>No surprises here. I used to use Plausible, but that costs money and you can't beat free. Monthly cost: $0.
Thanks hard pass. I checked and Plausible starts from 9 Euros a month. This is hardly what I'd consider advertising material. But likely I'm just not their target audience.
There's also no consent banner, so this doesn't seem GDPR compliant
Edit: I must have phrased this in a confusing way. They are NOT using Plausible but GA. I'm well aware cookie-less analytics does not require consent
I was referring to "There's also no consent banner, so this doesn't seem GDPR compliant". They explicitly state that they are GDPR compliant without the need for a consent banner.
> There's also no consent banner, so this doesn't seem GDPR compliant
I'm not sure what you mean by "consent banner". If you're referring to an annoying "allow us to track everything you do" pop-up, that's probably because Plausible doesn't attempt to track everything you do.
They have plenty of privacy information on their site [1].
Even if you are using Google Analytics, you can skip the consent banner by denying on behalf of the user. GA will then update your counters without doing anything that would require consent.
ad_storage='denied'
> Requests are sent through a different domain to avoid previously set third-party cookies from being sent in request headers.
> Google Analytics will not read or write Google Ads cookies, and Google signals features will not accumulate data for this traffic.
> IP addresses are used to derive IP country, but are never logged by our Google Ads and Floodlight systems and are immediately deleted upon collection. Note: Google Analytics collects IP addresses as part of normal internet communications.
> There's also no consent banner, so this doesn't seem GDPR compliant
Plausible uses this one weird trick where they don't use cookies and don't store PII, and thus don't need consent (neither for cookies nor for GDPR). You lose some data because you can't track users across visits, but you gain accuracy (because you can track every visit, not just people who agree) and don't have to annoy your visitors.
Basically. Except that htaccess logs won't correctly track page visits on your fancy react page, so you still need a bit of JavaScript.
And once you have that you can also track visit time a bit better, or track goal conversions like how often people click some specific button. But other than that it's basically a more modern webalizer.
And Plausible won't correctly track visits for users who block their JavaScript :)
Also, last year I found a bug in Plausible's script which silently stops it working on some pages, so your data won't be accurate in those scenarios either.
Those “free” credits on these cloud providers are akin to how drug dealers sell unadulterated product in small amounts to new users to get them hooked. Then repeat business is the cut product with high markup.
The big brained maneuver is probably to build a cloud agnostic stack and migrate between these clouds, negotiating free credits as you go. Done right you could probably pay no cloud bills for several years.
The problem is that one does not simply build “cloud agnostic” services and you would probably spend a not insignificant amount of time migrating between them every year or two. But if you were really good at cloud and could do this efficiently and quickly it would probably be worth it
> one does not simply build “cloud agnostic” services
Why? From personal experience, everything I do is either:
- a static site, which you can run on anything from Netlify to some static CDN to GitLab Pages to a potato, or
- a bunch of Docker containers, which I usually run on Docker Swarm on any random VPS, and manage with my own little tool, https://lunni.dev/; AWS of course has some “cloud native” options for that, too (ECS + Fargate?), but I find predictable monthly bill a bit better on my nerves.
This and some S3-compatible storage do wonders for me.
Spinnaker project was supposed to make multi-cloud somewhat feasible. Assuming your application stack was fully "cloud native" with none to minimal vendor lock in to specific cloud provider products.
I would NOT call this frugal at all, actually. You are overspending in a lot of places and under spending in others. $281.32 per month for such a simple service is...astonishing. Change Vercel and Intercom with Cloudflare and Chatwoot (self hosted), and you have saved $120/mo. Get rid of Ahrefs for Ubersuggest (or no SEO tool at all) to save about $80 more. That leaves you with $81.32/mo. If you try even harder, you can easily bring that under $50.
But I get it. If X tool saves time, why not pay for it? I wouldn't call that frugal though because you are still spending where you can save.
Ubersuggest has a comparison page[0]. From my understanding it boils down to Ubersuggest being cheaper and most likely enough for what most users will need, but Ahrefs has a larger tool suite (and much higher cost).
Use one of the OSS password managers to save on 1P. If you have a cloud sync service you already use, KeePassXC + cloud sync will solve having the passwords available everywhere you are. You can even use Syncthing for free (though I'd at least donate once off) since you can use it peer-to-peer between your devices.
SEO really needs disruption here on price. Ubersuggest seems decent.
On Intercom: OP can probably go into more detail on why Intercom's widget is so much better. If it's just animations and UX, then yeah go find some alternative or self host.
What's useful in the self-hosting space is finding either a tool to manage server provisioning & management (e.g. Ploi), or a hosting provider that makes it easy to install (e.g. DO marketplace). Then you're only spending a small fee on the management and only spending on hosting cost (which for a long time can be negligible).
Spending some time to DIY with reliable tools (automated where possible) is IMO still preferable to spending this much money if starting from scratch.
Looks to me like the 1P biz rate is $7.99 per chair per month, so once you hire employee number 100, you're spending $9600/yr. That shouldn't put you out of business, but still could be spent elsewhere?
Does nobody here see the irony in bending over backwards to avoid spending money on software products when the entire purpose of your business is to sell a software product?
I could see if we were talking about a side project with no commercial intent. But this post is about an indie Saas founder...if a few hundred dollars a month is a barrier to success, then that's a pretty good sign your indie startup isn't starting up.
I find these threads endlessly amusing considering that, for hundreds of years, starting a business involved massive investments in real estate, physical equipment, etc. We're living the dream, yet somehow quibbling over spending $10/month on a piece of software. If it saves you just 2 hours, per year, then it's worth it.
I take frugal to mean being smart or conservative with spending. However, optimizing to the extent some people are talking about here would redefine the term to mean "irrational" for me.
In the context of a business, nearly any piece of software is a trivial cost compared to hiring a human (or yourself) to do a task.
That's just being nice. We developers are a right bunch of cheap bastards. At least I am. I'm working on it, but it's just so ingrained and pervasive, it's weird. I've been trying to figure out where it comes from for a while now, and how to get people to work past it, with little success.
If there was a magic debugger that made debugging 100x easier and faster, but it wasn't free and open source, I'd still have a hard time using it even though it would save me a bunch of time, and easily pay for itself in time and frustration saved.
I generally agree, but I'd say it depends. I'm not too surprised that software developers are less likely to spend money on software - they do have another option non-programmers don't have. I'd also not be shocked to hear that trained accountants starting a company are less likely to hire accounting firms than non-accountants.
And most software I deal with isn't exactly free of opportunity cost - I dabbled a bit with accounting tools to the point of frustration and time sunk that I decided to roll my own (built around ledger, in case anyone cares) in about a day. Writing my own tool is sometimes faster for me than trying to figure out how to use one that's not designed around my specific needs.
Sure, 10$ doesn't seem like much, but all these little invoices can creep up to a non-trivial sum.
Yes, but shouldn't I be making more even money by using these services to optimize those other things? If, say, Calendly helps me get booked for appointments, appointments that wouldn't have happened without that app, then isn't that spending money to make money?
If you think so, it's probably true. That's why I can't really disagree with parent :D
I find the cost and value of tooling options quite difficult to fully grasp, it often comes down to gut feeling IMO.
What I would want to caution folks for is that the cost of a tool is not only what it actually costs. It's also the amount of time that has to be invested into usage, and potentially migrating to something else a while down the line in case of discontinuation, disappearing capabilities, new requirements, ToS or pricing changes, ...
Is it bending backwards though? The problem is that saas solutions scale in cost as well, so when you grow you pay more. For a small startup, there is always a cut off where the price of the saas service(s) explode well over the utility but you are hooked and you can afford it now. But should you? Personally I enjoy profits and more profits are better. I cannot accept using tools that I don’t really need, but, at success time, I cannot really get rid off anymore easily. And then suddenly that $10/mo is $23k/mo and I knew that when I integrated it in the first place, but thought it will save me 2 hours per year. And the new price utility is not suddenly saving me 100 or more hours a year; more like 4 hours.
If you have employees, you’re paying hundreds of thousands of dollars per year for them. Paying an extra $10/m to save 2 hours of their time is still a profitable trade.
Two things: the first is that it's easier to estimate costs with services like 1P due to it being mostly static, unlike other services where costs are quite dynamic ranging from different factors like usage and hours spent. The second is that once you reach that scale (more than 10 members), your costs are so large on personnel alone that $10/person is an absurdity to be frugal about when it can save you time.
I don’t doubt it’s possible, but I would love an example of a Saas product that successfully scales pricing from $10/m to $23,000/m on customer segments that aren’t enterprise-scale and has the kind of lock-in you describe.
I ask because I would like to quit my job and build a competitor ASAP.
Also, the problem isn’t the potential cost down the road. If you’re scaling on any Saas product to the $23k/m tier, then congrats, you’ve made it!
The problem is, there’s a 90%+ chance all your cost optimizing upfront was a giant waste of time, because a vast majority of startups don’t work out. And the ones that do often pivot.
It depends a lot on your time and experience... Also, what you want to know. There's no guarantees at all with a SaaS startup, and if you can increase your knowledge base at a minimal time sink it may be worth it. I have a dozen things I tinker with and no expectations, so I treat it more as a hobby.
I am spending about $140/mo on a dedicated server currently, mostly because I want to run my own mail server without depending on a relay host or nickel and dime costs ($2-10/mo per address). This now runs just about everything I use. Nearly all my apps are dockerized so I can port them to another host pretty easily.
I've also been playing with writing for Cloudflare directly (workers & pages) as well as CockroachLabs (CockroachDB Cloud). Both with relatively low startup pricing and can grow as needed based on usage/demand, both with a relatively clean self-host strategy as an exit.
It would be too easy for me to wind up paying $20-30/month or more on a dozen or more projects. I try to keep this down, as if I don't touch something for a month, or don't complete it for a year, it adds up quick... If I keep baseline costs low, it's less of an issue.
If I was 20 years younger, I'd probably have gotten more of these things done or put more effort in. The tech costs are a lot better today than 25-30 years ago when I was starting out. But I'm still pretty mindful of it.
> Does nobody here see the irony in bending over backwards to avoid spending money on software products when the entire purpose of your business is to sell a software product?
No, because nobody thinks that it'll happen to them. Their company's/startup's/side project's product will be worth paying money for, and surely nobody's going to try to start an open-source product that will compete with them.
I think others have addressed your comment well, but I'll add my 2c:
I think it's more about minimising the cost with effective alternatives rather than just taking whatever the "brand name VC startup/scaleup" is in the space and forking over hundreds of dollars.
You will more likely always have to spend money to make money, I agree. But if starting from nothing, having an immediate -$200/month (up to $500+ once the free credits run out) before you actually make any money does not sound like a good time.
Also take into account when running multiple such projects. Survivorship bias in our space is real, and it takes a lot of hits to "make it big". That can easily pile up to thousands for a handful of projects while none of them get near break-even before you go bankrupt.
All it takes is finding a good-enough alternative in the space, or if you trust the software enough to self-host, do that.
There's nothing wrong with actually trying to be frugal. We're supposed to have the skills to be more self-sufficient in this space, and I don't think it's a negative to want to actually do so. And neither is paying for something if you feel the premium is worth it.
Pay more, no matter how unreasonable it is because you can afford it. You’re living in the US bubble, where VC funded startups feed each other in often unsustainable ways. People were taught this argument to keep the wheel spinning.
The right way to build a startup is: offload your infra, self host everything you can while balancing the maintenance effort/cost.
I used to be CTO of an Uber Eats style service that ran on hosting costs of about £100pm. Thousands of orders every few days, real time driver tracking, self hosted invoice generation, the works. Hetzner servers, Docker Swarm and Cloudflare for the tech.
It is truly amazing to me the value that is provided by renting hardware instead of VPSes. As long as you're willing to roll your own infrastructure instead of buy into a cloud provider's infrastructure.
I did that once and the data center burned down. Sure I could have the service spread over several centers and build distributed backups etc.
In the end self hosting and self managing is a money / time trade off, especially for a side gig I’d us SAS and managed solutions. The one thing one has to make sure off is to not get locked in with a particular provider, so knowing how to do everything yourself is a very valuable skill.
To be fair, Datacenter burning down is pretty much on the bad luck side of risk management. It's in the same category as your distribution center being hit by an earthquake... I'd guess it wouldn't happen again, but who knows...
Didn't that happen from someone leaving a sink on or something, causing flooding and then the UPS batteries to short and explode?
I was there when someone at AWS accidentally unplugged an entire region.
Shit happens, it doesn't matter where it's hosted. People act like the cloud is infallible or something. You're literally sending lightning bolts worth of electricity through bricks of metal. Anything can happen.
If you know what you're doing it's easy enough to roll out a multi-region distributed system with HA and backups on a pretty modest (<100pcm) budget that can handle competitive QPS.
However, most people do not - some will learn, but most will fall for the cloud marketing depts and become infra renters for life. Teach a man to fish, and so on.
It's a pretty common abbreviation of high availability on this context, with a heavy implication on active-passive redundancy (although the GP looks like an exception here). It's more used than the full wording, and part of the name of some important tools.
So, yeah, it's good that you asked, because it's not as widely known as the people that use it think it is.
Another reason is that if you look for an investor, one of the first things they ask is how / where you host it. If it is anything remotely DIY you'll be turned down.
It was actually painful to see start ups spending thousands a month on hosting, that could be easily achieved as you say under 100 pcm plus. They would have to get a contractor to set it up and for support, but it would have worked out much cheaper and they could have bumped the salaries of their workers.
I'm sure you can guess - pure risk aversion. Your business idea is risky enough, and they would need engineers to assess your (possibly ever-changing) DIY stack.
You see the same thing in the corporate world for in-house stuff. Your manager (and your manager's manager) don't want to hear about in-house or self-hosted things that AWS can provide.
This is totally understandable. It's a repeat of the whole "nobody ever got fired for buying IBM" mantra of computing's early decades.
> some will learn, but most will fall for the cloud marketing depts and become infra renters for life
Do you have any learning recommendations for someone looking to start down this path? I've only ever worked in an infra-renter context, and I've begun exploring the 'rent from Hetzner, manage your own infra' for personal projects, but I would love to learn from the paths of experts where possible.
I'm no expert but hopefully I can still point you toward the happy path: start very small and increase distributed complexity at your own pace until you can fully appreciate the entire end-to-end system and all the processes involved. The book: DDIA is a well-known 101, if a little primitive, and has references you can dig into as well.
Ideally, you also have some exposure to this at $job as simply building DIY infra horrors without seeing the real-world context, tradeoffs, etc. in which they typically operate will be misleading.
I'm working on a middle-tier solution for this gap -- I call it Nimbus, but basically the idea is to provide managed services at low cost cloud prices.
There's no reason someone should have to run a service like Chatwoot themselves, the software is so good that it's mostly set & forget for most small use cases.
That's where I come in. Unfortunately I don't have ChatWoot yet, but I have (and use) Umami for page view tracking extensively on my own projects now, with Nimbus[1]. The dogfood tastes decent so far.
Yeah, but all improvements will likely be upstreamed/made open source (their license isn't AGPL or anything but just makes sense to me since it's MIT).
The first thing I want to do is add a backup mechanism that isn't just take a snapshot -- There are a bunch of similar tools to Umami and I don't think that any of them have a really good cross-project way of taking backups.
Feels like there should be a page view/analytics backup standard, so you can easy move from a tool like Plausible or Umami and try out a new one, like Fugu.
But outside of advanced functionality I think my platform is just a lot closer on cost. The instance costs don't go up per traffic served (especially since 99% of people won't need that) -- it's more like parts + maintenance (and since Umami is good software it doesn't need a TON of maintenance either, just regular patches and some monitoring/extremely light resilience engineering).
Hetzner is great and I love it and use it for all my dev needs. But there availability is in limited zones and for production I do suffer from the latency. Haven't found any other provider which costs anywhere near the same(every comparable server is at least 2x the cost). I use their AX41-NVME. Any suggestions of an alternate?
I've been pretty happy with my OVH server, but agreed on support... had better luck with chatgpt and a lot of additional reading getting my CIDR block of addresses configured under ProxMox (which made me far less worried about ChatGPT taking my job any time soon).
In the end, was an interesting learing experience, that I hope to not have to repeat. Only went for a single server as I had several smaller VPSes on DigitalOcean and wanted to add a mail server in the mix, and couldn't reliably send via DO or Linode, so was easier to consolidate and run on a single/larger host for hobby projects.
Vultr has some very cheap offerings and maybe worth a look. I can't speak to their reliability or service, but they have DCs everywhere even in South America and Africa where there are few cloud options outside of AWS. They're more expensive than Hetzner, maybe even a little over your 2x range depending on your load, but if latency is your issue they probably got you covered.
You'd be amazed how much you can get out of one of the Hetzner $3/mo ARM servers with the right code.
I use a $6/mo box for my primary business hosting, but I have a $3/mo one that I'm using to build v2, really just to prove what's possible. If you set up your DB and caching right you can do so much with so little...
Those ARM boxes are incredible for the price, I'm using them to document a hobby K8S cluster because the overall cost is low enough that it won't price a hobbyist out.
Similarly, but on the opposite spectrum, I also wonder how far you can go on vertical scaling nowadays. For 200 EUR/month on Hetzner you get a dedicated 80-core ARM CPU, 128GB ECC RAM, 2TB SSD...using a good performance multi-threaded language, what _can't_ you run on that? It's ridiculous value.
> what _can't_ you run on that? It's ridiculous value.
Yeah - I remember a StackOverflow talk[1] where they basically said that they just vertically scale their database.
The fact that they were able to make it work tells me that most businesses should probably just go that route and avoid the headache of distributed systems[2].
2. Obviously a business should probably invest in redundancy when it comes to data (as did StackOverflow), but a pure "Raid 1" setup is the easiest of distributed systems to understand.
People used to make websites by uploading some php or perl scripts to a shared server (or just buying a domain and pointing it at their home ip). In that context, I can't imagine someone writing an article like this.
I think it's kind of sad and insane that people seem to have forgotten (or came into the industry too late to know) that it's possible to build fun and useful things that way.
I built a forum/social site that way back in the day and it was active and profitable for 10+ years.
It also feels like the art of optimization has been lost.
For small and medium sites a lot of today's crazy build pipeline and distributed asset hosting complexity can be sidestepped if you just focus on optimization and making sure that cache expiration dates for your assets are set correctly.
On the server side, it is considered "normal" these days to have maybe 50+ database queries per request. People then reach for expensive and complex database solutions (clustering, etc) before doing simple app-layer optimizations like caching.
The "shared server" concept is funny because as much as it seems old fashioned, it embodies the spirit of all the modern things:
* Serverless - in that you don't need to do server admin
* Online IDE - CPanel lets you make changes online to your code
* Continuous Integration - When you save the file online it is in immediate production
* Branching - you can copy a folder from one subdomain folder to another - this effectively gives you a poor man's git and environments in one swoop!
It is like a REPL of web dev. Nothing can come close to it, because once you made your changes they are live, there is no infuriatingly slow publishing step.
There are a lot of use cases for which you can’t actually save by kicking out Ahrefs. Of all the tools here it is the hardest one to replicate. I agree w the rest - but in many situations Ahrefs is not replaceable.
Its provides detailed analytics on the popularity of different websites and the keywords that people search for; it goes really, really deep on this, much deeper than you can go through manual google searches and various scripts. You can easily recreate Vercel on your own if you're willing to forgo some convenience, for example; there's no equivalent to that for Ahrefs - you can't realistically 'roll your own' or code your own Ahrefs. The web largely works on an advertising model, and Ahrefs is what tells you what has value for advertising purposes and what people will visit, or are visiting now. If you're doing something simple, it may not matter. If you're not, Ahrefs is indispensable.
> I use AWS EC2 for Waitlist's backend python webserver, RDS for the postgres database, S3 for file storage, and an Elastic IP. I use T2 and T3 instances. My AWS bill would be $200-300 a month
Seems like a high bill for what sounds like basically a hosted email list? I would have expected compute and storage costs to be very low.
E: This was NOT a point about how cloud costs are ridiculous. Even for cloud this seems high.
> except it's free because I got $5,000 in credits from AWS Activate
Free credits are the ingenious marketing ploy of all big clouds to delay price comparisons until some amount of lock-in is achieved. Sure, you could likely host it for 1/10th to 1/5th of the cost elsewhere. But then you would have to pay from the get-go, and on AWS it's free™ (until it isn't)
RDS is very expensive. Just following the console's advise for production instances (meaning Multi-AZ with a Proxy), a t3-medium (2 vCPU, 4GB) Postgres instance with 5GB storage is $130/month.
You can get it down to $12/month by going with a t4g.micro (2 vCPU, 1GB RAM), single-az, no proxy, no backups, 5GB storage; with price after that scaling roughly linearly with RAM. But the interface will scream at you the entire time that this isn't reasonable for a production setup.
We thought that at the SaaS I’m working at but AWS scheduled maintenance in our AZ and that led to significant downtime with nothing we could do about it, no compensation, nada. It was basically our fault because we didn’t use multi-AZ. I can’t remember if this was RDS or ECS but we figured if it could happen to one critical service in a single AZ, it could happen to another.
I think Proxy _is_ overkill and perhaps expensive for what it is. For a pg install, running pgBouncer isn’t difficult. But I guess that depends on whether you’re authenticating with IAMs, which doesn’t sound trivial with bouncer (it might be).
Multi-AZ is freaking awesome because the literal first 10 minutes I'm forced to give a shit about my RDS DB, its already a waste of money vs the monthly cost.
Or just rent a bare-metal server with 32 real cpus and 128gb of ram for <100 bucks a month... My bare metal k8s cluster running on rented bare-metal costs <300 a month and I have hundreds of cores and hundreds of gb of ram available.
Is it really that hard to install netdata? It's a single command.
It is really no different than managing VPS's, in all honesty (just less expensive and tangible). I spend less than one hour a week doing updates and routine maintenance. Drive failures happen rarely, and can usually be swapped out within a few hours (this is what RAID is for).
More time is spent dealing with k8s and it's shenanigans than dealing with the actual servers.
Many people don't do that. My company uses EKS and we just send dockerfiles and yaml files at it. We don't ever actually interface with an ssh prompt on our servers
I'm so thankful my career as a developer and sysadmin preceded the cloud era. We now have a whole generation who manage cloud infrastructure but have never setup a Linux server from scratch.
I am sure there are numerous services costing $5-10/mo that would cover this and I would certainly consider that to eliminate the hosting and maintenance cost.
If I had to build it, I would expect using Lambda, DynamoDb and S3 would likely bring the hosting cost to down to a couple of a dollars a month max.
I'm not sure what you're trying to say? I have no idea what your 1M requests are doing. They could be 1M requests for static HTML. It does make the monthly costs seem a bit more understandable, but 1M/24hrs is still only ~12 RPS.
Btw my comment wasn't an insult, "hosted email list" genuinely seems like the primary functionality and I wouldn't have expected it to use many resources.
1Password probably has best UX. You are thinking of Lastpass for security problems, 1Password has not had a breach. Source: went through all of these as possible options for the company I work for last month
Be careful with using their CI tho, even if it's good, you hit the limit quite fast (especially storage, not minutes; and they don't report properly where the storage expenditure comes, so it's very frustrating).
You can quite easily plug in your own CI runner on Github and reduce the cost by 20 times or more, especially if you're already running k8s. (I wrote a post about it for the cheap managed k8s service I built https://symbiosis.host/blog/using-actions-runners)
I use ARC and I wouldn’t call it trivial to set up. Took awhile to tweak all of the things necessary to get it working right. It is way cheaper though once it is working
I didn't know that some of these tools offer a startup credit program. I know some of the tools mentioned would have been helpful for a project that I have been working on in the last couple of years[0].
Interesting, thanks for sharing . Especially Ahref, Intercom and at some point the AWS bill make up the vast majority of the expenses. Chatwool on Railway for 5$/month would probably do the trick instead of intercom. At first glance 200-300$ per month for AWS once your credit runs out seems like _way_ too much. If I understand your product correctly then you serve customizable widgets for signup campaigns and persist the emails with some metadata for analytics in a database. I would like to be proven wrong but to me that sounds like a 29$/month supabase pro bill or a 20$/month neon.tech bill if you like managed services. Or given the numbers on the landing page this should(!) easily be doable with a hetzer server for about 30$/month. Depending on the exact usage Cloudflare R2 might also be significantly cheaper than AWS S3.
It seems to work pretty well, it could be entirely self-hosted, but I like using a 3rd party service, mostly to make sure the emails are delivered and because I don't yet have my own self-hosted mail service (planning to start one soon, especially for such transactional emails that I receive myself and can make sure they don't arrive in spam).
Are we the only ones? Do our users think differently? I have wrongly assumed that my users think like me before. Is there any research showing how Intercom is received by users?
I would say technology wise, you don't need to launch with more than a VM for hosting if frugality is important to you. That way your costs don't explode. I would put the rest of my money into the sales pipeline. I think the requirements for initial scale are usually grossly overestimated, and the throughput of a simple $30 VM underestimated. Unless I built my stack out of mollasses, or I was anticipating 1000s of concurrent users, concurrent being the important part, then I wouldn't touch cloud stuff personally.
2. IMHO intercom is overpriced relative to competitors depending on how you use it, with the caveat that I hate in-app messaging widgets and consider them a blight on the modern web. (HelpScout is my preferred app.)
A friend of mine who was at one point a GH employee gave me a (lifetime? it was around four years ago) F&F discount that IIRC amounts to $200/mo free across all SKUs.
> Auth: I know it's increasingly trendy to outsource auth to a paid vendor but I'm not a fan of the lock-in and I like having full control over the experience.
I use very little of the outbound/link-building things, but find the alerting on broken content and notifications of which specific pillar pages are hitting traction very actionable and useful.
> I use AWS EC2 for Waitlist's backend python webserver, RDS for the postgres database, S3 for file storage, and an Elastic IP. I use T2 and T3 instances. My AWS bill would be $200-300 a month
My guess is that RDS is eating up the cost here?
I've been using Litestream on datasciencesouth.com - it has flaws but if you can get away with it, it's much cheaper than RDS.
The thing about RDS is that you’re paying for database backups and failover etc to be someone else’s problem. Given that the data is often an extremely critical art of the application and the high chance that you are probably not doing it right if you’re not an expert it’s often worth it to pay for managed db. RDS is a pretty raw deal in that space though. Someone else rightfully pointed out that you can get Supabase managed db for $30 USD per month which is significantly cheaper
It would be nice if we can see a breakdown of P&L. I see the indie dev claims to have “over 7500+ businesses” as customers. But also a very generous free tier.
I see the post is getting eviscerated in the comment section. But if this person is spending mere cents in operating costs when compared to the profit earned each month or year (I think in business terms this is known as customer acquisition costs). Then I would say that is truly “frugal”
I think people want to nitpick certain things they have expertise on and say that you can save $XYZ if you know what you’re doing. Well, duh.
On the other hand, while OP’s stuff generally looks like some good decisions were made, there are a few items in there that just seem like criminally bad value for what you get and you wouldn’t spend near that much time and effort switching to something much cheaper. Vercel stuck out to me as one of these
Yeah but let's say your time is worth $60/hr. If you save only 40 minutes per month then you've broken even. Basically one lunch break. Some things are just a hassle to do, personally I would try to self host the Next app but I definitely see where the author is coming from purchasing Vercel.
I love this list. One thing I'll add is I like to avoid using S3 or google's cloud storage AT ALL until I have > 30 GB. i.e. with google free tier of their compute engine you get a 30GB hard drive, 1GB ram, and two AMD EPYC 7B12 2250 MHz processors. So I make a fake bucket system using those free 30GB:
May be a weirder question to ask, but curious about the Delaware LLC decision. If I was creating a C-corp, Delaware makes sense, but for an LLC it feels weird unless the author is in Delaware. I also don’t see any cost for registering as a foreign LLC wherever they are located, so maybe they are in Delaware.
Not OP & IANAL… by having the LLC in Delaware, it’s probably nominally easier to convert to a Delaware C corp at the desired time. Or maybe OP lives in Delaware.
As you said it’s not that much harder to convert. For startups it feels like it’s just “the thing you do” when in reality it ends up doubling your paperwork at the end of the year come tax time and siphoning off critical money if you’re bootstrapping. Sure if you have VC pockets go ahead and do it but for everyone else in bootstrap land just make your LLC in your home state if you live in the US. You’ll save yourself time and money.
You don’t get any tax benefits having the LLC in Delaware since you still have nexus in your home state and you’ve just made your situation more complicated by also having to file the nexus paperwork on top of your Delaware paperwork and fees. There are some ancillary things like being able to hide you own the LLC if you register in Delaware (or a few other states like Wyoming) but those don’t really matter for the vast majority of people.
When I formed my first company, I tried to min/max my costs and tax savings. In the end I accomplished neither buy forming a California LLC with C-corp taxation. I did a very good job of not correctly reading a DIY book how to set up your own business. When my accountant saw it, he said “Wow, I’ve never seen one of these before. You picked the absolute worst possible choice. I’m going to show this to my friends!”
Linear is more of a UX thing, not quite the same product. GitHub issues is steadily working to be better though. They’ve basically already made Zenhub pretty much obsolete in recent releases
> but self-administering the modern React deployment chain is tedious
Is it? It's not the React part that's adding to the tedium, I am sure; that part can be trivially automated away. Is it any more tedious than self-administering the deployment of older-school php apps?
I run a Vercel equivalent in Cloud Run’s free tier. The fact is that Vercel charges a LOT for what you get. IMO the UX is nice but what they offer is not something that you critically can’t build yourself in an extra day of coding
40 dollars for hosting your frontend is bit much you might want to check other alternatives like https://www.stormkit.io they allow you to deploy other cdns as well. Or use something like github pages.
If anyone looking for free alternative for certain tool that you can use to help your startup. Here is the open (re)source project: https://freestuff.dev/ (free stuff for developer)
218 comments
[ 5.5 ms ] story [ 232 ms ] thread>No surprises here. I used to use Plausible, but that costs money and you can't beat free. Monthly cost: $0.
Thanks hard pass. I checked and Plausible starts from 9 Euros a month. This is hardly what I'd consider advertising material. But likely I'm just not their target audience.
There's also no consent banner, so this doesn't seem GDPR compliant
Edit: I must have phrased this in a confusing way. They are NOT using Plausible but GA. I'm well aware cookie-less analytics does not require consent
Edit: If you mean this is stated in the blog post, I can't find it
I'm not sure what you mean by "consent banner". If you're referring to an annoying "allow us to track everything you do" pop-up, that's probably because Plausible doesn't attempt to track everything you do.
They have plenty of privacy information on their site [1].
[1]: https://plausible.io/privacy-focused-web-analytics
Plausible uses this one weird trick where they don't use cookies and don't store PII, and thus don't need consent (neither for cookies nor for GDPR). You lose some data because you can't track users across visits, but you gain accuracy (because you can track every visit, not just people who agree) and don't have to annoy your visitors.
And once you have that you can also track visit time a bit better, or track goal conversions like how often people click some specific button. But other than that it's basically a more modern webalizer.
Also, last year I found a bug in Plausible's script which silently stops it working on some pages, so your data won't be accurate in those scenarios either.
Tangent, but cookie-less analytics may also require consent. It's about PII, and cookies are just one mechanism to tie information to a person.
For example: browser fingerprinting doesn't use cookies, but serves the same purpose.
The problem is that one does not simply build “cloud agnostic” services and you would probably spend a not insignificant amount of time migrating between them every year or two. But if you were really good at cloud and could do this efficiently and quickly it would probably be worth it
Why? From personal experience, everything I do is either:
- a static site, which you can run on anything from Netlify to some static CDN to GitLab Pages to a potato, or
- a bunch of Docker containers, which I usually run on Docker Swarm on any random VPS, and manage with my own little tool, https://lunni.dev/; AWS of course has some “cloud native” options for that, too (ECS + Fargate?), but I find predictable monthly bill a bit better on my nerves.
This and some S3-compatible storage do wonders for me.
But I get it. If X tool saves time, why not pay for it? I wouldn't call that frugal though because you are still spending where you can save.
Ubersuggest has a comparison page[0]. From my understanding it boils down to Ubersuggest being cheaper and most likely enough for what most users will need, but Ahrefs has a larger tool suite (and much higher cost).
[0] https://neilpatel.com/blog/ubersuggest-vs-ahrefs/
Use one of the OSS password managers to save on 1P. If you have a cloud sync service you already use, KeePassXC + cloud sync will solve having the passwords available everywhere you are. You can even use Syncthing for free (though I'd at least donate once off) since you can use it peer-to-peer between your devices.
SEO really needs disruption here on price. Ubersuggest seems decent.
On Intercom: OP can probably go into more detail on why Intercom's widget is so much better. If it's just animations and UX, then yeah go find some alternative or self host.
What's useful in the self-hosting space is finding either a tool to manage server provisioning & management (e.g. Ploi), or a hosting provider that makes it easy to install (e.g. DO marketplace). Then you're only spending a small fee on the management and only spending on hosting cost (which for a long time can be negligible).
Spending some time to DIY with reliable tools (automated where possible) is IMO still preferable to spending this much money if starting from scratch.
I could see if we were talking about a side project with no commercial intent. But this post is about an indie Saas founder...if a few hundred dollars a month is a barrier to success, then that's a pretty good sign your indie startup isn't starting up.
I find these threads endlessly amusing considering that, for hundreds of years, starting a business involved massive investments in real estate, physical equipment, etc. We're living the dream, yet somehow quibbling over spending $10/month on a piece of software. If it saves you just 2 hours, per year, then it's worth it.
In the context of a business, nearly any piece of software is a trivial cost compared to hiring a human (or yourself) to do a task.
If there was a magic debugger that made debugging 100x easier and faster, but it wasn't free and open source, I'd still have a hard time using it even though it would save me a bunch of time, and easily pay for itself in time and frustration saved.
And most software I deal with isn't exactly free of opportunity cost - I dabbled a bit with accounting tools to the point of frustration and time sunk that I decided to roll my own (built around ledger, in case anyone cares) in about a day. Writing my own tool is sometimes faster for me than trying to figure out how to use one that's not designed around my specific needs.
Sure, 10$ doesn't seem like much, but all these little invoices can creep up to a non-trivial sum.
I find the cost and value of tooling options quite difficult to fully grasp, it often comes down to gut feeling IMO.
What I would want to caution folks for is that the cost of a tool is not only what it actually costs. It's also the amount of time that has to be invested into usage, and potentially migrating to something else a while down the line in case of discontinuation, disappearing capabilities, new requirements, ToS or pricing changes, ...
I ask because I would like to quit my job and build a competitor ASAP.
Also, the problem isn’t the potential cost down the road. If you’re scaling on any Saas product to the $23k/m tier, then congrats, you’ve made it!
The problem is, there’s a 90%+ chance all your cost optimizing upfront was a giant waste of time, because a vast majority of startups don’t work out. And the ones that do often pivot.
I am spending about $140/mo on a dedicated server currently, mostly because I want to run my own mail server without depending on a relay host or nickel and dime costs ($2-10/mo per address). This now runs just about everything I use. Nearly all my apps are dockerized so I can port them to another host pretty easily.
I've also been playing with writing for Cloudflare directly (workers & pages) as well as CockroachLabs (CockroachDB Cloud). Both with relatively low startup pricing and can grow as needed based on usage/demand, both with a relatively clean self-host strategy as an exit.
It would be too easy for me to wind up paying $20-30/month or more on a dozen or more projects. I try to keep this down, as if I don't touch something for a month, or don't complete it for a year, it adds up quick... If I keep baseline costs low, it's less of an issue.
If I was 20 years younger, I'd probably have gotten more of these things done or put more effort in. The tech costs are a lot better today than 25-30 years ago when I was starting out. But I'm still pretty mindful of it.
That implies bare minimum spend not splurge-in-just-a-few-areas cost to time savings benefit analysis.
No, because nobody thinks that it'll happen to them. Their company's/startup's/side project's product will be worth paying money for, and surely nobody's going to try to start an open-source product that will compete with them.
It's always somebody else.
I think it's more about minimising the cost with effective alternatives rather than just taking whatever the "brand name VC startup/scaleup" is in the space and forking over hundreds of dollars.
You will more likely always have to spend money to make money, I agree. But if starting from nothing, having an immediate -$200/month (up to $500+ once the free credits run out) before you actually make any money does not sound like a good time.
Also take into account when running multiple such projects. Survivorship bias in our space is real, and it takes a lot of hits to "make it big". That can easily pile up to thousands for a handful of projects while none of them get near break-even before you go bankrupt.
All it takes is finding a good-enough alternative in the space, or if you trust the software enough to self-host, do that.
There's nothing wrong with actually trying to be frugal. We're supposed to have the skills to be more self-sufficient in this space, and I don't think it's a negative to want to actually do so. And neither is paying for something if you feel the premium is worth it.
The right way to build a startup is: offload your infra, self host everything you can while balancing the maintenance effort/cost.
It is truly amazing to me the value that is provided by renting hardware instead of VPSes. As long as you're willing to roll your own infrastructure instead of buy into a cloud provider's infrastructure.
In the end self hosting and self managing is a money / time trade off, especially for a side gig I’d us SAS and managed solutions. The one thing one has to make sure off is to not get locked in with a particular provider, so knowing how to do everything yourself is a very valuable skill.
I was there when someone at AWS accidentally unplugged an entire region.
Shit happens, it doesn't matter where it's hosted. People act like the cloud is infallible or something. You're literally sending lightning bolts worth of electricity through bricks of metal. Anything can happen.
Uh huh. This is why DevOps/SRE roles pay more than development roles.
“Step 2: draw the rest of the owl”
However, most people do not - some will learn, but most will fall for the cloud marketing depts and become infra renters for life. Teach a man to fish, and so on.
So, yeah, it's good that you asked, because it's not as widely known as the people that use it think it is.
It was actually painful to see start ups spending thousands a month on hosting, that could be easily achieved as you say under 100 pcm plus. They would have to get a contractor to set it up and for support, but it would have worked out much cheaper and they could have bumped the salaries of their workers.
Why is that?
You see the same thing in the corporate world for in-house stuff. Your manager (and your manager's manager) don't want to hear about in-house or self-hosted things that AWS can provide.
This is totally understandable. It's a repeat of the whole "nobody ever got fired for buying IBM" mantra of computing's early decades.
It also totally sucks.
Do you have any learning recommendations for someone looking to start down this path? I've only ever worked in an infra-renter context, and I've begun exploring the 'rent from Hetzner, manage your own infra' for personal projects, but I would love to learn from the paths of experts where possible.
Ideally, you also have some exposure to this at $job as simply building DIY infra horrors without seeing the real-world context, tradeoffs, etc. in which they typically operate will be misleading.
1. a DNS monitoring with failover (DNSMadeeasy has a decent solution)
2. a Haproxy setup with health checks for switching to a working upstream service
3. a distributed filesystem
4. a master-slave replication with monitoring (something like Mariadb + orchestrator service)
and nightly backups for all this. Database and FS are latency sensitive so they shouldn't be too far apart.
There's no reason someone should have to run a service like Chatwoot themselves, the software is so good that it's mostly set & forget for most small use cases.
That's where I come in. Unfortunately I don't have ChatWoot yet, but I have (and use) Umami for page view tracking extensively on my own projects now, with Nimbus[1]. The dogfood tastes decent so far.
[1]: https://nimbusws.com/managed/umami/
The first thing I want to do is add a backup mechanism that isn't just take a snapshot -- There are a bunch of similar tools to Umami and I don't think that any of them have a really good cross-project way of taking backups.
Feels like there should be a page view/analytics backup standard, so you can easy move from a tool like Plausible or Umami and try out a new one, like Fugu.
But outside of advanced functionality I think my platform is just a lot closer on cost. The instance costs don't go up per traffic served (especially since 99% of people won't need that) -- it's more like parts + maintenance (and since Umami is good software it doesn't need a TON of maintenance either, just regular patches and some monitoring/extremely light resilience engineering).
In the end, was an interesting learing experience, that I hope to not have to repeat. Only went for a single server as I had several smaller VPSes on DigitalOcean and wanted to add a mail server in the mix, and couldn't reliably send via DO or Linode, so was easier to consolidate and run on a single/larger host for hobby projects.
I use a $6/mo box for my primary business hosting, but I have a $3/mo one that I'm using to build v2, really just to prove what's possible. If you set up your DB and caching right you can do so much with so little...
You get a lot of bang for your buck out of them.
Yeah - I remember a StackOverflow talk[1] where they basically said that they just vertically scale their database.
The fact that they were able to make it work tells me that most businesses should probably just go that route and avoid the headache of distributed systems[2].
---
1. https://www.infoq.com/presentations/stack-exchange/
2. Obviously a business should probably invest in redundancy when it comes to data (as did StackOverflow), but a pure "Raid 1" setup is the easiest of distributed systems to understand.
I built a forum/social site that way back in the day and it was active and profitable for 10+ years.
It also feels like the art of optimization has been lost.
For small and medium sites a lot of today's crazy build pipeline and distributed asset hosting complexity can be sidestepped if you just focus on optimization and making sure that cache expiration dates for your assets are set correctly.
On the server side, it is considered "normal" these days to have maybe 50+ database queries per request. People then reach for expensive and complex database solutions (clustering, etc) before doing simple app-layer optimizations like caching.
* Serverless - in that you don't need to do server admin
* Online IDE - CPanel lets you make changes online to your code
* Continuous Integration - When you save the file online it is in immediate production
* Branching - you can copy a folder from one subdomain folder to another - this effectively gives you a poor man's git and environments in one swoop!
It is like a REPL of web dev. Nothing can come close to it, because once you made your changes they are live, there is no infuriatingly slow publishing step.
3.375$ a year: call it frugal...
Seems like a high bill for what sounds like basically a hosted email list? I would have expected compute and storage costs to be very low.
E: This was NOT a point about how cloud costs are ridiculous. Even for cloud this seems high.
Free credits are the ingenious marketing ploy of all big clouds to delay price comparisons until some amount of lock-in is achieved. Sure, you could likely host it for 1/10th to 1/5th of the cost elsewhere. But then you would have to pay from the get-go, and on AWS it's free™ (until it isn't)
I have a hard time understanding where the cost is coming from. I can only assume it's because of the DB as opposed to the instance or S3.
I wouldn't be surprised if it was over 10x cheaper if you didn't use cloud hosting.
I am not sure I know the reason behind it honestly.
Either you go with something like Dynamodb or planetscale or the least you pay is 40€/m using AWS Aurora serverless.
As soon as databases start to charge for computing time the cost balloon out of reach of hobby projects.
You can get it down to $12/month by going with a t4g.micro (2 vCPU, 1GB RAM), single-az, no proxy, no backups, 5GB storage; with price after that scaling roughly linearly with RAM. But the interface will scream at you the entire time that this isn't reasonable for a production setup.
I think Proxy _is_ overkill and perhaps expensive for what it is. For a pg install, running pgBouncer isn’t difficult. But I guess that depends on whether you’re authenticating with IAMs, which doesn’t sound trivial with bouncer (it might be).
It is really no different than managing VPS's, in all honesty (just less expensive and tangible). I spend less than one hour a week doing updates and routine maintenance. Drive failures happen rarely, and can usually be swapped out within a few hours (this is what RAID is for).
More time is spent dealing with k8s and it's shenanigans than dealing with the actual servers.
Many people don't do that. My company uses EKS and we just send dockerfiles and yaml files at it. We don't ever actually interface with an ssh prompt on our servers
If I had to build it, I would expect using Lambda, DynamoDb and S3 would likely bring the hosting cost to down to a couple of a dollars a month max.
Btw my comment wasn't an insult, "hosted email list" genuinely seems like the primary functionality and I wouldn't have expected it to use many resources.
It is cloud software. I expect them to get hacked sooner or later.
Be careful with using their CI tho, even if it's good, you hit the limit quite fast (especially storage, not minutes; and they don't report properly where the storage expenditure comes, so it's very frustrating).
How do you reduce the cost of zero?
I didn't know that some of these tools offer a startup credit program. I know some of the tools mentioned would have been helpful for a project that I have been working on in the last couple of years[0].
[0]: https://zesfy.com/
It seems to work pretty well, it could be entirely self-hosted, but I like using a 3rd party service, mostly to make sure the emails are delivered and because I don't yet have my own self-hosted mail service (planning to start one soon, especially for such transactional emails that I receive myself and can make sure they don't arrive in spam).
https://buttondown.email/stack
---
A couple notes on where I differ/concur from OP:
1. ahrefs is absolutely worth its price in gold.
2. IMHO intercom is overpriced relative to competitors depending on how you use it, with the caveat that I hate in-app messaging widgets and consider them a blight on the modern web. (HelpScout is my preferred app.)
3. I deeply miss my Free AWS Credits Era.
> …it's free thanks to a generous friends and family plan
Amen. I thought I was alone in this.
My guess is that RDS is eating up the cost here?
I've been using Litestream on datasciencesouth.com - it has flaws but if you can get away with it, it's much cheaper than RDS.
I see the post is getting eviscerated in the comment section. But if this person is spending mere cents in operating costs when compared to the profit earned each month or year (I think in business terms this is known as customer acquisition costs). Then I would say that is truly “frugal”
On the other hand, while OP’s stuff generally looks like some good decisions were made, there are a few items in there that just seem like criminally bad value for what you get and you wouldn’t spend near that much time and effort switching to something much cheaper. Vercel stuck out to me as one of these
https://github.com/andrewarrow/feedback/tree/main/filestorag...
All the code I write is in bucket form, ready to flip a switch and use real buckets but 100% free up to that hard drive limit.
You don’t get any tax benefits having the LLC in Delaware since you still have nexus in your home state and you’ve just made your situation more complicated by also having to file the nexus paperwork on top of your Delaware paperwork and fees. There are some ancillary things like being able to hide you own the LLC if you register in Delaware (or a few other states like Wyoming) but those don’t really matter for the vast majority of people.
Is it? It's not the React part that's adding to the tedium, I am sure; that part can be trivially automated away. Is it any more tedious than self-administering the deployment of older-school php apps?
If anyone looking for free alternative for certain tool that you can use to help your startup. Here is the open (re)source project: https://freestuff.dev/ (free stuff for developer)