Taking a Simple Approach to Analytics ![rw-book-cover](https://img.youtube.com/vi/8JTRVoCzWBw/maxresdefault.jpg) ## Metadata - Author: [[YouTube]] - Full Title: Taking a Simple Approach to Analytics - Category: #articles - Document Tags: [[video-notes]] [[youtube]] - Summary: Taking a Simple Approach to Analytics - URL: https://www.youtube.com/watch?v=8JTRVoCzWBw Hi everyone and welcome to another episode of the WP Minute. My name is Eric Carvak and I'm your host. Uh today's guest is Danny Vancouin. He's the founder of Iberode uh which is a European software company and you may know him from some very popular WordPress plugins such as Mailchimp for WordPress and Coco Analytics. Danny, welcome to the WP Minute. >> Thank you, Eric. I'm happy to be here. >> So, I want to get into Coco Analytics. So, I kind of want to dive into that, but first I I read your bio and we'll put a link to this in the show notes, but um you have a very unique story about how you started with uh creating WordPress plugins. Um apparently you were recovering from appendicitis. Is that right? >> Yeah, that's right. Um, I had this idea to create a WordPress plug-in and I had it in my head for about 5 years, but never actually went for it because I was so busy uh working for clients and just um yeah, making a living. But then uh I was on holiday in Vietnam and uh I was hospitalized with appendicitis. So, I had like 10 days uh of bed rest and it was a great forced uh rest which allowed me to start some long-term projects. It's kind of amazing how something like that I mean no one wants to get appendicitis obviously and especially not while you're uh visiting another country. Uh I can only imagine how uncomfortable that must have been but uh you turned it into something very positive. Um, and I see that, uh, Mailchimp for WordPress has over two million active installs. So, congratulations on that. That's quite an accomplishment. >> Thank you. Uh, I've been very happy uh with the appendicitis just because of what came out of it. Um, I just know for sure that otherwise I would never have taken the time because I've already uh been meaning to do it for 5 years and never went for it. So, um, and even the hospital, it was I had a great time. Uh, everything went fine. uh as soon as I knew what was actually going on with me. So, um yeah, I'm just very happy that um it happens weirdly. You never know when something like that is going to change your life for the better, right? Um very very different uh way to start uh getting into plugins. So, I mean, were you uh like freelancing before that? >> Yep. uh I was uh freelancing and I think I've created a Mailchimp signup for like over over five times or maybe 10 times and I just have to keep doing the same thing. So, uh I figured I should make a plug-in for this. Um never did it obviously until the hospital happened and I finally got around to it. >> Well, I bet a lot of people are happy that you did that. I mean 2 million active installs is is is quite something. Um, and then I don't know um if you want to go into the uh genesis of Coco Analytics like how did how did you um decide like that was was that like your next project after Mailchimp for WordPress took off? >> Um I I've had a few plugins in between which did quite well but uh not on the scale of Mailchimp for WordPress but about uh maybe 50,000 users at most. uh one for pop-ups that I'm not so proud of anymore. Uh one for contact forms, but uh I think I did a toy project. I was trying to learn a new programming language uh back in 2016. And uh I'm I used it to build an analytics product. Um and then two years later, I think GDPR uh came into effect here in Europe. So the topic was quite well hot. Um, and I started Fetam analytics. I'm not sure if you've heard of it. It's it's quite popular nowadays, but um, I think we grew it to about uh, 500 or maybe 1,000 monthly users. Um, but I became a new dad during his first year and then during the second year, my second daughter was born and it was just very hard to struggle. So, I uh, got out of the project and it continued with another founder stepping into my place. Um, so and it's been doing very well after that as well. So it was uh very well I I've had some mixed feelings seeing it grow without me obviously but I couldn't let the topic go completely either. So that's where I started Coco Analytics just focusing on WordPress users and also >> Yes. Oh, go ahead. I'm sorry. >> Yeah. One thing that I really didn't enjoy back then with Fetam was keeping everything online for our customers. Uh I had no experience with it and it was just quite stressful, especially if you're not sleeping because of your uh children. >> Yes. Yes. >> So I figured if I build it on top of WordPress, I can piggyback on all the amazing WordPress hosting out there and um I don't have to deal with any hosting. I can just focus on the code for uh recording and showing the uh data. >> Yeah, that makes sense. So I mean you had uh with previous product you had a server I I assume then that that tracked all this for your clients and if that went down I can only imagine the stress that is uh caused by that. Yeah, it was very stressful especially well uh we had a lot of small clients but as soon as we had a big client with with millions of page views um everything had to change and I just had to uh get uh behind a computer immediately to scale up the server. It was a manual thing back then um especially because we were growing at such a crazy pace that I couldn't do it correctly. Um in hindsight I should have but uh hey you live you learn. Well, yeah. This this is all a learning experience. And I I think it seemed like a logical step to go to that next place, which is WordPress, which already is self-hosted. You don't have to worry about a server. You can focus on making the plug-in simple to use, privacy friendly, which I think is is a huge deal these days. I mean, GDPR is all, you know, is is very much a concern. Even here in America, we we have to comply with it if we're, you know, um serving European customers. Um how how did you decide on on um WordPress though? Like what what drove you to that? >> Um well, initially Fet was an open source project. Um but after I left I had no control over it obviously and they decided to turn it into a closed source uh product which I think makes sense for them but for me the goal was never um to make it a big commercial success although it would be nice to make it sustainable. So I figured uh how can I reach a large number of people um and not have to deal with hosting people. Well, WordPress is obviously a huge part of the internet. Um, there's a lot of amazing hosting uh setups out there. So, and I had a lot of experience building on top of WordPress. So, for me, it was quite logical to build it on top of WordPress. And well, I obviously had to make some design choices as well um because it's not that easy to store years worth of analytics data into a WordPress database. But it happens that to um be a great match with the privacy friendly aspect because if you're not storing all the visitor specific stuff um then you can also build a quite a good product on top of a WordPress database. Yeah, I mean that that is nice and it's portable as well. Um you know we move your site to another host we still have all our analytical data which is a big deal. Um I I I think you know I I've been in in the web on the web for many years and Google Analytics has always kind of been that standard right for for analytics. People kind of expect it these days and yet I don't know a single person who likes Google Analytics 4. U Google's UI has become in at least in my opinion very dense and very difficult to use. Um, so like when you're building out Coco Analytics, like did you take a look at Google and kind of look at it and say, "Okay, this is where I need to improve on what Google does. This is where, you know, they fail and maybe I can make a positive impact." >> Yeah. Well, to be brutally brutally honest, um I didn't look at Google Google Analytics at all because even back in my freelancing days, um my customers expected me to uh install Google Analytics on their site. Uh but then when I came back to their sites months or years later, they never even looked at it even once. Uh or maybe they tried to, but they just couldn't find like how can I just see the number of visitors because it's not on the front page of your analytics dashboards. it's buried three uh pages down or something like that. So, I didn't really look at Google. I just um wanted one page that gave you an overview of what happened on your site. So, uh no custom reports, no multi-levelled pages, just a single page showing visitors and sources. >> That makes sense. I mean, because I I think the simplicity is what people are really after. We we talk about Google Analytics and you said as a freelancer people want you to install it and then um just that it's almost like an industry unto itself now that you have to be an expert at Google Analytics in order to set up the reports in order to get the emails sent out uh and and just customize things to to match what your clients are looking for. Coco Analytics is quite the opposite of that. You install it, you can see the stats right away. Um, our own Matt Maderos did a uh a little video tour of it recently, which we'll put in the uh show notes, showing just basically you turn it on and it works. And of course, the the pro version you can create reports and things like that as well. Um, but there's still like this this expectation, right, that that Google is going to be what's used. like what what advice would you have for like a freelancer or an agency whose client says I really want Google Analytics um but as a freelancer you're saying well I think Coco is probably the better option what what would advice would you give to them? Yeah, what I would really like people to do is like maybe start out with something simple like Coco Analytics because it covers a large number of use cases like uh if you just want to see what kind of traffic you're getting whether it's increasing month over month and where it's coming from uh which I think is covering 80% of small uh business owners website or even uh private users. Um, I just want the default to be something simple, like as simple as it could be. And then, uh, over time, maybe your needs are becoming more advanced, and you can always slap on something, uh, on top of Coco Analytics because it's so lightweight, it it doesn't get in the way. It doesn't take up any space, and it's not slowing down your website. Um, so it's easy to grow into more advanced users. What we uh experience with Koko is usually when when the default is not enough, they want custom events anyway. And even with Google Analytics, you will have to set that up. Um and with Coco Analytics, you can now track any type of custom event as well. So even when your use case is more advanced, it can probably be done in a simpler way than you would have with Google Analytics. >> Yeah. And the other part of that is that you don't have to worry about Google changing their platform or making it, you know, more difficult. Um, >> that that's been a problem too. I think, you know, just the switch from Google Analytics 3 to four. Um, that was a hassle for a lot of uh website owners and especially I think agencies that had a lot of clients, >> you know, having to to go through that. um a plugin like Coco, you're just basically hitting the update button when a new version comes out and you know everything still works. Uh which I think is nice in an age where everything seems to be getting more complicated. It's a it's a nice win. One other thing that uh I remember from from when I started Coco Analytics is I think back in 2016 the Google Analytics tracking script was was like 48 kilobytes after compression and minification. That's that's a crazy amount of JavaScript for for initiating uh well maybe one or two HTTP requests to actually lock the visitor. uh with with co with Google Analytics, it's 500 bytes added to uh your pages and that's it. >> Yeah, the performance part of it is definitely important. I mean I I could even recall I mean and I still see this with with some sites I manage that even if you you load up the browser console during uh analytics coming in it it litters your console with errors at times it slows everything down. Um sometimes you are waiting for the Google server to um respond and so that kind of you know it's one of those why is my website slow type of things. Well analytics can definitely do that anytime you're making a third party call like that. So with Coco everything it really is just on your site. It's not making third party calls right. >> Yep. And most likely there's already a connection open that can be reused for logging the visit. Um, so yeah, it's it's really I have a hard time believing it could be any faster than than this. >> Well, I I noticed also that you um you went on to a standalone version for folks that aren't using WordPress. Um, how is that going so far? Wow, that's I'm surprised you found it because uh I I've not been putting it out there as much because I think uh the relationship between the standalone version and Coco Analytics can be a little weird, but um I'm using it myself and I know of a couple of people who are self-hosting it and it's it's quite uh useful as well because you can host a single instance and you can use it to track multiple websites. So for instance, if you have uh a couple of WordPress sites but also a couple of uh static websites or or anything like that, uh you can use a similar looking interface um for both of these sites. Um so I'm not sure if I'm going to make it like it's public, it's out there, it's open source, people can use it, but it does require uh some Yeah. Well, some more advanced setup in terms of deploying it to a server that you can actually use it on. It's definitely not as easy to use as Coco Analytics, uh, the WordPress plug-in, but it it can be used and it's really nice if you have non WordPress websites, I think. >> But that's neat. It's nice to have the familiar interface. I mean, I can I can see like an agency that has a server that maybe wants to set something like that up to track all of their clients in in one in one go of it. Um, but I I thought that was an interesting um interesting side project. Um, so I I I I have to ask about AI because everything's AI now. Um, so your product is very simple. It's very straightforward, but we're also like in this age where everything is getting AI integrations. Um, I have to ask, do you see any sort of AI integration coming to Coco and maybe it's it's the going the opposite way might actually be better, but but what do you what do you say about that? >> Yeah. Well, if I'm if I have to be honest, personally, I'm I'm getting quite fed up with uh everything turning into AI or for Coco Analytics. I'm not sure I see a use case except for maybe um the surrounding um products like the documentation that you can search or that you can uh use an AI chatbot to summarize it for you or to help you write code snippets to track custom events or something like that. So I can definitely see some use cases for it. But in the product itself, um maybe to forecast uh projected visitor data, but I'm not sure. Personally, I would rather uh keep it u clear of any AI. >> Honestly, I'm happy to hear you say that. >> Really, I I I think I think we we've put it into so many places where it's not really that useful. It's just more of a race to say, "Oh, I've added AI to this. Okay, does it make it my job any easier? Does it give me information that I can't get by just looking at the screen?" Uh, you know, some of that stuff is just more of a um a marketing hype to me than it is um useful for the user. I think so. And >> to give an example, like we're using Help Scout to talk to our users and uh like a year ago, they started pushing uh us to start using AI to talk to our users. But uh and they also started pushing this to my team and I told them like don't click it. Please don't use it. I want to I want us to talk to our users oursel and learn from them and use it to make the product better. I don't want to save any time on it because it's fine. Like I like talking to users. I like learning from my users and I definitely don't want to hand it over to a computer. Um my users probably get maybe a faster answer but also a less accurate answer. So if they want to use AI to learn something about Coco Analytics, they can of course do so. But if they start emailing us, I want them to be actually emailing with us like a human and not uh the AI. So >> yeah, that you bring up a good point about just being able to learn from your users, right? I mean, how many I'm just curious like I h what kind of things have you added to Koko just based on user requests or maybe you saw like a use case that you hadn't thought of before? Well, uh, a lot of things like like most things are actually coming from users because um, my own use case is quite limited. Um, I don't really look at analytic data all that much. So most of the product as it is right now is is what I learned from users like they're using it in such a way. Um, they want to aggregate referers on a hostname basis. uh they care about which specific Google domain someone came from. Um they don't care about the specific Pinterest someone is coming from because Pinterest is one of the last few domains out there which is actually sharing the full referral information. uh just general stuff like that, but also which uh types of custom events are most users tracking and maybe we should make it a built-in event, stuff like that. >> Yeah, it something I hadn't considered with AI and we're using AI for support and all these other things. You are kind of disconnecting yourself from the people who are using your product, right? So you're kind of in a in a vacuum building features and yet you know you release them to users and they're kind of wondering what is this for? Who is this for? And you might um you know it's a missed opportunity as a product maker. >> I think so. like I I talk to AI a lot and I enjoy it, but um one thing I really missed from my freelancing days is that I ran into a lot of problems um that I could then use to scratch my own itch and maybe turn into a product because it's been my dream to build products ever since I started programming. But the ideas are a lot harder to come by now that I'm disconnected from building websites every single day. >> Yeah. And really the users are the only way to kind of get into that, right? I mean to to see what people are facing and and where, you know, you can take things. That leads me to, you know, what's next for you? Is there anything um new coming to Coco or or any of your other products we should be looking out for? >> I'm focusing on Coco Analytics right now. Um, and I'm also still involved with Mailchimp for WordPress, uh, which is still uh, like our largest product by far, but but Coco has been growing really well and I hope uh, one day that it can match the success of Mailchimp for WordPress. Um, we recently added geoloccation. Um, not because again, not because I care for it because but because a lot of our users actually uh, needed it. Um, and we're also adding um more advanced tracking of like what type of device or screen or browser is viewing this specific website because a lot of people are using that to base their uh design decisions on. So that's one thing that's coming and we're also working on adding uh tracking UTM parameters by default. So with something like uh just I'm just curious like something like uh browser and device tracking how do you do that in a way that's privacy friendly? I think the important part is like to not track any anything down to the version number or you have to be quite generic uh in a way. So maybe uh just the most popular uh operating systems and if it's a very specific operating system then maybe bucket it bucket it into a more generic um like a name instead of well if I had a visitor and I see in my analytics that it's using Linux Mint and it's been viewing uh five pages then I know for sure it's it's the same visitor uh because no one uses that uses that on my So I think it's important to to bucket things um so everything becomes slightly more generic but you can still use it for um well it's still a lot more info than not knowing anything about the operating systems. >> Exactly. I mean I think and we're in a we're in a time now too where all the browsers are automatically updated. I mean, I started in this industry a long time ago when, you know, a couple of versions of Internet Explorer came out a year and you had to manually update and not everyone did and they were still using an old version and all that. So, I I think the version numbers probably aren't as important now, I think, as they were, you know, 15, 20 years ago when we only had a few browsers that had to to to really work for. >> Yeah. And I think the standardization of browsers has also been improving a lot since um well Internet Explorer 6. >> Yeah, thankfully. Yeah, I mean I I can't remember the last time I had to look at just a specific browser and say, "Oh man, this just doesn't work." Um you know, things generally do work across browsers these days, which is thankful for for web designers and site owners, too. Um, so any anything else coming down the line that we should look for? >> Well, uh, we are doing our own anti-PAM uh, pro protection lately. And, uh, it's funny because we we have been suffering from a lot of bot signups which would just go through our purchase flow and then get stuck on the final step where you're making a payment. And, um, we we've been getting very good at stopping it. Um, one uh funny example is that they're all using the same user agent for every couple of weeks. So, we've been uh logging that extensively and we're building up quite a nice database um on user bots and the types of methods they use. So, I've been thinking of maybe building something and offering it uh to the to the general public. So, maybe I'm doing something like that. So you can protect your site from bot signups or uh comment form spam without requiring a a third party. I think that would be nice, but uh don't um uh pin me down on it, please. Well, I I can tell you there's definitely a need for it. Um and yeah, privacy focus, not using a third party would also be a huge bonus for a lot of people. Um because we have uh Cloud Flares, Turnstyle, we have Recapture from Google. Um, but both of those are thirdparty solutions. So having something that you can host on your own site and uh maintain the privacy uh aspect of it, I think would be awesome. So definitely let us know if you if you go ahead with that because we'd love to talk to you about it. >> I will. Thank you. >> Well, Danny, I want to uh say thank you for for being on the WP Minute. And uh where can folks find you and and connect with you online? >> I'm on Blue Sky since a couple of weeks. Uh I've not been as good in checking it like daily, but uh I'm getting there. So uh people can search for Danny Fan Cotton on Blue Sky and uh if they do, please uh send me a what do you call it on Blue Sky? Uh I think it's just a chat. >> Okay, chat with me there. Awesome. Well, thank you again for being on. Um, and uh, everyone, thank you for uh, for joining us here on the WP Minute today. Uh, be sure to uh, check us out at thewpmin.com/subscribe. There you can receive our newsletter. Um, get information on new episodes and also uh, become a member of our community. Um, thanks for being with us and we'll see you next time. >> Thanks, Eric. Thanks for having me. >> Thank you., Taking a Simple Approach to Analytics, YouTube, 01k9q29gprk20hgyfxtsb66s2h.