For nearly two decades, Google Analytics has been the go-to tool for marketers and data analysts.

But, as we get toward AI-powered search and privacy-first data collection, one question is quietly emerging: will there even be a Google Analytics 5?

The 2 main things that limit the number of people you track are:
1. Cookie consent banners
2. Ad blockers

Depending on the profile of your visitors these things alone, can stop you from tracking more than half of your visitors. In this episode of Growth Detectives show I am going to tell you how you can prevent that from happening.

So, let's start with consent banners.

The truth is that privacy laws in over 60 countries require that website owners ask their visitors for consent before tracking. This consent should be given by clicking a button in a consent banner. Unfortunately, a big portion of visitors do not do that. And as a result, they cannot be tracked.

So what to do about it? Should you just disrespect that and risk being fined?

There is a better solution.

The thing is you need to ask visitors for consent only when you want to track personal information about them and / or use techniqies that can identify them. One of them is cookies, from hich the banners took their name. However there are in fact many more of them.

So what to do about it?

Do not use a tracking tool, that tries to identify your visitors or collect their personal information. As simple as that.

Then, you won't have to ask visitors for consent and will be able to track many more visits (although a little bit less accurately).

However, there is still one problem, that can stop you from tracking a large portion of your visitors - ad blockers.

This is because ad blockers actually not only block Ads themselves, but also prevent tracking.

They do it either by blocking scripts from specific sources (like Google Analytics) or by stoping them from sending data to specific domains (like Google's).

However, there are 3 ways to make it more difficult for them:
- server-side tracking
- local hosting
- ad file proxing

let's start with the server-side tracking.

The way it works is simple. Instead of sending traffic information directly to the tracking platform, the information is sent to your server and only then to the server of the tracking tool.

The downside of this, however, is that your server has some additional work and may require an upgrade.

Next we have local file hosting.

In this case, the tracking script is not loaded from the domain of the tracking tool, but directly from your own server. Sometimes they are also renamed to cheat ad blockers. The downside of this solution, is that only few ad blckers can be cheated this way and the tracking file needs to be frequently updated.

Last but not least, we have file domain proxying. In this case, the tracking file is still kept on the server of the tracking tool, but ad blockers believe that it is in fact hosted on your own server. This technique however has the the smallest impact of these three.

So, what solutions should you use to track as much visitors as possible?

Definitely a tool that does not require visitors consent and - if possible - offer server side tracking option.

And these are all the solutions you can use to maximize the number of visitors that your website tracks. Thank you.

Hello, and welcome to another episode of Growth Detectives. I am your host, Chris Planeta. And in today's episode, we are going to talk about conversion attribution and different attribution models. And also I'll tell you why I don't like using attribution modelling, why I will never include attribution information in custom reports that I make for my customers and what I use instead.

So let's dive in.

First, what is attribution modelling? What's conversion attribution?

For those of you who don't know, conversion attribution has one goal - to attribute part of conversion's value to different traffic sources that took part in the customer's journey. So for example, if somebody visits a page five times, every time from a different source from Google Ads, Facebook Ads, Google search, et cetera; each of these, let's call them, touchpoints should get 20 percent of the conversion value.

But in reality, it's never so simple. Why is that?

First because for the data to be precise, you actually need to know the full customer journey with all the touchpoints. However, nowadays, it is almost impossible to know it.

Why is that? First of all, because users use different devices, they switch from their mobile phone to a laptop to a company computer. And so you can never know. You can never be 100 percent sure if somebody visits your website for the first time or is it a consecutive visit.

Next, people use browsers or browser extensions that block tracking scripts. And these tracking scripts are indispensable to start tracking attributions. So this is reason number two.

And reason number three - the problem with Apple iPhones.

The thing is that iPhones or applications in iPhones that open websites or open links - usually I'm talking like about social media applications - they don't pass referral information to the website when it gets opened after somebody clicks a link in those apps.

And so Google Analytics or other tracking tools actually have no way of knowing whether someone came from an application or from a different source.

There is simply no referral information.

And the fourth problem is with those tracking tools themselves. Because even though there is a problem with referrals or missing referral information in iOS devices however, this information is present in Android apps. So when somebody clicks a link in a Facebook Messenger on Android, there is this referral information getting passed through.

However, this information is simply not understood by Google Analytics or other tools. And, all of this means, that the touchpoints, the traffic sources that you're seeing in your reports, are very much incomplete. So, in my view, this is one huge disadvantage of conversion attribution.

The second reason why attribution modelling, in my view, is something that you shouldn't actually touch, is the fact that, well, there are many different models of attribution. You can attribute the conversions or the whole value of the conversion to the first source to the last source, divide it equally among all the sources or some combinations of them.

There are also AI engines that try to solve this riddle - to put different values to different sources. But depending on which model you choose, you will get different data. So, well, actually, so what's the best model? There is no, best model. And that's the problem. So if the data that you based your conversion attribution on is incomplete and the attribution and every attribution model is imperfect, then why to use attribution modelling? So, these are the two reasons why I don't use attribution modelling. However, I will tell you what I use instead, and this is a very simple solution. However, I have not seen it discussed anywhere on the internet. So let me share it with you.

The idea is very simple. When people visit your website, they do some actions. Some of them leave the page after three seconds. However, other people can stay there for longer.

They can, I don't know, sign up to newsletters, they will open galleries, view products they can add products to cart, et cetera. So what I do is, I track these actions and apply point value to all of them. And this point value is then sent to tracking tools.

And after some time, I create custom reports and see how many points were gathered by specific traffic sources.

Like for example Google Ads or a particular Google ad gathered 650 points, then a different one got 20, things like that. So even though I don't attribute conversions or a part of the conversion value to these sources I can clearly see whether people who come from a specific source are potential customers, whether this source is valuable or not. And then this information is also very, very valuable to the marketing department.

So yeah, this is basically it. Take care. Bye bye.

Hello and welcome. I'm Chris Planeta and I'm a host of Growth Detectives show. This is our fourth episode to date. In this episode, we are talking about making web analytics reports dashboard - a dashboard that will let you see how your company or how your business, how your website is doing and how it will be doing in the near future.

To create such report we need to start with the quality of traffic sources that you need to have, a number of active users that you are getting from each traffic source. And what I mean by active users? An active user is not a person who comes to your website and leaves 10 seconds later.

It is a person who spends there some time, for example, a minute. It is a person who views a few pages. It's a person who scrolls your website and clicks some buttons, makes some actions that can tell you that this person actually likes what they are seeing. They are interested in the content of your website.

And if you have these numbers of active users that you're getting from different traffic sources, then you can compare them to the cost of their acquisition. And this will tell you how much it costs you to get an active user.

It will be also a number that you will be able to improve in the future to lower the cost and to increase the number of the number of users you are getting.

But this is step one, but there are also other things that you should put in your analytics dashboard.

Next things you should put are actually data of how many people you have on each step of your customer journey. So for example, if you have a web store, you could put their information, how many people are interested in your products, how many people added something to cart, how many people started checkout and checked out and made a purchase.

Some of this data is very easy to get some, some of it can be more difficult. So you need to set up some some custom segments in your analytics tools.

However, it is all achievable and when you get those numbers, you can put them in your analytics reports and turn them into a funnel, which will show you not only how many people there are in numbers like 300, 400, et cetera, but also how many people drop off on every stage. For example, out of a thousand active users, I'm getting only 300 people who are interested in specific products, only a hundred people who added something to cart, only 50 people who started checkout, things like that.

And yeah, and actually this is it.

You don't need anything more to know how your website is performing and how it will perform in the future, or actually how your business will perform in the future. Because if you know your conversion rate, which is like 2%, for example, from the number of active users to the purchase, then you can easily say that, okay, so now I have like a thousand active users. The conversion path takes about a week and the conversion is 2%. So in a week, 2 percent of those thousand users will probably buy something from me. So this is something important, something that will let you plan for the future.

And also all these numbers you are getting, even though they are very simple, they will also act as something to improve on, because you will be able to improve the number of people you have on each stage and also increase the percentage of people that move from one stage to another one.

And yes, and this is everything you need to have great web analytics reports dashboard. And for those of you who are more into web analytics, who are professional marketers and are wondering "Hey, why there isn't anything about attribution conversion or attribution modeling" I will actually record another episode of this podcast. It should be out tomorrow or in two days and I will tell you there why I think attribution modeling does more harm than good and it shouldn't be used at all, actually, and especially not in your web analytics reports.

Thank you.

​