community_about_changes_to_our_api/jnkd09c/?context=3]. Meta is profitable [https://investor.fb.com/investor-news/press-release-details/2023/Meta-Reports-First-Quarter-2023-Results/default.aspx] but that’s more because they are a world-consuming surveillance entity that has a few social network products than they are a social media company. all of these companies are ad-supported (the alleged Way To Make Money) and they’re still not making it work. at a certain point you have to ask: is it even possible to make money as a social media company? Ad-free social media is, thus, at somewhat of a disadvantage. by being subscription-based, we’re dependent on our users deciding that they want to pay us, as opposed to advertisers wanting to pay us. advertisers are pretty easy to lie to [https://www.theverge.com/2018/10/17/17989712/facebook-inaccurate-video-metrics-inflation-lawsuit]; users won’t pay for a site they don’t like. those users also have to exist to begin with. assuming infinite user growth: if you need [x] paying users for sustainability, and need [y] total users for [x] to be possible, but you can only moderate [z] users with your current staff, every time [y] is larger than [z] you need to hire, which makes [x] bigger, which makes [y] bigger, etc etc etc. more users costs more money which means you need more users to offset the increased expenses. this continues forever.3 cohost is important to the people using it, but these users aren’t enough for sustainability. with flat growth over the course of the year so far, it’s not viable for us to bank on Just Getting More Users. IS THERE A ROAD TO SUSTAINABILITY? great question! no idea! DISCLAIMER: THE FOLLOWING SECTIONS ARE NOT FIRM PLANS OR NECESSARILY THINGS WE ARE SERIOUSLY CONSIDERING IMPLEMENTING. DO NOT ASSUME WE ARE GOING TO LAUNCH ADS JUST BECAUSE I TALK ABOUT ADS HERE. THANKS. the easiest tool to reach for here (and the one that I think most would reach for) is ads. ADS we are an ad-free social network. we pride ourselves on that, it’s one of our selling points, it’s part of the Identity of the platform. is it even possible to transfer from ad-free to ad-supported without alienating your users? to start with: any ads we sold would bear almost zero resemblance to conventional internet advertising. we do not want to be beholden to Brands. we think there’s benefit to being able to promote your work as an individual or small creator to people who don’t directly follow you, but no one’s coming on cohost to see ads about Toyotathon. the only direction we find palatable is entirely untargeted ads, where everyone on the site (except cohost plus subscribers obvs) gets the same shit regardless of who they are. to be clear: we would rather shut down the site than collect or infer the user data required to make targeted ads possible. we refer to this as the “webcomic ads” model, as the first provider that comes to mind here is Project Wonderful, Ryan North’s now-defunct ad network that was the go-to within certain niches. Tumblr also has a promoted post system (”Blaze”) which is effectively untargeted advertisement; you pay Tumblr for a fixed number of impressions and your post gets served at least that number of times. Blazed posts are moderated before they go live4. Tumblr was able to easily adopt this because they already had ads. adding a different form of ad to an already ad-supported platform is a no-brainer. we don’t already have ads so we wouldn’t get that benefit; adding a Blaze-like system would be the first form of advertising on the platform. there are ways to accomplish this ideal beyond an ad network (a single time-limited site-wide sponsor, like how Blaseball managed sponsorships) but the general theory holds. PIVOT TO SOMETHING ELSE let’s assume social media is a forever-unprofitable cesspool that no one should touch. what do you do? something completely different! unfortunately, this requires a fair amount of capital. you have to either shut down your existing product (would rather not!) or hire people to keep it in maintenance mode while you fuck off to build something new and shiny. we have joked before about “pulling a Panic” and pivoting to becoming a game publisher. however, Panic had a successful boxed5 software business before they published Firewatch, so that’s not actually viable. FAKE IT ‘TIL YOU MAKE IT (MORE FUNDING ROUNDS) eventually, the money will run out. our funder is one person. while we’ve been cagey in the past identifying them, that’s solely because they like their privacy; it’s not a big secret, it’s just not relevant.6 they are not a billionaire, they do not have infinite money, they can not (and should not) fund us forever. our current funding is in the form of interest-bearing bonds; these are debt that need to be paid back. every time we do a new funding round, we take on more debt. the fewer rounds the better. one option would potentially be to open up funding to other people. as we’ve stated before, we do not and will not ever sell equity to non-employees. because of the interest on the bonds, our funder will eventually get back more money than they put in, but that amount is fixed and there aren’t a lot of investors who are interested in funding something they can’t hypothetically make 10x on later. you need equity for that and we aren’t offering it. additionally, the collateral for our existing bonds is ownership of ASSC’s IP rights (at this point that is exclusively cohost’s source code). we’re not able to offer that same collateral to other funders because you can’t easily split that up. what other collateral does the company even have to offer? everything comes back to what I said before: more funding rounds = more liabilities. we would like to avoid this. NON-FINANCIAL SUSTAINABILITY: BURNOUT as I said earlier this week [https://cohost.org/jkap/post/1677028-today-in-man-i-wish], we don’t have the money to hire additional people. this means we don’t have the ability to do things like hire our own replacements. burnout is inevitable. eventually you will reach a point where you just can’t do it anymore. this job is exhausting and, speaking frankly, has gotten more exhausting almost continuously over time. relative to the industry as a whole, we make very little money. money doesn’t buy happiness but it does help offset some of the stress7. while we could just decide to pay ourselves more, that would (obviously) make every single bit of the math involved here worse and is kind of not an option. even if we ignore burnout, the reality of a four person company is that the bus factor [https://en.wikipedia.org/wiki/Bus_factor] is Pretty Bad. there’s some things we can do to offset this, but realistically the only way to fully mitigate it is to hire more people. we must consider burnout to be an incredibly slow moving bus that is moving towards us all at an unknown rate. the bus gets everyone eventually. how do you prepare for the bus? you hire more people. how do you hire more people? you get money. how do you get money? please see the rest of this post for more information there. this is largely an abstract concern, but it’s one that has the ability to become concrete at any moment. could happen in a year, could happen in five years, could happen tomorrow. any one of us could wake up one day, say “fuck this actually”, and hand in a resignation8. it is not currently possible for us to protect against that. IN CONCLUSION sorry to end on such a bummer. I am incredibly depressed, stressed out, and have been for a while now. things are not getting better and that means I am forced by my brain to be a “realist” (pessimist) about these things. as I said before, I’m teaching someone else how to prep these so someone less depressed can work on it next time9. if you want to help in the immediate term, subscribe to cohost plus [https://cohost.org/rc/user/settings#cohost-plus]. we recently added support for EUR and GBP pricing, which means that regional payment methods should now work. if you’ve got other currencies you want us to support, please e-mail support@cohost.org [support@cohost.org] and we’ll look into it. if you’re an existing subscriber and want to give us more money (thanks!) you can click “manage subscription” at the link above and update your “quantity” there. at the moment, buying extra subscriptions doesn’t do anything but give us some walkin’ around money. we’re going to add gift subscriptions down the road, but we’ve got a lot of more immediately useful stuff to ship before then. if you want to help out for free and you’ve got ideas for making the site better, please submit them to the feature request forum [https://help.antisoftware.club/support/discussions/forums/62000112864]! we’ve got a pretty sizable backlog at the moment, but we do sincerely want to hear from y’all on what you want from the site. that’s all for now. we’re hoping to move back to a monthly cadence with these moving forward so you should see the next one in a few weeks. thanks for your support, and thanks, as always, for using cohost. :eggbug: ~jae ---------------------------------------- 1. first off: “cyberattacks” is a really dumb word but I kind of love it. I will probably talk more about the nature of these attacks at some point, but this post isn’t the place for it. 2. lol it’s the weed number 3. because of this, it’s absolutely bonkers to me that reddit has failed to be profitable. their moderation staff is largely free labor! how do you fuck that up! absolutely wild. steve huffman is a Business Genius and if he puts out a book i’ll pirate it day one and screenshot all the dumbest bits. 4. at least in theory; there’s been several instances of posts (mostly hate speech) that a reasonable moderator wouldn’t let through still getting in. 5. once literal, now metaphorical 6. they actually are fairly active on cohost and might be reading this right now! if you’re reading this, hi [redacted]! thanks for believing in us! 7. it also makes things like “moving out of your hometown in Florida in an expedient fashion because you are a trans person and it’s pretty generally known that being a trans person in Florida is not a great move right now but was a much more reasonable decision when you actually moved back here” much easier and less financially devastating10. 8. I have come closer to this than I would like to admit 9. Hi! Aidan here. I'll probably write the next one of these with jae’s expert guidance. 10. on a personal note, I burned all my savings paying bay area rent for the six months before we had funding, and then I burned all the stock I still had left from past jobs funding our move from the bay area to Tallahassee. I can’t fuckin afford to move, so I instead choose to spend my time stressed out about this job. better than the alternative.