I don’t know @arp242, but recently when I asked how to do something on the Vi Stack Exchange, he said “hm, you can’t do that” and then landed a patch in Vim to implement the feature I was asking for. So it seems safe to say that he’s a friendly person and an asset to the community.
Your life will be much easier with 3 customers paying $500, than 500 customers paying $3.
I get that it seems nice to have an accessible price that’s fair for tiny usage. But lower barrier to entry attracts customers who aren’t invested in the problem, won’t have any developer to help, and may not even fully understand what your product does, but at any price, as paying customers they’ll expect to be treated seriously and demand that you make your product work for them.
I’d take personally take on tremendous amounts of pain to have 500 people paying $3 dollars a month, though I’m still pretty skeptical about the amount of pain involved.
Why not just do both and tier based on hits and domains like it already is? Can you not just tell the $3 dollar tier to expect basically no support?
For every SaaS companies I worked for (3), most of the support pain was coming from the lower-tier customers.
There are people out there that feel entitled regardless on the level of support that was advertised. It’s quite emotionally draining to read these messages. And because they contain a lot of negative emotion it’s hard not to reply to them in the hope to address those. And replying takes a lot of time because every word has to be measured. Anyways… that was my experience at least.
Since GoatCounter will be open source I would let the lower tier handle their own hosting. It will also promote more developers to come and join the party.
That’s my experience as well. If the product doesn’t work for them, they’ll demand you fix it, or give them money back. If you don’t refund the money right away, they’ll contest the charge with the bank and you’ll have a $35 loss on a $3 customer.
And even if you do a great job helping them, and they’ll be so satisfied that they’ll buy a whole year of service… that still earns you a low-wage tech support job.
Congratulations on shipping, how exciting for you!
I would remove the Personal / $3 tier and instead offer a 14 day trial period that does not require a credit card. The $30 Business price is in the ballpark. Consumers don’t buy services, businesses don’t want to bother with anything that costs less than the morning coffee. IMO.
And you could easily add a $99/mo Enterprise tier which promises GDPR compliance. It might be exactly the same as your Business tier; all they are buying is the assurance it’s compliant and your willingness to deal with that headache.
The entire GDPR doesn’t even apply right now as no personal data is collected at all. This might change in the future (although it’s a good selling point, IMO), but right now I’m not sure what promising “GDPR compliance” would entail?
I would remove the Personal / $3 tier and instead offer a 14 day trial period that does not require a credit card. The $30 Business price is in the ballpark. Consumers don’t buy services, businesses don’t want to bother with anything that costs less than the morning coffee. IMO.
The reason I added that right now there are surprisingly few options outside of Google analytics that are in an acceptable price range for consumers (basically, none). If there was a service like this for $3/month I probably would have paid it, instead of making this. But maybe that’s just me…
I should probably make it a bit clearer that it’s for personal use, only.
To be fair, the worrying about what “promising GDPR compliance” is what you’re charging the extra $69 for. It may be nothing, but you took the time to worry about it.
I might be wrong but I believe you’re still required to respond to GDPR queries (e.g. “No, we don’t have any data on you”). Might be straightforward but someone has to respond to that email.
How good is it at separating humans from goats^W bots?
Also, would it be against GDPR to collect aggregate data on user-agents etc., basically increase a counter when this or that user agent is seen?
Just the page visit data is useful, but it’s hard to satisfy one’s curiosity regarding the visitors and answer development directions questions like “should I improve the CSS for mobile users?”, “is anyone still using IE?” etc.
RE GDPR, aggregate data is generally quite safe (anything you definitely can’t link it to a person is safe - so e.g. IP address is riskier as there’s a reasonable chance there’s only one user).
Also worth bearing in mind that the goal of GDPR is compliance, not revenue- I’m unaware of any case where a fine has been issued without a “please comply or we’ll fine you” letter being ignored.
Just the page visit data is useful, but it’s hard to satisfy one’s curiosity regarding the visitors and answer development directions questions like “should I improve the CSS for mobile users?”, “is anyone still using IE?” etc.
Yeah, the User-Agent header is already stored for every visit. It’s just not processed and displayed yet. It should be added soon.
I too was interested in Fathom Analytics, but the reason it is not active is because they decided to completely rewrite Fathom for version 2 and make that new version proprietary. When I asked them to reconsider, they said no.
Oh, I didn’t know about that. I find the complaints about lack of contributions rather strange, since they’re just ignoring people’s PRs and issues :-/
I wrote a rather detailed comparison several months ago of all solutions I could find (before I started work on this), but then I accidentally rm’d that and never bothered re-doing it 🤦 From the top of my head:
Many add large amounts of JavaScript. For example I believe Matomo (formerly PiWik) adds 120k or something; I believe Google Analytics is even larger. Not just a lot to download (comparatively to the rest of the site) but also a lot to execute, that can go wrong, etc.
Hard to use UI: my main question I had about my personal website was “who linked to it”, so I can follow HN conversations and whatnot. I found it surprisingly hard to get this information out of Matomo’s UI.
In general, I’d like an UI/frontend that doesn’t suck. I don’t know why so many of these modern web apps are such a pain to use.
I have little need to persistently track people (that is, recognize you as a repeat visitor when you come back next week), and don’t want to display annoying faux-consent banners. Pretty much all existing solutions track people (Fathom is one of the few that doesn’t).
I tried Fathom originally (self-hosted) but there were a few pain points. I would have preferred to contribute to Fathom, but they don’t seem open to contributions, so 🤷
Awesome work, and congratulations on taking the step! Any analytics platform that aims to be open and privacy-respecting should be lauded.
My website is phenomenally small at the moment, and I don’t feel the need to use any analytics right now, but I will keep you in mind if ever I head that route for any of my personal projects!
Reminds me of Fathom which I can see is something you have looked into; I think the community edition of Fathom has been left to just exist while they focus on the paid product which is why the PR you raised has sat for half a year.
I like to see these kinds of “tracker-less” web statistics becoming more in vogue. This is a project I will be keeping a close eye on.
I am happy with anything that takes a little bit away from Google’s stranglehold on web analytics, something I rant about from time to time. Personally, I am against third-party trackers altogether but if they have to exist then yours looks like a pretty good deal.
Interestingly, the domain www.goatcounter.com is blocked by my work’s firewall; I don’t even want to know why.
It’s not really a “tracker”, though. It just counts which pages get visited, similar to the “visitor counters” in the web of old. Hence the name: GoatCounter, and not “GoatAnalytics” or some such, and why I avoided the terms “tracker” and “analytics” as much as possible.
Or, to give a real-world analogy, it’s simply counting how many people walk in front of your house.
That “bestasciitable” site isn’t great, for the sole reason the table isn’t the first thing I see, centre of frame. The table is also really really big. Attention lost, website closed.
Ehm, okay. That’s … rather off-topic here? But there’s a “hide” button and it’ll remember in localStorage.
You might also want to reconsider your choice of words in delivering feedback because quite frankly, you could’ve been a lot more friendly. I spent my spare time making that in the hopes that maybe, perhaps, it will be useful to others. If it’s not for you, then that’s fine. If you think it could be better in some way, great! I’d love to hear about it. But no need to be abrasive about it.
Crudely put, I don’t understand why some people in the open source community feel the need to constantly shit all over other members at the drop of a hat. What goal did you hope to achieve with your comment? How did you expect it to be received? Right now it is serving no goal as it contains practically no actionable criticism, and I can assure you it’s not received well.
I strongly urge you to consider your attitude, and how it is received, because right now the only effect it has is making my day a bit worse and me spending 10 minutes writing this reply.
I wasn’t trying to shit all over your project, sorry. The table is great, I thought I put that in my message, I must’ve deleted it when moving text around. It probably would’ve made my comment sound less harsh, it wasn’t trying to be. Just when people open asciitable.com they get the ascii table, in full view, in the middle of the page (the whole thing usually fits in one page); and it’s an image, so you can open it in a smaller tab or on your desktop if you need to. The design is very “Don’t make me think”, which is what people are usually looking for.
That’s okay; we all make mistakes and it’s hard to communicate by text anyway.
Especially the “Attention lost, website closed” sounds rather abrasive, IMHO. A better comment would have been something along the lines of:
Bit off-topic, but that “bestasciitable” could be better: the table isn’t the first thing I see, centre of frame. The table is also really really big.
Even less words, too!
As a side-note, I tend to write these messages in PMs a lot on Lobsters (e.g. when I have problems with someone’s sites, especially if it’s not directly related to what they’re writing about). It’s a good reminder for myself that I’m writing to an actual person, and sometimes feedback is better received in private anyway. When I first joined Lobsters I was like “why do I need PM?” but I’ve learned to love it.
Just when people open asciitable.com they get the ascii table, in full view, in the middle of the page (the whole thing usually fits in one page); and it’s an image, so you can open it in a smaller tab or on your desktop if you need to. The design is very “Don’t make me think”, which is what people are usually looking for.
Yeah, that’s fair enough. I actually went through a few design cycles where the table was on top and the description below that. I eventually opted for the “hide” link which is remembered accros visits, but maybe I’ll change it in the future.
I don’t know @arp242, but recently when I asked how to do something on the Vi Stack Exchange, he said “hm, you can’t do that” and then landed a patch in Vim to implement the feature I was asking for. So it seems safe to say that he’s a friendly person and an asset to the community.
Hah, I just did that because I wanted it myself, too. But I’m glad that it’s appreciated :-)
Your life will be much easier with 3 customers paying $500, than 500 customers paying $3.
I get that it seems nice to have an accessible price that’s fair for tiny usage. But lower barrier to entry attracts customers who aren’t invested in the problem, won’t have any developer to help, and may not even fully understand what your product does, but at any price, as paying customers they’ll expect to be treated seriously and demand that you make your product work for them.
I’d take personally take on tremendous amounts of pain to have 500 people paying $3 dollars a month, though I’m still pretty skeptical about the amount of pain involved.
Why not just do both and tier based on hits and domains like it already is? Can you not just tell the $3 dollar tier to expect basically no support?
For every SaaS companies I worked for (3), most of the support pain was coming from the lower-tier customers. There are people out there that feel entitled regardless on the level of support that was advertised. It’s quite emotionally draining to read these messages. And because they contain a lot of negative emotion it’s hard not to reply to them in the hope to address those. And replying takes a lot of time because every word has to be measured. Anyways… that was my experience at least.
Since GoatCounter will be open source I would let the lower tier handle their own hosting. It will also promote more developers to come and join the party.
That’s my experience as well. If the product doesn’t work for them, they’ll demand you fix it, or give them money back. If you don’t refund the money right away, they’ll contest the charge with the bank and you’ll have a $35 loss on a $3 customer.
And even if you do a great job helping them, and they’ll be so satisfied that they’ll buy a whole year of service… that still earns you a low-wage tech support job.
I was against this, but you won me over.
I guess people who would pay $3 can host it themselves on a super cheap VPS.
Congratulations on shipping, how exciting for you!
I would remove the Personal / $3 tier and instead offer a 14 day trial period that does not require a credit card. The $30 Business price is in the ballpark. Consumers don’t buy services, businesses don’t want to bother with anything that costs less than the morning coffee. IMO.
And you could easily add a $99/mo Enterprise tier which promises GDPR compliance. It might be exactly the same as your Business tier; all they are buying is the assurance it’s compliant and your willingness to deal with that headache.
The entire GDPR doesn’t even apply right now as no personal data is collected at all. This might change in the future (although it’s a good selling point, IMO), but right now I’m not sure what promising “GDPR compliance” would entail?
The reason I added that right now there are surprisingly few options outside of Google analytics that are in an acceptable price range for consumers (basically, none). If there was a service like this for $3/month I probably would have paid it, instead of making this. But maybe that’s just me…
I should probably make it a bit clearer that it’s for personal use, only.
To be fair, the worrying about what “promising GDPR compliance” is what you’re charging the extra $69 for. It may be nothing, but you took the time to worry about it.
I might be wrong but I believe you’re still required to respond to GDPR queries (e.g. “No, we don’t have any data on you”). Might be straightforward but someone has to respond to that email.
As a counterpoint, I think a simple service like this is great for personal users, but if you are a business you might want more features?
It looks like the value proposition is similar to Fathom and Simple Analytics so you can use their pricing as a benchmark as well.
Good luck. Also, love the simplicity of your site. Old school.
Best of luck with this project! I hope it eventually pays the bills.
I’m suggesting joining the #fediverse aka a masto instance.
How good is it at separating humans from goats^W bots?
Also, would it be against GDPR to collect aggregate data on user-agents etc., basically increase a counter when this or that user agent is seen? Just the page visit data is useful, but it’s hard to satisfy one’s curiosity regarding the visitors and answer development directions questions like “should I improve the CSS for mobile users?”, “is anyone still using IE?” etc.
RE GDPR, aggregate data is generally quite safe (anything you definitely can’t link it to a person is safe - so e.g. IP address is riskier as there’s a reasonable chance there’s only one user).
Also worth bearing in mind that the goal of GDPR is compliance, not revenue- I’m unaware of any case where a fine has been issued without a “please comply or we’ll fine you” letter being ignored.
It checks the User-Agent header, which is an external library. It seems to work fairly well. This is also exactly the same library/method that Fathom uses btw.
Yeah, the User-Agent header is already stored for every visit. It’s just not processed and displayed yet. It should be added soon.
Great stuff, definitely a product I have wanted. I don’t want heavy piwik or spyware from google.
I too was interested in Fathom Analytics, but the reason it is not active is because they decided to completely rewrite Fathom for version 2 and make that new version proprietary. When I asked them to reconsider, they said no.
Oh, I didn’t know about that. I find the complaints about lack of contributions rather strange, since they’re just ignoring people’s PRs and issues :-/
Curious to hear what motivated this. You say there were no good solutions for your requirements, so I’m curious what those requirements were/are?
I wrote a rather detailed comparison several months ago of all solutions I could find (before I started work on this), but then I accidentally
rm’d that and never bothered re-doing it 🤦 From the top of my head:Many add large amounts of JavaScript. For example I believe Matomo (formerly PiWik) adds 120k or something; I believe Google Analytics is even larger. Not just a lot to download (comparatively to the rest of the site) but also a lot to execute, that can go wrong, etc.
Hard to use UI: my main question I had about my personal website was “who linked to it”, so I can follow HN conversations and whatnot. I found it surprisingly hard to get this information out of Matomo’s UI.
In general, I’d like an UI/frontend that doesn’t suck. I don’t know why so many of these modern web apps are such a pain to use.
I have little need to persistently track people (that is, recognize you as a repeat visitor when you come back next week), and don’t want to display annoying faux-consent banners. Pretty much all existing solutions track people (Fathom is one of the few that doesn’t).
I tried Fathom originally (self-hosted) but there were a few pain points. I would have preferred to contribute to Fathom, but they don’t seem open to contributions, so 🤷
Awesome work, and congratulations on taking the step! Any analytics platform that aims to be open and privacy-respecting should be lauded.
My website is phenomenally small at the moment, and I don’t feel the need to use any analytics right now, but I will keep you in mind if ever I head that route for any of my personal projects!
I wish you all the best of luck!
I love this and have already stuck it on my site on a trial basis. Not much gets me to change my Content Security Policy but you have won me over.
Thanks, you’re probably the first :-) Let me know if you have any problems, or if there’s anything missing.
Reminds me of Fathom which I can see is something you have looked into; I think the community edition of Fathom has been left to just exist while they focus on the paid product which is why the PR you raised has sat for half a year.
I like to see these kinds of “tracker-less” web statistics becoming more in vogue. This is a project I will be keeping a close eye on.
I am happy with anything that takes a little bit away from Google’s stranglehold on web analytics, something I rant about from time to time. Personally, I am against third-party trackers altogether but if they have to exist then yours looks like a pretty good deal.
Interestingly, the domain www.goatcounter.com is blocked by my work’s firewall; I don’t even want to know why.
It’s not really a “tracker”, though. It just counts which pages get visited, similar to the “visitor counters” in the web of old. Hence the name: GoatCounter, and not “GoatAnalytics” or some such, and why I avoided the terms “tracker” and “analytics” as much as possible.
Or, to give a real-world analogy, it’s simply counting how many people walk in front of your house.
That “bestasciitable” site isn’t great, for the sole reason the table isn’t the first thing I see, centre of frame. The table is also really really big. Attention lost, website closed.
Ehm, okay. That’s … rather off-topic here? But there’s a “hide” button and it’ll remember in localStorage.
You might also want to reconsider your choice of words in delivering feedback because quite frankly, you could’ve been a lot more friendly. I spent my spare time making that in the hopes that maybe, perhaps, it will be useful to others. If it’s not for you, then that’s fine. If you think it could be better in some way, great! I’d love to hear about it. But no need to be abrasive about it.
Crudely put, I don’t understand why some people in the open source community feel the need to constantly shit all over other members at the drop of a hat. What goal did you hope to achieve with your comment? How did you expect it to be received? Right now it is serving no goal as it contains practically no actionable criticism, and I can assure you it’s not received well.
I strongly urge you to consider your attitude, and how it is received, because right now the only effect it has is making my day a bit worse and me spending 10 minutes writing this reply.
I wasn’t trying to shit all over your project, sorry. The table is great, I thought I put that in my message, I must’ve deleted it when moving text around. It probably would’ve made my comment sound less harsh, it wasn’t trying to be. Just when people open asciitable.com they get the ascii table, in full view, in the middle of the page (the whole thing usually fits in one page); and it’s an image, so you can open it in a smaller tab or on your desktop if you need to. The design is very “Don’t make me think”, which is what people are usually looking for.
That’s okay; we all make mistakes and it’s hard to communicate by text anyway.
Especially the “Attention lost, website closed” sounds rather abrasive, IMHO. A better comment would have been something along the lines of:
Even less words, too!
As a side-note, I tend to write these messages in PMs a lot on Lobsters (e.g. when I have problems with someone’s sites, especially if it’s not directly related to what they’re writing about). It’s a good reminder for myself that I’m writing to an actual person, and sometimes feedback is better received in private anyway. When I first joined Lobsters I was like “why do I need PM?” but I’ve learned to love it.
Yeah, that’s fair enough. I actually went through a few design cycles where the table was on top and the description below that. I eventually opted for the “hide” link which is remembered accros visits, but maybe I’ll change it in the future.
My goal was mainly to answer stuff like: https://www.reddit.com/r/vim/comments/cqvbvs/how_to_map_ctrl_number_looking_for_help/ewzt3ig/ – I started writing the website after a similar question, wanted to link to a good ASCII table, and … couldn’t really find one.