The Changelog: Software Development, Open Source - De-Google-ing your website analytics (Interview)

Episode Date: May 27, 2020

Plausible creators Uku Täht and Marko Saric join the show to talk about their open source, privacy-friendly alternative to Google Analytics. We talk through the backstory of the project, why it's open source, the details behind a few viral blog posts Marko shared to bring in a ton of new interest to the project, why privacy matters in web analytics, how they prioritize building new features, the technical details behind their no cookie light-weight JavaScript approach, and their thoughts on a server-side option.

Surely people look at this and say, oh, I would love switching if it just did X, Y, or Z. How do you guys decide what to build and what not to build? That's actually really easy. I decided that I wanted to make it as open and the development of Plausible is going to be as open and transparent as possible. So the reason it's easy is we have a public roadmap and we have a public forum for feature requests and pretty much people upvote on what they want and I just go in order. You build it.

Yeah. Just like that. The one that's upvoted the most, I go with it. But have you ever seen the Homer Simpson car? You know what a car would look like if Homer Simpson designed it? No. And it's just got like knobs and widgets and like horns sticking out the side. Like if you just give people what they want you end up with a monstrosity

two years from now, don't you? Being with her changelog is provided by Fastly. Learn more at fastly.com. We move fast and fix things here at changelog because of Rollbar. Check them out at rollbar.com and we're hosted on Linode cloud servers.

Head to linode.com slash changelog. This episode is brought to you by DigitalOcean. DigitalOcean's developer cloud makes it simple to launch in the cloud and scale up as you grow. They have an intuitive control panel, predictable pricing, team accounts, worldwide availability with a 99.99 uptime SLA, and 24-7, 365 world-class support to back that up. DigitalOcean makes it easy to deploy, scale, store, secure, and monitor your cloud environments. Head to do.co.changelog to get started with a $100 credit. Again, do.co.changelog. Welcome back, everyone. This is the Changelog, a podcast featuring the hackers, the leaders, and the innovators in the world of software.

I'm Adam Stachowiak, Editor-in-Chief here at ChangeLog. On today's show, we're talking about a website analytics tool called Plausible. It's open source, it's privacy-friendly, and it's not Google Analytics. We're joined by Uku Tatt and Marco Cyrus, the founders of the project. We cover why it's open source, the backstory of the project, the details behind a few viral blog posts Marco shared to bring in a ton of new interest to the project, why privacy matters in web analytics,

how they prioritize building new features, the technical details behind their no-cookie, lightweight JavaScript approach, and their thoughts on a server-side option. So in April, Marco, you wrote a blog post which was quite intriguing. Why you should stop using Google Analytics on your website. We thought we'd start with you just giving us the hard sell. What's the pitch? Why should we?

By the way, changelog.com, Google Analytics. That's right. Hasn't always been the case, but it has been for the last couple of years now. And I'd say we do that in anger, but we do it. And so maybe we'll be an easy sell. But why should folks stop using GA on their website? Okay, well, for me personally, the kind of privacy aspect is a big one. But let's take it from what most people care about.

So one thing that I mentioned there that a lot of people talked about was the aspect of being lightweight. So Google Analytics is not. And the kind of the official way to install Google Analytics uses a second script as well. So you end up having, you know, quite a lot of, you know, things loading, even though it's something that most people don't really look at. So that will be one. The second one will be quite an important one over the last couple of years are all

these different privacy regulations. We know the GDPR in Europe. There's the CCPA in California. And there's one in the UK and so on. And these regulations require webmasters or website owners to insert this different, you know, like a cookie banner to ask for permission to store cookies, the permission to get consent to share data with the third parties. And all this, for me personally, I use the ad blocker and I have to block all of those annoying pop-ups and so on. And I agree with GDPR. It's something necessary for the web to keep it healthier. But I also think that website owners that care about it

can use a different solution that doesn't use cookies, that doesn't collect personal data, which means also that you can get analytics without needing to give all this prompt to your visitors and kind of annoy their experience. So these are the kind of the major ones that I believe most people will get quite a good benefit from. So speaking of the first point, which you started with the bloated script and then

moving on to the GDPR and privacy concerns, do you know how big it is? And I mean, this is something that I haven't necessarily considered, even though I've had my reasons for not wanting to have it. Are we talking about, you know, hundreds of of kilobytes are we talking about megabytes no no no not at all but basically if you if you take any you know the speed tests people because of the seo reasons like google announced like you know speed was one of their factors recently like for their search results and if you take like for example the google page speed test and one of the things they will actually mention is the third party aspect of you having the the analytics script from google analytics like as in that slows down your site so you know

this is why i mentioned it because you remove google analytics from your site or you use something you know something very small or lightweight that kind of error or that kind of warning gets away and you get a better score. So basically, even though it's, I think, let me check how much, how big the script is. Even though the script is not the largest one ever, it's 45.7 KB. So that's not much, but still. It's not nothing. It makes a difference in, you know, in the site, in the site speed for sure, even according to Google themselves. So Adam, you and I have talked a lot about Google Analytics

and different analytics trackers and solutions and what we should do. I have my reasons why I don't particularly like Google's offering. Do you have yours? What are your thoughts on Google Analytics, Adam? It's been, I think, hard to grok the dashboards and the data. It's pretty complex. It's not very clear, like, this is what you really need to know.

It seems built for a one tool for many different kind of customers, and the kind of needs we particularly have aren't exactly the ones that are surfacing easily. And just for that one reason alone, it's pretty difficult. I mean, I only really care about a couple pages in there, and even then they're just difficult to sort of like, I guess, bend to my will, give me the information I actually want to know, analyze my actual information. It just seems like that's what it should do well.

And that's the one thing it does, in my opinion, pretty poorly. Yeah, so basically when I did the research for this post, I actually went through Google Analytics. And I basically counted all the different reports they have. And I counted more than 125 different reports in the left-hand side. And all combined, these 125 reports or so have about 300 different metrics between them. And I'm thinking from my own site i probably use five ten most of those the other ones some of them i've never heard about or never look at them but still i'm running the script there that's that's kind of calculating this and collecting this data all

the time and i'm using it either never or maybe once in in in a while so and i think i'm not like the i'm not the like special case here. I think this is quite common. People install Google Analytics because it's the most popular tool or it's something that they've told that they need to have. And they have this,

this that collects like 300 different metrics from their visitors and they use maybe five of them. And, you know, it's a base in terms of, you know, the KB load on every, on the website visitor. And, you know, it's a base in terms of, you know, the KB load on every website visitor. And, you know, thinking in the sense of, you know, the climate change, even like how to decrease the carbon footprint of a website. There's like a carbon footprint calculator. And if you decrease something like 50 KB of your site, your score gets high that you're something like of the top 10 of the sites that load fast so so

it does make a difference and something people don't use to look at and if you don't use it then you can remove it for me personally it makes sense to do so i'm just like you adam like for me if i say like what's the number one reason why i don't like it i'm just like i log into the thing and i just want to log out again. There's so many features that I don't care about. I don't care about conversion tracking. I don't care about AdSense things and goals. It seems like the information I want is further away than I want it to be,

and the information I don't care about is right there, and then the filtering and stuff is convoluted. And then the other thing is I don't really trust it. I don't trust it in a privacy way. I don't not trust it in a privacy way, although I think that's there as well. I'd say my second reason is I feel like they have so much of our information

and we're just giving them more. Like here, here's all of our website traffic information. That feels naive to me. But I don't actually trust their analytics. I think because of our audience demographic and Change.com's traffic demographic, most of the people that read our website are blocking things.

I'm blocking things. I'm blocking things too. I know it's not right, just intellectually. I know it's incorrect. I don't want to look at that thing that's wrong. As an example, when we have live shows, so this show is not recorded live, but Go Time, every Tuesday, the JS party, every Thursday,

we have a live page, and our live page shows how many people are on that page, and it knows that because it's connected to the actual audio stream. And it could be a couple dozen people listening to GoTime, and I'll go into Google Analytics, and I'll look at the real-time website traffic for that page, and it's got like two. And I just know factually, demonstrably, that's incorrect.

And that bothers me. If you think about it, like Firefox, Brave, they're quite popular browsers, especially in the tech community. And these things block it by default. And then not even that, people that use Chrome, they have extensions on such as the AdBlocker or uBlock, and they block it by default as well. So it's not uncommon to see 40, 50% of a tech site going hidden, as in their visitors block Google Analytics. And so using a different service perhaps also

gives you more accurate data because that different service is less of a target, as in less popular service and it might not be blocked by some of these services such as Firefox and Brave and so on. I'm sure eventually, plausible though, we'll get to a popularity point that you do get that visibility.

And maybe we can talk about how you block scripts or how you block tracking in a way that respects the user. I mean, because if a user comes to our site, doesn't want to be tracked, I'm not going to force my way to track them. That's wrong. So I want an opt-in world, and that's what I really care about. And I suppose when it comes to data and traffic and analytics, you just have to sort of assume that there's a hidden or an untracked spectrum of your actual analytics that's just not ever going to be there.

And you just have to sort of take that into account, even when reporting to yourself or others that care about the performance of your site or lack thereof. Let's talk about plausible analytics a little bit here and set the stage because this post that you wrote was a brilliant piece of content marketing

for plausible analytics, which is a service and a tool and an open source application that you two are working on. So the pitch is uninstall GA. And I'd say, why do we still use Google Analytics? It's because, well, what else are we going to do? It's free, it's easy to set up.

What else is out there? And so this was a nice piece of marketing because it's like, here's this great post all about it. Here's some alternatives. By the way, plausible analytics is something that we're building, which is an alternative to that. So I would love to hear all about it. Here's some alternatives. By the way, plausible analytics is something that we're building, which is an alternative to that. So I would love to hear all about how that works and some

of the stuff Adam's bringing up. Let's pull Uku into the conversation because you've been waiting in the wings here. Uku, when did you start building this and was it because of the reasons that we've been discussing with regard to kind of the status quo of tracking and analytics with Google and other large providers?

For sure, yeah. When I first started writing the Plausible Codebase, I didn't want to use Google Analytics, but I didn't have much of a problem with its UX because I had never really used it. I'm a developer. I don't spend much time in analytics. But I was working on a different project

and the marketing guy asked me to install an analytics tool. And he asked me to install the industry standard Right. services by Google. I was just getting off Chrome and trying to replace Gmail and things like that. So being in that mode of de-Googling my own life, I thought, well, I don't like installing Google Analytics for my project. I had to do it because I also didn't have a good alternative. And there were some alternatives, but I thought some of them were just very simplistic and quite expensive, to be honest. It's hard to justify paying for analytics when there's this standard solution that's free for everyone. But you do realize that you end up paying with data, essentially. So I thought there's room for an interesting alternative there. And I started writing something. I didn't know where it was

going to go, but I had a proof of concept in mind. So I just thought I'll get started on it and I'll run it in parallel with Google Analytics and see how it works. And it took about three months, I think, to get a sort of simple beta going initially. What was involved in that? What were some of the basic, the initial features you focused on?

Like even how did you focus on those initial features? Right, yeah. I just figured, well, what's kind of the basic stats? I didn't have any really experience in analytics before that, so I had to kind of learn about analytics of what's even like a useful stat to have. So I just started using Common Sense. I want to know how many visitors visit my website.

I want to see how many pages they're viewing, what's the top content, what refers they use. And I kind of started building things from scratch. Obviously, I took a look at all of the other analytics tools and tried to distill the most useful sort of UX and what features they surface on their dashboard. But yeah, the first proof of concept was just having a graph with the visitor numbers,

how many visitors there were in a given timeframe and giving the top referrers and pages for that time. There is one thing that was interesting about that early stage. I had tried to build side projects before. I had ideas. I felt like if I could just get something going and try to market it, I'd been lurking on indie hackers for ages and communities like that. But really what changed, I think, with Plausible was that when I started the project, before I wrote a single line of code, I wrote a blog post that said,

here's what I'm doing. Here's how the proof of concept looks like. I don't know how long it's going to take, but if you want to join the beta, send me an email. So I shared that on Twitter and some communities. And I think I had not many people, like 20, 30, but that was enough to give me the motivation to finish the proof of concept. So that's something I'd recommend to everyone who is thinking about writing a side project, getting something going.

Blog about it. Is to get an early audience and commit to something publicly. I think it's really useful. Did the early interest inspire you, or was it more the commitment? Yeah, I felt like both. I felt like the fact that people cared enough to send me an email about it, but also I felt like I have something to show to them in a few months so um i felt like i made a commitment to some real people and that changed it that actually kept me going for three months to get

it out there well you'd be surprised what happens when you feel like somebody's in the fight with you yeah you know i mean like that kind of motivation it's pretty intriguing how being uh responsible to someone somehow yeah changes your motivation like an accountability partner or something yeah exactly accountability only your partner is a bunch of strangers on a forum that's right well but this is a great segue to building the team and the motivations behind that yeah just also remember i've been in situations before where developers don't think of these things before they actually spent let's say three six months building something without thinking of what happens the day i i'm ready to release it and only then they actually start thinking who do i release it to you know who does want this and, who do I release it to? Who does

want this? And what happens next? And doing this earlier not only gives you motivation, like Uku said, it also helps you when you're ready to actually release it out to the public. It helps you have some type of an audience or even really a better idea of do people actually want this or what do actually people want? And this kind of works more ways this aspect of you know building an audience before the product is ready yeah absolutely so you had this you had at least that much intuition and you had something going pretty quickly but there are so many people announcing launches and so many alternatives to things and so many open source projects even. This is a common theme on the show. People we talk to is like,

I built a thing and now how do I get people to use the thing or interested in the thing? And there's this old meme about how you're successful on the internet and it's two steps. You make cool stuff and then you tell people about it.

And that's the two steps to success. And that's true in a sense, but also not true because tons of us are telling people about our things and yet no one's listening because there's so many people telling you about their things. And so you have a nice one-two punch here with Uku and Marco. Marco wrote that post, which brought a lot of attention to Plausible and definitely striking a chord with people

who are already angsty against GA or looking for alternatives. Uku, how did you guys meet? Did you decide, I can't tell this story on my own and I need a helper? How'd that go down? Yeah, it was two reasons. One was, like you said, I'm good at writing code,

but I'm not very good at writing blog posts, communicating stories and ideas to people. I'm good at writing code but I'm not very good at writing blog posts, communicating stories and ideas to people you can get good at anything I think if you put enough time into it but at some point it's also useful to focus on what you're good at and trying to bring someone else in who can complement your skills so that you can both be just experts in your own field

I like this idea of a broken comp theory where you're as a person you're supposed to have broad knowledge of many many things but then deep expertise in a few things so i i realized i'm not going to build deep expertise in marketing and content writing. That was something that after doing it, I was trying to tell the story to get people to use it for about a year. And I had minor success, but nothing to write home about. What were the stuff you were doing? I was trying to write blog posts. I was writing emails to people.

Hey, can you include me in your blog post about Google Analytics alternatives? I was writing updates on indie hackers. I was trying to post stuff on Hacker News and sort of haphazard. I didn't have a strategy. I was just opportunistically trying to get in front of people with Plausible. But at the same time, I felt like it was taking time away from what I really enjoyed doing, which is development. So I really wanted to involve someone who could help with marketing, with getting plausible in

front of people and telling its story. But the other aspect, which is why I reached out to Marco, is that accountability aspect that we talked about earlier. I felt like working on it alone, I started going a little bit crazy sometimes. If you don't have someone to talk to, if you don't have someone to hash your ideas out, someone to tell you when you're wrong, it's so hard to make up my mind. I was going back and forth on a lot of decisions. I didn't commit to a strategy. I didn't like working alone. It's interesting.

There's a lot of talk about being a solo founder and whether that's a good thing or a bad thing. And I felt like it was a bad thing. I didn't enjoy it at all. I think people are different. But for me, yeah, I felt like I was going crazy at times. Yeah. When you have somebody there with you, it's good to get direction from that person.

They're your litmus test in many cases too, but it's also more fun. Absolutely. When things are more fun, you produce your best work, right? That's so true. With Marco, we have a daily call, and it's a great time. It's always a great time. We talk about the product and what's happened in the last day,

and it makes it more interesting. Plus, I'm going to celebrate those victories with the champagne glasses. When you clink a champagne glass against the other champagne glass in your hand, just by yourself, that just seems so lame. You need two people to do that transaction, at least. So how did you find Marco? Marco, we're going to get to you and some of the thoughts you brought immediately to Plausible because this was not very long ago.

You write this, perhaps your first blog post in promotion of the tool, and it has been so far a massive success. We want to get into the numbers. But did you know marco was the guy i think it was january this year when i was just scrolling on my twitter feed and i saw this post about de-googling your life moving from proprietary ad tech company tools over to more open source solutions. And Marco was telling me about, was in that blog post talking about what you can use instead of Chrome and also Google Analytics.

And he had alternatives for all the Google products. And I felt like this is awesome content, right? I just stumbled upon a great blog post. Kindred Spirit, is that the AMP one? Adam, didn't you put that on Change.org News? Sure did, yeah. Was that the AMP one or not? I know that was the December time range marker

when you wrote that one. Well, for whatever reason, I've wrote several Google posts. It's like the Google killer. Google assassin. AMP was one back in December, and that one went well in Hack News and so on. I think the one Uku saw first is I wrote

one about how to Google-ify your website and kind of in general how to use less of this kind of, well, basically two posts, one for your personal life and one for your website. And I'm not sure now which one, was it the personal life or the website one that you saw, but I think the personal life

went better in terms of views, so it might that one but i i wrote two uh and yeah that was basically me in my life over the last two years or so trying to figure out you know what can i do to make the web a healthier place what can i do to kind of support uh you know the smaller players and and and kind of uh what can i do to kind of get, you know, the smaller players and kind of what can I do to kind of get away from the, you know, the ad tech and so on. And one thing was, you know, Google is obviously the target there because, you know, Facebook is easy because if you don't use Facebook, you don't see it much.

But Google is like any website has Google Fonts, has Google Analytics, has, you know, Google has everything, the AMP and so on. And Google is so much more difficult to get away from. And this is maybe why I focus more on Google rather than something like Facebook, which I think is probably even worse company. So just because the fact that Google is so much more difficult, so much more ingrained into pretty much every website that we visit.

And that's where some of these posts have come from that was the motivation really yeah in a lot of cases this reminds me of the base camp story jared back when base camp first came around it was the power of story is what we're kind of talking around it's like this they have a story of this david and goliath which is mentioned in one of the posts on plausible.com and or is it dot io dot io i was just i was like yeah did i say that wrong it's plausible.io but this whole idea of like when base camp first came around its its claim initially which got it its headlines was we're not microsoft word or we're not microsoft this or whatever and it was like this idea that these anti, you know,

you sort of knew what you didn't want to be. So it's, it was fairly not easy, but it was sort of easy to see what you don't want to be. And you can kind of see what you do want to be. And people can grab ahold of that. But Mark, as you're saying,

you're right. If you don't log into Facebook, you kind of don't see it, but Google is everywhere. Yeah. It's almost as if their business, you kind of don't see it, but Google is everywhere. It's almost as if their business strategy is to embed themselves in the structure of the web and then be the middleman for ad buys. Right. And they've always provided more useful tooling.

Oh, the tooling is great. I complain about the interface, but Google has provided to developers and just to techies for years, very valuable. I mean, Google Reader was a hugely valuable tool. Google Search is by far the best search out there, I think. But I don't use it. This is actually something we spoke about this actually,

this point that we were both big fans of Google. Look back, I don't know, three, five years, I was the one using, I don't know, five, seven different Google products every day. And I was the one telling to my parents and my friends, you know, check out Google Inbox. It's the greatest inbox for email. Check out this, check out this. And I don't know, over the last two years or so, my personal opinion has changed about these things. And now my thinking is a bit different. And I'm not the only one.

There's a kind of a growing movement, if you want, of people that want to de-Google their life or de-Facebook their lives and kind of support some different alternatives. Yeah, so I came across Marco's blog post about that and I thought, he's telling the story. He's doing an awesome job at it. I read more blog posts. I thought the content was just amazing. And then I went to the landing page of his personal site and it said, he's a marketer.

I was surprised, honestly. I thought he was a developer by the content that he was writing. I thought he was a developer by the content that he was writing. I thought he was a developer when I first contacted him. Yeah. Good job, Margot. You know, I got a compliment from Uku the other day. He said that my blog or my personal website doesn't look

like the typical WordPress site that he thinks about. When he thinks about WordPress, he thinks about all these banners and all this flashing stuff and lots of stuff right in your face. And my one is more basic, like something you might have a developer do on some smaller CMS or static site. And that was a compliment to me. And I mentioned to him, WordPress by itself is not what you have the image of in all these marketing sites with all these call to actions and so on. It's what people put on top of it

that makes it so. Just a nice little compliment to a non-developer. I want to go back to what you said though, Marco. You said you wanted to play your role in making a more healthy web. And this isn't just simply Google bad, plausible good.

It's not just simply saying that. It's more this notion of power. How do you make a more healthy internet is probably decoupling away from one person or one large entity controlling the data. been freely available and is so accessible to many people and i think the stat is like somewhere above like 80 of most websites out there are using this free tool then that means in aggregate over many many years potentially decades they've got a lot of data and as an ad tech company not saying that they're using it in bad ways or they're bad people or they're a bad company there's varying degrees of that but the point is just like when you put that kind of data in one organization's power or control or whatever potentially bad things could happen you know when you control your own data and you have your own data well then we don't have to worry that some other organization has our data whether it's can be used against us or not it's just a matter of like you don't know when you want privacy until you need privacy. Yeah, I mean, I agree with you.

I think my kind of thinking of this started to change, you know, with all those, you know, the Snowden stuff and Cambridge Analytica and all these things happening a few years ago. And I like, I'm a marketer and, you know, I was using these tools personally. I was using these tools in my profession as well. And, you know, that kind of made me aware of the issues. I was ignorant about these issues before, I guess. And it took Snowden and it took all this media campaign

and all these people to talk about it for me, myself, to realize maybe I should rethink what I'm doing here. Maybe this is not the healthiest in the long run to have Facebook and Google pretty much control everything we do online. So I have my own blog now that's disconnected from everything. You know, Plausible is trying to kind of

get at least some websites to choose the same, you know, to disconnect it one way or another from these big companies. And yeah. Does that make you unique amongst marketers though? Because when I think about who wants more analytics generally or more information, more tracking tracking it seems like marketing folks do because they can then do better at their job I when Adam and I talk and sometimes he'll put on more of his marketers hat and he'll start to say

if we knew x we could do y and I'm always like yeah but x is gross we can't do that right not always but you know we have a balance because he's thinking like a marketer and then when he thinks more like somebody else he's like yeah that's not a good idea are you unique in that way or is there a groundswell of marketers who are more hands off with the tracking I think many marketers for example use ad blockers themselves

so this just tells you that maybe even within the marketing world, this whole thing of collection of data and privacy invasions is not the kind of optimal, it's not something that many people like. But yeah, normally, and this is what we discussed about Google Analytics, the fact that Google Analytics says 300 different metrics is because somebody wants as many metrics as possible of other users. But I think, like mentioned earlier, I think that knowing the core, kind of the most impactful

metrics, the kind of the metrics that make a difference to your company, to your bottom line is better than having access to 300 different ones and that you don't really use. So, yeah, I mean, there's an argument that the more you know, the better it is. But I think that doesn't mean that you need to collect as much as you can. You better kind of limit it down and kind of actually understand

what do you need to know from your website, from your customers and so on. How can you use it? And what's the best way to get that without going overboard and collecting everything and making all these behavior profiles and all the other kind of tracking across the web as people browse different websites and all that stuff. Kind of a balance like you mentioned.

Marco, I think in my view, you're quite unique being a marketer who thinks that the web is a little bit broken and wants to fix it. I remember telling to my friends before I stumbled on your website, I want to find a marketer who cares about privacy and open source. And I couldn't find one. I looked, I was trying to find someone who could help me with this, but someone who wasn't just a

marketer, as I thought, no offense, but to my, it is also my understanding that it's kind of marketers who want more and more data usually. And from the marketing departments, it's also my understanding that it's kind of marketers who want more and more data usually. And from the marketing departments, it's where some of this data collection issues are coming from and the privacy issues. So I thought it was very unique to find someone who can do marketing but also is in that same sort of mind space in terms of understanding what Google is about

and trying to fix that. Yeah, maybe I'm just a bad marketer that I don't use all this, that I don't collect all the data. Well, you don't need it. Yeah, I think in general, it's something that you can argue that there's a need for it, but it's not something that's necessary. I think the more important thing is being able to have a product that people want and then being able to communicate about the benefits of that product to people that are interested in it,

to people that that product solves issues for. And that thing you kind of really do as well just by knowing a lot of data. You actually need to speak to real people. You need to kind of get into their shoes and kind of understand them better. And in order to communicate with them or in order to create a product that actually solves the issues they have. If you need secure and PCI compliant payments for your website or an app with iOS and Android, where would you get started?

Well, our friends at Square invite you to develop on the platform that sellers trust. Head to squareup.com slash go slash changelog to learn more and create your Square developer account. When you build with Square, you get to outsource all the payments complexity to them. They take care of maintaining PCI compliance, detecting fraud, and managing disputes on your behalf. You can customize your checkout experience to match your style guide or save time by using the pre-designed payment flow. Thank you. account today at squareup.com slash go slash changelog. Again, squareup.com slash go slash changelog. So guys, whenever you build a product or a service as the anti-X, as the David to some Goliath, you have to be very intentional with features. You have to set yourself apart.

Sometimes you have to pick which features you're not going to develop, especially in the case of a privacy thing. You say, we're not going to do that because that's privacy. But whenever people are looking at it and they're thinking, okay, I would love to go away from Goliath and Google Analytics, but I've been using it for so long, it's free. I like this about it. I do complain about it, but there are things I like. I like the event tracking features.

We do use it. We like to know on our website when people click the play button to play our episodes. And then we also like to know how far they get through because we think that's useful for us to know. Not any particular person, hey, you listened to our episode, but anonymized how many people listened to this episode on the website, how far did they make it through.

We use the event tracking. That's just one example of one feature that we appreciate. Surely people look at this and say, oh, I would love switching if it just did X, Y, or Z. How do you guys decide what to build and what not to build that's actually really easy i decided that i wanted to make it as open and the development of plausible is going to be as open and transparent as possible so the reason it's easy is we we have a public roadmap and we have a public uh forum to for

feature requests and pretty much people upvote on what they want and I just go in order. You build it. Yeah. Just like that. The one that's upvoted the most, I go with it. But have you ever seen the Homer Simpson car?

You know what a car would look like if Homer Simpson designed it? No. And it's just got like knobs and widgets and like horns sticking out the side. Like if you just give people what they want, you end up with a monstrosity two years from now, don't you?

That's a good point. I mean, you have to, I guess, weigh it against some of the values or the vision that you have for the product. So there's a feature that I'm planning to add that no one has requested, for example, and that's going to start in somewhere, which is being able to just query ad hoc anything, basically, by clicking on whatever you want in the UI.

There's no request for that, and I'm going to do it at some point. And I've said no to things, for sure, but more or less, I'd say 80%, the prioritization comes straight from users. I guess maybe, Jared, what you might be asking is, what's the backbone of your roadmap? Do you have things you weigh against?

So sure, even if it already is in there, I think it's great to have an open forum to invite people to. How do you gauge the judge on, is it simply just votes, I suppose at this point to some degree, but what else do you weigh it against? I think I weigh it against my own vision for the product. I guess what I'm seeing in about a year's time and then thinking are these the things that will lead us to that point is one of those things.

I just want to build a good product. We didn't have a thing that would differentiate us that would give people a really good reason to switch for the longest time and I really wanted to have one. So being able to do analytics without cookies is now a big one and uh it wasn't the highest requested feature but uh that was one that i brought in to the top basically because i really wanted to have that reason why someone would look at you know what do i get by switching to plausible i'm going to switch on from a free tool to an analytics tool that basically gives me the same stats, and I'm going to pay for it.

So it wasn't a great value proposition for the longest time, but now that we do analytics without cookies, which I don't think Google Analytics is able to do, I think that gives people a good differentiator. That means you don't have to have that stupid banner, right? Exactly. So that is great. That's a big deal.

What are the trade-offs? What do you lose going cookie-free? You lose some accuracy, right? In my testing, I was running both approaches. So currently the unique user tracking is basically based on the amount of IP addresses that access your website, anonymized. The numbers are very similar to what the unique counts

that I was getting using a cookie. I was actually surprised. I thought that I was going to see numbers that are quite a bit off, but they're actually very similar. Unique user tracking with a cookie is not accurate completely either. Everyone has three devices now. You're basically tracking devices, not visitors.

In some cases, IP address might actually be more accurate when you have multiple devices on one IP address. There are interesting trade-offs between cookies and IP addresses. In some cases, maybe I got three people at my house there are interesting trade-offs between cookies and IP addresses. Yeah, I was going to say, because in some cases, maybe I got three people at my house and we're all natted

and we all show one public IP, but that actually is three visitors. Or maybe I'm just using three devices and I'm still one person. So yeah, it's fuzzy on either side, isn't it? Or maybe you're on a mobile device and driving and you get a new IP address every few minutes.

It's fuzzy either way. I was running both side by side for a while. It's plausible. Yeah, it's good. But remember, if you take the whole concept of the product, then it makes sense to make this slight trade-off here and there. Oh, for sure. We are catering to people that care about these things,

just as we do. And they also want to remove Google Analytics from their site. They also want less tracking, but they still want to see something. And this is what allows us to kind of make these decisions because we know we're

pretty much on the same pages as a lot of our audience. And this also helps with the roadmap because if you go to the roadmap right now, which is on our site, a lot of our audience. And this also helps with the roadmap because if you go to the roadmap right now, which is on our site, a lot of those requests, they're pretty much fit with what we're thinking about. Now it's just about prioritizing them

and getting time to do them and doing them right so it fits with the product. But it's not like we have completely different people asking for something that really does not fit. It's a niche product that the people that come to it are actually interested in this thing and then they kind of think in similar ways which which really helps in and kind of build a product that that's kind of unified and that that makes sense for for this audience talking about prioritization there's a there's

a really interesting new input which is marco you know Now that we have a marketer on board, we're dogfooding our own products more and more. Previously, we didn't have much traffic, and it was me looking at the stats. Just 50 people today, cool. Wishing there were more of them. Now we have actual traffic. Yeah, now that we have real traffic and we have more of a

marketing approach to and focusing more on selling the product, I think that will start, the dogfooding aspect will really start feeding into the product as well. And like, I know the tools from my past experience working for different companies. I've spent hours doing tracking heat maps

and looking at what people click on on the conversion, on the funnel to sign up and so on. So I know what is out there and what is useful. So now it's like me feeding back from that side and then Uku feeding back from the, you know, this is not possible because of, you know, you cannot do this without cookies or this is not possible because we would have to track, you know, identify people, you know. So this kind of the bounce back and forth, it works in that sense that I can come from more marketing side

and kind of I can push some of the more marketing aspects reasonably. And then Uku can tell me from the tech side and kind of more from the privacy angle, like we can do this, we can do it this way. What do you think of that? And so on. Or this we cannot do because of these decisions we have made in order to make the product privacy friendly.

So you're in an interesting spot in a niche where you are your customer base or your perfect user is privacy oriented, but not so much so that they have to run all their own things because you're still hosting the data. And so it is open source. It's an Elixir app. I assume there's lots of moving.

We also host an Elixir app. There's things to do to host that yourself. Maybe you can make that click a button deployable at some point, whether that's in your interest or not, I'm not sure. But you think the real privacy-oriented people, they don't want to host their data with you. They want to run their own plausible.io. Have you run up against that? Those are the really, I would love this, but I'm not going to host it on your guys's stuff. Yeah, and I want to offer that. It makes sense. The main reason why I haven't done it so far

is because the product is still fairly early stage. There's a lot of not only moving parts in the infrastructure requirements, but a lot of moving parts in terms of just upgrades to the database schema, for example. I'm now working on adding a second database to the infrastructure. It's worked to just upgrade my own servers,

but having to upgrade a hundred other ones and having documentation and the click button convenience for that, it's a bit too much for me right now. Yeah, it takes away from other things you'd be working on. Yeah, but I have nothing against people, it's released under MIT, it means you can do whatever you want with it. You can start your own company running the same code if you want. But recently there's been more and more interest in self-hosting plausible, and there's a GitHub issue with now three people involved,

including, well, excluding me, so four people involved now. And I'm trying to offer my own help as much as I can to make it self-hostable, have a Docker image ready to go on Docker Hub so you can just pull and go. Yeah, I think that will net you a lot of goodwill over time when you get to it. Yeah, I think the way we make money

should be hosting the open source solution and committing to it. But it shouldn't be from guarding the secrets or having some kind of walled garden that you can't access. Yeah. You mentioned it was MIT. What was the thought behind making it MIT? Why is the transparency important?

Obviously Google Analytics is not. One of the things I thought about was, what would stop a company from becoming Google? I think one of those things would be, if you don't trust the company anymore, you can just take the code and build another one using the same product, for example. Just having the code in the commons rather than in trademarks,

I think is valuable to the community because it stops a company from going haywire, I think, in terms of what they do with the data. You can't lose customers' trust because you have the threat of forking. And the threat of forking is what keeps a company in check, I think, when their code is open source.

I'd like to take this opportunity to announce our brand new service. It's called More Plausible Analytics. It'll be up here real soon. Just kidding. Let's see if we can do content marketing as good as we can. There we go. That's the secret sauce. Well, that's the question.

Are your blog posts Creative Commons? Share like 4.0? We can just rip off everything you're doing. It's pretty easy. we'll just follow you you know one interesting aspect was that I never experienced before

the other day I got an email which was just from GitHub automated email somebody telling me he was reading my post before I published it and I didn't realize

that we're now using a CMS which goes through GitHub so every time I do my draft and I save something it's there and people can see it and read it. And I was like, wow, because normally I write the draft

and it's just me looking at it. Maybe I save it and it's finished. But now I'm actually drafting something and it's going every time I save it because I want to preview it. I want to check how it looks like, whatever. It goes to GitHub and people can actually look at it there. So I just thought like,

oh, maybe I should watch what I'm talking about Google here. Maybe I should not write something that I'm eventually going to delete because it's going to stay there on GitHub, which is just a funny experience. And even if you remove it, it's still in the Git history. This is why this is just a funny aspect of it

that I did not consider, but it just makes the whole thing more open. So from being open source to having this open roadmap, to even our silly little blog posts being on GitHub while they're being written. Before they're published. That's one webhook away from easy plagiarism. Just every time Marco pushes to GitHub,

just webhook that sucker and publish it. Don't start me about plagiarism. Are you getting ripped off? Yeah, maybe I should not talk about it because I'm trying to be quiet. But it's been going on since the successful story of ours. Yeah, it's been going on, unfortunately.

From even bigger companies, if you want. Which is funny. Things happen when you ruffle feathers. you know, bigger companies if you want, which is funny. Well, it's, things happen when you ruffle feathers. You know, when your story makes sense to a large majority who have been paying attention. And when you're in the limelight and people are like,

I want some of that limelight. Yeah. Or they get threatened. I think this is the way I look at it because for me, you know, I'm looking at Google Analytics as a competitor, if you want. We're trying to find some of those hundreds of millions of sites that use Google Analytics to maybe consider plausible.

And being threatened by someone of similar size, I'm not. So, yeah, it's just the way some people react to it differently. And then they see somebody having a little bit of success. And because we're so open, people can actually see that we're having success, which is kind of, you know, it backfires in that sense that actually some competitors can actually, you know, try and steal our limelight from us. Remember, imitation is the sincerest form of flattery, right?

It means we're doing good. Absolutely. Great. So before we get too far afield from the product, I do want to ask you this, Uku. So you have lightweight JavaScript. You've gone no cookie at this point. I've thought a lot about just removing JavaScript altogether

would feel even better. And I've thought if we had some sort of plausible, if we had just good enough, even server-side analytics that just did log analysis and was at least smart enough, whether it was like Blacklist or machine learning algorithms on the user agents, whatever it was, to just get rid of the real body stuff, the scrapers and stuff,

I would actually be just fine with that. I wonder if you've considered server-side logging or if it was JavaScript all the way, because your JavaScript's not doing very much at this point. Yeah, this is one of those things that I don't think it's been requested on our feedback page, or maybe it has, or it has one or two upvotes, but this is one that I'm raising the priority up by.

Yeah, I want to provide that, for sure. Really? Yeah. Sign me up. I think there should be a module for Nginx or Apache, whatever you're using. If you're running Elixir, I can just write a plug and send that over to Plausible from the back end.

And you will miss some of the things. I can't get the client width, for example, of the browser. Right. I think it's nice to do device detection based on the actual width of the browser rather than the user agent. Viewport size is important for responsive design. When you're writing CSS, you care more about how it's displayed.

There are things that I'm planning to add to the product that might require JavaScript, but the general idea that the basic things like getting visitor accounts and referrers and user agents doesn't need JavaScript. And it's something that I actually, yeah, I

would recommend everyone do that once it's available with Possible. It's how we used to do it back in the 90s. I mean, well, 90s is, maybe that is how they used to do it. I wasn't doing it in the 90s. I was doing it in the early 2000s. that is how they used to do it. I wasn't doing it in the 90s. I was doing it in the early 2000s. And I'd run this Perl script. I can't remember the name of it anymore.

And it would just analyze my Apache logs and it would spit out reports. And it was pretty good, but then it just became such wrong data because there's all these bots. And the nice thing about JavaScript is at least you know you have a device that runs JavaScript, right?

Bot detection is hard. And it's one where google has so much data and they're not releasing that so it's tricky i'll have to see what i can do currently the bot detection is very basic based on user agents and the fact that the client has to run javascript for it to be counted that that counts out a lot of you know totally yeah there are challenges there and there's a challenge of convenience. The front end is standardized, but the back end isn't. So there will be many, many different modules

and libraries for people to hook into when you do offer back-end tracking. I mean, it can start with a simple HTTP API for you to just shoot requests from your back end. It's a hard problem. So it's not one that I have spent too much time on right now, but it makes total sense to me. That should be available.

It's the ultimate differentiator. It goes the complete opposite direction. And I haven't seen, because I've been looking for it, and I was like, is this something that we should build open source or something? There are some, there's obviously open source log analysis tools and a few things our community members in Slack have pointed us to nothing's quite been what I thought it should be and so it would be a differentiator

like no JavaScript you know that sounds nice, that feels nice I've actually said publicly to a few people if you feel like you want to deal with the log analysis tools and manage that, I think that's better than including our JavaScript in terms of privacy it's great

how does that impact real time metrics? as fast as your server knows it's there, it depends. Where are you introducing that analysis? Is your analysis streamed? Do you have to change the way you log then? Would you have to change how you log or how you're reading logs?

Anything with your logging whatsoever would much change or have to change to do it that route. Yeah, implementation details in terms of, are you deciding to have a standardized log format that's plausible or whatever this tool reads? Or are you streaming data directly into an API? Are you batching that? Is it real-time?

There's lots of things to decide. You've probably thought about this more than I have. Honestly, I haven't. Because it's one of those things where when I decide this is what we're doing, then I'll do a bunch of research on all the different options out there. But log analysis is its own little world.

I think, I don't have much expertise, but what I do know is that the biggest challenge is probably bot detection. I would tend to agree with that. What's interesting though is that the biggest challenge is probably bot detection. I would tend to agree with that. What's interesting, though, is that you had said once this is in place, you would recommend everyone go that route versus JavaScript client front end. Sure, if you don't need the extra stats

that come from JavaScript, of course. Would you still provide both then at that point to give people the option yeah yeah there's stuff you can do with JavaScript that you wouldn't be able to do from the back end like event tracking in browser event tracking it's much more convenient to track events

yeah you can do more accurate time on page if you want scroll depth you can do interesting stats like that which require JavaScript more accurate time on page. If you want scroll depth, you can do interesting stats like that which require JavaScript. But I guess most people wouldn't be interested in that, would they?

I guess it's up to them. If you want to know their mother's maiden name, you need JavaScript for that. Mother's maiden, yeah. Social security number, you need JavaScript for that. Well, so plausible is open source. Is it open source in the fact that you're taking contributions

and looking for contributors? And what I mean by that is if someone out there has specialized in or has a lot of information they can contribute towards, say, the server-side methodology, and they have a lot of experience around logging and et cetera, is this a call to arms, so to speak, to say, if you've got expertise in that area, we're looking for people to work with us

or contribute or share ideas. To what degree can someone get involved in Plausible? Yeah, I think once we figure out the self-hosting aspect in the coming weeks, it will be very easy to take the code and run it yourself. And that's where things can really get going in terms of contributions. Currently, I've merged one or two pull requests from other people, but not a lot

of people are running it on their laptops. Not a lot of people are running it on their servers. I don't even know if anyone has a self-hosted version running right now, because it would be tricky to get it going, to say the least. So there's work going on to make it really easy for people to run it themselves right now. Once we're there, that's when the contributions can start really happening. This can become more of a community project. So far, it's been pretty much all me.

But I'd love for people to get involved. About the contribution aspect on github this is again first time that i'm dealing with this kind of github and open source aspects of running a startup when this blog post you know went viral a few weeks ago i think the day after there was a a new thread on github like quite long beautiful thread where a guy said you know you're using unnecessary code. I don't even know what he wrote. Uku is much more familiar with that. But basically, he came, a guy that really knows his stuff,

he came in there and he wrote basically pieces of code how we can improve our tiny 1.4 KB script to go even lower to perhaps get it to 1 KB or even under 1 KB. And I thought that was amazing that somebody would take, I don't know, maybe one hour, two hours to write quite a long GitHub thread and to help us without us asking for it, without us knowing him,

just because he read that post perhaps on Hacker News and he thought, okay, these guys don't know what they're doing. Let's get it from 1.4kb to 1.1 or 1.0. And I just thought it was amazing that this kind of contribution can happen. Yeah. I know from experience that running an open source project

is a lot of work as well. So, so far I've focused on the product aspect and the business aspect of things. Because one of the trickiest subjects is how do you get open source funded? How do you get people working on open source in a sustainable way? So that's kind of what we're trying to figure out here, but it's definitely a direction I want to go in is more community involvement and more of a community project where we charge

for hosting it. How much time does your team spend building and maintaining internal tooling? I'm talking about those behind the scenes apps, the ones no one else sees the S3 uploader you built last year for the marketing team, that quick Firebase admin panel that lets you monitor key KPIs, maybe even the tool, your data science team hacked together so they could provide custom ad spend analytics. Now, these are tools you need, so you build them, and that makes sense.

But the question is, could you have built them in less time, with less effort, and less overhead and maintenance required? And the answer to that question is, yes. That's where Retool comes in. Rohan Chopra, Engineering Director at DoorDash, has this to say about Retool. Quote, The tools we've been able to quickly build with Retool have allowed us to empower and scale our local operators, all while reducing the dependency on engineering. End quote. Now, the internal tooling process at DoorDash was bogged down with manual data entry, missed handoffs, and long turnaround times. And after integrating Retool, DoorDash was able to cut the engineering time required to build tools by a factor of 10x

and eliminate the error-prone manual processes that plague their workflows. They were able to empower backend engineers who wouldn't otherwise be able to build frontends from scratch. And these engineers were able to build fully functional apps in Retool in hours, not days or weeks. Your next step is to try it free

at retool.com slash changelog. free at retool.com slash changelog. Again, retool.com slash changelog. Well, you're well on your way. Let's get back to the blog post because, as I mentioned, it's never as much fun to celebrate by yourself. We're here to celebrate with you guys. Marco had a hit post. We talked about it, why you should stop using Google Analytics.

Can you guys share some of the impact of this? Because everybody loves that moment where the thing they've been toiling over gets some attention and gets some users and you've got a little bit of open source starting to trickle in and there's interest how big did this blog post go and what did that result in terms of users trials people giving you money yeah i mean so so basically the post was published in like 8th of april i believe it was. And our stats for April were just over 60,000 visitors, almost 63,000, which is 2,500% increase compared to March. That's one number to kind of look at.

We also had, because we offer a free trial, 30-day free trial people can sign up for to test us out before actually deciding if you're worth it. 272 people signed up in April, which is six times more than signed up in March. And actually, all the signups in April, they were more than previous nine months combined. So just that aspect that we got a huge boost in visitors and huge boost in new signups for trial was great. And now we're kind of reaching those 30 days or so from that first day, which means that some of these trials are expiring. And now we're basically went over a hundred customers. So we have seen a 70% increase

in total in paying customers from the day of the post until today. And we've seen MRR as well increase by 80% from the day of the post to today. And basically, these are very concrete numbers, which you can go and say, doing this type of marketing, it can work as well. You don't have to go and go to Facebook and pay Facebook. You can do some content marketing. You can reach out to your audience organically and you can actually still achieve results. And this kind of is at least one that proves it.

Maybe we were just lucky. I don't know, but it can be done. Well, you find out when you write your second big post and you see if you hit a home run again, right? Yeah, like I wrote in the follow-up to that one, I was like, the only way from here is down because i mean i i don't want to like promise uh people that just start like was talking before you know he was writing posts before but that doesn't mean you know you can

create the greatest post you think it's an amazing post everyone should read it but you know only your mom comes to your site and reads it maybe. So this kind of thing of publishing content works in general in the long run if you're kind of consistent with it and if you produce value. But you also need some luck here and there in order to get a spike.

So this is why I mentioned the only way to go is down. As in, don't expect us to have a spike every time we post something or even every month you know it's just it's something uh you know that rarely happens but the important thing is that you're you're building up the content you're building up the value you're creating you're answering people's questions you're kind of building up the authority and the name of the brand and this kind of kind of slowly rises from from day to day from week to week and then if you zoom out and look at the long-term picture,

oh, there's actually a big achievement. We've actually gone gradually up. And there's now, in the last six months, there's a huge increase, rather than one spike and then nothing again. Yeah. It really just shows you how one blog post can really change the traction of your startup.

Yeah, that sounds familiar. Touche. That's plagiarism, man. You got me, you sounds familiar. Touche. That's plagiarism, man. You got me, you got me. Yes, I stole the title of your blog post written on April 17th. However, it's very true, though. I mean, and this is something that we share a lot, too, because we have a news feed. We populate that news feed with lots of native content, but we also have it sponsored.

And one of the biggest things we tell people who want to sponsor our newsfeed and newsletter is write awesome content. Don't point people to landing pages that are terrible, that have pop-ups everywhere. Like literally think about the kind of content that is high value and then potentially highly convertible. And not just convertible in the fact

that it gets you new users, but starts to chip away at that idea of trust in a good way, meaning that it starts to establish roads of trust towards your brand, towards your product, towards your service, towards whatever you're trying to do. Provide valuable content, high value content that is convertible in that trust factor that says, I should now trust your brand or believe in who you are or get involved in your mission or care about what you're doing. And when we see people do that with the kind of content that we want to promote and the fact that it's promoted or sponsored news content, we see great results.

So storytelling is huge, but high-value content that developers actually want to listen to or read is paramount. Exactly. And let me ask you a question. How many companies take you up on that? A lot, yeah. And a lot of what we do too is even evolve that education too because some will come to us with not really understanding how to leverage us. This isn't a pitch to people who out there

want to use our newsfeed or newsletter to promote their stuff, but it might be in some way. It's an education process. We help them understand the power of good content marketing and not just like content marketing for the negative sides of it, just doing it just to do it. But really, truly sharing your story, which is what you guys are doing here today, is like when you have a story to tell, people care. And if you can be really good at telling that story, which you are good at, Marco, is there's proven dividends. So yeah,

a lot of people do. And some people come to us with not a really good clear direction towards it. And we educate them. We help guide them. We're very much a guide in that process, like how to leverage promoting their content in our newsfeed. So in many cases, we'll help them understand how to best use us. They don't often come to us and say, here's all these ideas, you know, go and run with them.

We more or less kind of help them and guide them. But it all starts with a great content funnel and a great team behind that to do all that work. It's like,

if a company came to us with no high value content, we would say, go create some high value content and come back. Right. Because we can't help you until then.

But it's easy for us to open with that because we're developers, and so we just ask ourselves what's interesting to us. It's not that rocket science. Sometimes you have to pull yourself out of the equation and say, okay, I'm not into that particular thing. That being said, is this good or bad? Is this interesting?

But what do we like? And this is what we spoke about earlier about marketing it's a different mindset you need to do content marketing rather than paid advertising so that's that's kind of another thing that maybe paid works better for some because it might be easier to get started before we get ahead with but content marketing you need to actually put yourself in the shoes of an audience you need to actually create something that they marketing, you need to actually put yourself in the shoes of an audience. You need to actually create something that they might like. You need to actually speak to people and so on. It's not as easy

to get started with as just put a credit card on your Facebook account and start running advertising. Yeah. Well, you might get clicks, but you might not get conversions. You might get people come to your site, but they'll come there and go nowhere.

Exactly. And this is what all marketers want. I mean, they don't want clicks. They want to get some real actual results, kind of like similar results to what we were speaking about, to actually signups and actually paying customers in the bottom line of the company. That's the one. You always hear about the vanity metrics and social media and all that but in the end what a marketer reports on to their leaders in the company at the end of the month or at the start of the month are the actual you know the core numbers of the profits the new signups and so on it's it's a long past those days where you could report on how many new facebook likes

you've had on your page that nobody cares about those numbers anymore. So Marco, as the only marketer I know who runs Linux on his laptop, I think you're well positioned to answer this question with regard to what developers are interested in. When we first spoke, you said the why to take GA off

your website was like the post you already had written in your head when Uku first approached you or met you. That one was obviously successful. Now you've had your in your head when uku first approached you or met you that one was always successful now you've had your follow-up like what traction looks like that's a interesting you know sophomore album what's some other stuff you have in the hopper like moving beyond the obvious google analytics bad plausible good what else can you write about what else can you talk about that helps tell your guys' story? I'm trying to think of things that are somewhat reproducible. I know we can't all hit a home run just by doing exactly what you did, unless we're just going to plagiarize.

But are there any sort of recipes, or how do you think about these things? I think it's important to understand who are you speaking to. And if you can do that, if you can see it from their perspective. So one of the things that many companies go wrong with is they think of themselves first. They're like, I'm going to sell my company. I'm going to sell my features and my product. They're not thinking about who they're trying to sell their product to, or they're not thinking about what those people actually want. And my experience is if you actually think of them first and actually provide value to them first

and only then indirectly you can talk about your product and what your product can do even better than whatever solution you kind of described early on, then you can actually see better results. But majority of sites go like, me, me, me, me, rather than, you know, you, you, you, you. And there's no like, you know, law of first sight online in online marketing.

You actually need to provide some real value to people for them to actually, you know, understand you, to take you seriously, to actually spend a few minutes of their time to actually explore your post

and your product because people are busy. People are impatient online. There's so many other distractions that if you're not actually providing value and speaking to them directly in words that they want to hear,

you will struggle. And there's so much other content out there that it can get difficult unless you really understand the people you're targeting and you speak to them and their issues and the problems they want to solve. So Uku shared his roadmap. Can you share your drafts folder?

What do you have? I know you share it on GitHub, but you got some titles. What's the kind of stuff you're working on, you're thinking about writing, things that are going to go out? You know, so because of this success here we've had, I've been doing more interviews

and speaking to other people because so many people are curious now. They've heard about Plausible. They want either features on their sites. They want to ask questions and things like that. So there's many questions on social media, email. So I've been more, if you look at our blog, it hasn't been that active since then.

We published, I think, twice or three times. But basically, one good example of what I've published since, obviously, I published one with the results and how one blog post can make a difference. But the second one I published is actually six or seven different people have asked me pretty much similar questions. Like, this sounds great to me. I also don't like Google and blah, blah, blah. But would I lose my search engine rankings if I remove Google Analytics?

Or does Google actually use Google Analytics and the data they get from there to help me rank better on search and get more visitors? And, you know, that was like something I didn't consider. That was not on my roadmap to write about. But I was like, I see, you know, I understand the opportunity.

I understand what people are asking for, what they're curious about, what questions they have. And then, you know, I did my research and, you know, I wrote whatever I could find directly from Google on they have. And then I did my research and I wrote whatever I could find directly from Google on this topic and I published that. So again, you got to put yourself in the shoes of people that you're targeting and just understand what kind of questions they have, what are they thinking about in these things. And then you base your roadmap

according to that. So it's not like I came to... you know, I might have said, yeah, I had this post kind of as a title already because I write something similar on my own blog. And then I saved it for plausible when Uku contacted me. But in general, it's not like I came with a roadmap of next six months content. And I said, we're going to do this, this, this, this, this, this. I might have an idea or two here. But then it's about listening, listening monitoring being part of the conversation you know i spend i don't know how many hours looking through hundreds of hacker news comments and lobsters comments and and people on twitter and and mastodon and so on i spend hours on like reading taking notes and

and kind of trying to figure out what the the situation is what's the feel and what what are people you know talking about and this then leads us not only to create more kind of interesting content in the future, but also to improve our product. Because people, one of the big comments on Hacker News was like, you guys are missing the pricing. And we had the pricing on the homepage, but it was not on the top navigation bar. And people don't scroll down so far that they saw it,

or at least a certain percentage of them. And within one hour of that comment being on Hacker News, we already fixed it on the site. So basically, it's really important to be connected to the people, to the community you're talking to, and kind of react. Don't have like, this is the set rules. We're going to do this the way we want. Be part of the conversation and kind of be flexible enough to understand where the conversation is going and kind of work according to that. There's one word that describes somebody who does something like that. Well, a phrase technically, but one word that describes it.

They care. You show up and you participate if you care. You solve problems because you care you can write these blog posts because you have empathy because you because you care it's important and that this is why when i said earlier people that come to us they care as well you know they're they're writing this long contributions on github they're sending us all these comments and and additions to our roadmap So it's important to be able to communicate. And if you communicate well, you can get people to understand it

and come to you as well, come on the same page. It brings it all to a higher level. So where will you all be at in a year from now? Considering the success you're at right now, the adoption rate you're working on, your open source roadmap, this idea that you also agree with on server side. I didn't give them the

idea. Yeah, I know you didn't. But the point is if things keep going the way they're going and more people look at Plausible as a plausible option against GA or others. I'm going to use that one.

Where will you be? What can we expect? I think we're going to still be on your podcast talking about the only spike we ever had in traffic and the only time Hacker News talked about us. Back in the day, that was a nice day. We should have a one-year retrospective and we can just reminisce. I'll bring the champagne.

From my side, I think my goal for the year would be to be sustainable as a business. I want to make sure that we can do this full-time without worrying about the runway. I love the product so much, and the project. It's really important to get to the point where we can just work on it without worrying about it. We're bootstrapping it, so it's difficult financially, but we don't want to have anything to do with VCs or that kind of stuff. Are you all in at this point? Are you still working?

No, I'm not. So I'm all in, I guess. I'm totally comfortable sharing the MRR, so we're at $800 a month. And we're very open about that. I write a monthly journal where I share that as a sort of a journey sharing the MRR, so we're at $800 a month. And we're very open about that. I write a monthly journal where I share that as a sort of a journey of how we're getting,

where we are in the journey to get to a sustainable business. That's the most important thing, I think. Because if we don't reach sustainability, then there just won't be anyone working on it, I guess. Well, in March, or sorry, I guess April's post about March, the recaps. One, I love these recaps you do. I think they're really awesome and very transparent in terms of your growth. And you already do mention MRR publicly. So it seems March was 4.15,

so you've doubled based on what Marco said earlier and what you're concurring against here in your blog post. I mean, that's a good thing, right? To get to sustainability means obviously you have to grow. And growing that number. What's that number look like at sustainable?

What's the threshold? We haven't talked about it with Marco. We can hash it out right here. Let's hash it out. I don't know. No, we have not. Let's hash it out. There you go. I don't know. I live a simple life. I can live on a minimum wage here in Estonia.

Which is what? Which costs the company about 700, 800 euros a month. I'm fine with that. It's a cheap country. There for one of us. Yeah, you're halfway there, right? Almost. We've got to double up.

The servers and the databases are quite expensive. Yeah, you're halfway there, right? Almost. We've got to double up. The servers and the databases are quite expensive. Yeah, what's your monthly expense? Oh, we're going, just the Postgres database right now is like $160, $170. The server itself is $7 a month, I think. So altogether we're running about $200 a month in terms of expenses for now. Is that just a VPS, or who's hosting your Postgres?

Is it a provider? It's a company called Aven. It's a Finnish company. It's a database hosting company but they're really using DigitalOcean or anything. I think DigitalOcean, AWS, you can choose. I didn't choose Google Cloud for obvious reasons. It's interesting. It's one of the first products I've worked on where this sort of

cost efficiency is a big deal. Previously I've only worked on sort of business applications that are just glorified spreadsheets where it doesn't matter. It can run on any machine and it might get a little bit slow, you just throw more metal at it and it's fine it might get a little bit slow you just throw more metal at it and it's fine but with analytics so where are the needs coming from just from the data storage aspect yeah so that's one aspect to think about when it when it comes to sustainability but uh we still have some way to go a year is a good i think time frame to to reach financial stability with the company

one good headache we've had over the last few weeks, again, since this post, is that several bigger sites have come to us and said, you know, on your site, you mentioned only, you know, whatever it is, 100,000 is the top level. How about 3 million? How about 4 million? And now Oku is in the process of, you know, like you mentioned earlier, like, you know, kind of making scaling plausible and making it more efficient. So we can actually from at some point

in a couple of weeks, perhaps, we will be able to start beta testing some of these sites that have, you know, four or five million per month of users or visitors. And this is kind of another interesting aspect. Yeah. Do you have to introduce enterprise pricing at that point? I haven't checked out your pricing very closely.

Does it scale up pretty well? If somebody's going to have massive data needs, are they going to be paying more? For sure, yeah. Well, then you're good. I think it's going to be fairly, not linear, but the more you use based on usage, the more page views you have, the more it's going to cost.

I want to keep the pricing as simple as possible just based on traffic pretty much. We've talked about some pro features, but personally I'm not a fan of having complex pricing where you try to navigate the features to get people to upgrade. I prefer to just have a simple, like, here's the product.

As much as you use, you're going to pay for it. I like that. Cool, guys. Well, we hope you get there. We will be rooting along. We'll be following the blog. We'll be trying out Plausible. I'll definitely be checking out that roadmap

and waiting for the server-side stuff to hit. Oh, yeah? That'll be fun. Absolutely. Well, just an interesting question. Are you comfortable with us hosting the data or would you also want to run your own instance of plausible? I'm not interested in self-hosted whatsoever.

Right. We use Google Analytics right now. I mean, we're already comfortable with the beast, let's say. I am not a privacy far-ender that we talked about where our stuff comfortable with the beast, let's say. I am not a privacy far ender, we talked about, where our stuff has to be all self-hosted at all.

I like the server-side idea. I like the simplicity of it. I like not shipping any JavaScript to my users. So I like more about the user privacy than the hosting thing. I guess at the end of the day, well, I think you can probably host it more securely than I can, probably.

So I guess that is probably pro-user privacy versus us having to maintain that. But I do like the server-side idea, and I like not having to ship any JavaScript whatsoever. And those are the main thing. I like the simplicity, in terms of the user interface, you already have that going for you

in the product. The way you display the information is nice and neat and doesn't have a bunch of the stuff I don't want. There are a few things I do want that aren't there, but like I said, I'll just check out that public roadmap and maybe I'll give it a few thumbs up to some things.

That's interesting to me is that aspect and really just kind of a return to the old way of doing things, server-side versus in the browser. I'm with you. I think allowing people to self-host, it makes sense for Plausible. We're in that sort of space. Absolutely.

But personally, as a user, I'd rather pay someone to take care of it for me. It's not something I want to... I want to minimize my headache as a developer. There are people that love setting this stuff up, though. They love self-hosting, they love running servers, they got their Ansible scripts, or they got their Docker files,

or they got their Kubernetes clusters., or they got their Docker files, or they got their Kubernetes clusters. I mean, I get it. I used to do that kind of stuff for as a part of my living, which is why I'm so allergic to it now. And so if they could slot this in as just another one of their Docker containers that they're already running this infrastructure for their company, or for their home, or whatever it is, there are people that love that stuff. I'm just not that person. But I think it's a worthwhile endeavor for you guys. I just am not interested in it for us. But I am interested

in the server-side stuff. I don't know where you're at with customer base. I can kind of see it based on MRR, but I would say, you mentioned the next few things for you in terms of focus around customer base and dollars is sustainability. So I'd say

if over the next six months you can focus on the customers that you need to get to sustain, focus on features and roadmap that sort of enables that, that would be a wise focus of your energy and time. And sure, maybe at some point if a self-hosted version is certainly possible now, but it requires documentation and stuff, that that not be your focus until you're ready to take on that kind of customer base. You're going to attract

the, Jared's words, privacy-oriented, right? And so you don't want to say no, but just say not yet. Yeah, we actually spoke about, I don't know if you guys know ProtonMail. I used them for my email. And I ended up, again, in my switch from Google, I ended up having to pay for email, which I never considered before.

So I pay for ProtonMail. And their concept I really liked and something we discussed is that they have paid plans and they have several levels. And people are donating money and giving quite high levels they have you can pay for. And they use that money they get from paying customers in order to support those that cannot afford it or that don't want to pay. And they allow people to also have a free email address. So this is something we discussed is that if we build, like you mentioned now, if you build plausible and get some sustainability there, and that will then allow us to actually spend some of our time rather than building new features and trying to get new customers, we can actually spend some of that resources, some of those resources into allowing, you know, easy self-hosting and things like that, that will make it easier for

people that cannot pay right now or that don't want to pay and so on. Well, it's really been fun digging through all these details. Certainly applaud what you're working on. I mean, it's definitely in the right timing, right timing for people like us even looking for alternatives. And thank you so much for sharing your story and your time. Appreciate it, guys. Thank you.

You're welcome. All right. We want to hear from you. Go in the comments and let us know what you think about de-Googling your website analytics. Are you going to do this? Are you going to try it out? It's open source. As a matter of fact, we just integrated with Plausible to give it a try ourselves.

So really curious what you think about this. Open up your show notes and click discuss on ChangeLog News. We want to hear from you. And we get asked this all the time. What's the easiest way we can support you? Well, the easiest way is to tell your friends. The best way for this show or any of the show out there, any podcast really, is word of mouth. Tell your friends. Send a text, send a tweet, write a blog post of your favorite podcast episodes each month. That totally works and helps every podcast out there. Of course, thank you to

our awesome partners Fastly, Linode, and Rollbar. And thank you to Breakmaster Cylinder for those awesome beats. And one more thing, we have a master feed that brings you all of our podcasts in one single feed. It is by far the easiest way to listen to all of our shows. Everything we ship is in that feed. You'll miss nothing. Head to changelog.com slash master to subscribe or search for ChangeLog Master in your podcast app.

You'll find us. Thanks for listening. We'll see you next week. Here he is. Sorry, guys. Linux crashed. Yes. I should not use Linux for this thing. The file was too big. It crashed. Yes. I should not use Linux for this thing.

The file is too big. It crashed my machine. Are your lights off there? You have a power outage? We're speculating why your Linux crashed. I have a request right now. If I said anything about Linux on record, can you remove that part?

What's it called? I think even Audacity crashed. So let's see if i can if i can save um the recording i'm very sorry oh this right here is going in the post show for sure this is our best moment in months everything crashed so now that i have you off air let me give you my feature request real quick. Okay. Yeah, go for it. So what I'm interested in most of the time, so our site is pretty well designed from an information architecture point of view. So if you think about our different podcasts, if you think about our news and then our posts,

so we think of these, they're almost like little verticals inside of our media company. So GoTime is its own thing. It's a show for the Go community. It has its own episodes. It has its own stuff. And what we're trying to generally track is the success and growth of our portfolio shows.

So what I mostly want to know, and this is hard to get even inside Google Analytics, but I think it's a pretty basic thing, is I want to know the performance of paths. If I can just see slash go time and everything underneath it, think of your top-level nav. If I can show

just the stats on this path, how's JS Party doing? How is news, which is a whole section? How is our posts? Is posts growing? Not the individual posts, cumulative, over time or dropping? That's beautiful for me.

And that's so simple. That plus what you currently have is basically, boom. Yeah. We are in progress of doing that, no? That's what's going to happen, to be able to dig deeper into these levels. Yeah, what really sucks is I can go into the database and I can run that query. Totally.

But surfacing that as a feature is something that I'm not very, I'm not a UX guy. We get a lot of compliments on the UX. It does look nice. Well, I think if you knew the routes, you could put that in the user's hands and say, which routes matter to you?

Instead of saying, here's all your routes. Yes. Or even top-level routes, like which top-level routes matter to you? Can I build a dashboard for you on that kind of thing? Yeah, I think we need to think about some sort of way. I can think easily of a way

to add a filter on top where you select the property to filter by, which would be the path name, and then add a regex that you can filter by. And then maybe we could store some sort of, have some sort of stored filters stored search yeah and you could have i don't know just tabs with different stored filter filtered searches i don't know something like that i can think of it now thinking about

building it it's going to take me ages no i know i'm not expecting you to build this i don't want you to have a homer simpson card about next week? Yeah, yeah. Anytime next week. We'll do it. If you want our, what is it, $6? If you want our $12 a month. We'll feel about a yearly plan for two years straight. Boom. Well done.

Now we're talking. Well, the trick is. We'll build a feature for you on the open source product. Sure. That could be fun. Oh, that'd be awesome. Yeah. Well, I am an Elixir guy.

Yeah. You can help me out. I actually would get involved in that way. I wouldn't force my features on you. But if you had like, I know you're working on the on-ramp, but like I don't need the self-hosted version, but I need like easy dev setup.

Yeah. That would definitely help out. Yeah. If we were using it for sure yeah one tricky part is you can run it okay it's actually not

that tricky I've been on board in client projects where it takes me like half a week to just get the setup running I think with this one

you should be able to download it and mix test well now you need a click house database as well which is a bit trickier but you can just

docker pull that documentation is what it allier, but you can just Docker pull that. Documentation is what it all needs for you to be able to do that. Yeah, just a little bit of get going. But one of the things that it really needs is

test data, because if you just download it, you're going to have no data in it. Right. And generating test data is really difficult, so I'd rather take the...

I think what I want to do is once we have this sort of self-hosted dev setup is I want to take a daily or a weekly dump from the live demo and just give it to you as test data for development. Yeah, totally.