Web Tracking in 2019, a Seismic Shift in Customer Analytics

Tectonic Collision and User Privacy

There are two major trends we have witnessed recently. First, internet privacy being driven by its users, which proliferate through the rise of ad blockers. These naturally also limit the tracking of users by third-party applications which are customer analytics for marketing. It is estimated that at least 22% of the internet’s users use these. If you are interested in details, check out this piece by Emarketer.

The second is driven a lot more aggressively through the dominant market force, the browsers themselves, especially Apple’s Safari or most recently Mozilla initiating a massive shift and imposing great restrictions on third-party trackers. The market share of these two is another 20% of internet users across devices. So this represents over 40% of internet users combined (but of course, there will be overlap) that your 3rd party analytics can’t see and track consistently over time, or at all.

Long story short, the reality is that as a result, your web analytics have more holes in their coverage than fine Emmental cheese. We would say this is kind of a big deal for third-party service tracking; and if I were a DMP vendor or someone who recently purchased a DMP, my palms would be sweaty.

New Era of Data Privacy. Thank You GDPR!

Once upon a time, the internet was a beautiful idealistic place of the free: Your data was yours, dear reader, and you were more than a cookie number in some transactional database sold to everyone who wrote a cheque. With a big push for user privacy through legal reforms such as the GDPR and PDPA, we are finally seeing some light of transparency and privacy as its global adoption grows despite the initial confusion. This is great news for the users, but for brands locked in current technological frameworks of web tracking, this poses a challenge to their current SaaS analytics toolkits especially for customer analytics. And brands will have to find a solution fast.

The Enterprise Problem

The downside of this current privacy sweep is that these initiatives are also limiting legit use-cases. In the digital age, if you are a company, you have your own website, online app, product, or e-commerce platform and you want to know what your users are doing in your web sites and apps and where they came from.

You care and have a right to care, to understand how your customer analytics - how they are behaving in your product and what behavior they exhibit. You need to be able to do this consistently over time. And for this, you need data.

The problem with SaaS tracking tools is that they use one backend for all of their customers. Meaning that from the point of view of browser, one entity (one server/one provider) knows a lot of data about web visitors from a vast number of different websites, effectively tracking a lot of behaviour across different sites about a huge volume of people. This is referred to as third-party tracking which is the target of ad blockers and privacy-focused browsers.

![data-collection4.png]

The Solution

To successfully solve this, you have to have your own, private, separate, data collection tool that you control completely and is perceived by browsers as a native part of your application. You could have your engineers build a custom framework, which is definitely a legit option for large companies where the web app is mission-critical. Alternatively, you could use a solution like Snowplow which is open source and allows you to run it on your own server and domain, which solves a big part of this problem. It is, however, not super user-friendly when you get into advanced settings.

How We Built For Privacy

We have built our solution with consideration for the privacy of customer’s data, accessibility, and reflecting on the current change of climate in this space. Meiro Events is not SaaS and we do not use centralized architecture. We never keep data from clients in one place - each client has their own, separate, isolated, data storage. More importantly, we are not doing and we will not do any cross-client customer analytics. Or if they want to, they simply host the solution themselves - either in their private cloud, or on-premise.

Given this property of our solution, the data collection Meiro Events provides uses the same domain and principles for data collection as the main web application, making it essentially indistinguishable from part of your app. Simply put - to all "Tracking Prevention Systems", extensions like Ghostery, AdBlock, uBlock Origin, and others, it will look like a regular web application call, and not tracking.

This is a good thing because you are not centralizing the data. You are not providing it to some other entity. You own and you use it to power your own first-party use cases which is perfectly legitimate and ethical because you are doing it with user’s consent.

Now The Technical Bit

In order to demonstrate it, I would like to compare it to the techniques used in this amazing article by Moz. It describes very well how there is almost no difference between how you set up or name your Google Analytics tracker. It does not make a difference if you use or not use GTM, or if you rename the loaded script, or if you self host it, or if you rename the trackers. And it has one very simple reason - All these trackers are sending the events data to the centralized location - Google servers. This is not desirable, because it gets picked up by ad blockers and privacy extensions. Extensions that are used by smart users, or by simply enabled by default in privacy-focused browsers like Brave or Firefox. But do you know who does not send data to third-party servers? Meiro Events!

If you think, because you are no doubt very smart since you are already reading this, that simple proxy server on your own domain is going to solve this, then let us save you some time, we tried it and it won’t. You will lose lots of valuable data like the location because all the web traffic will appear to be coming from your proxy server.

We have our website https://meiro.io and Meiro Events deployed to the events.meiro.io. It looks like it is a part of the app, we collect data just for this domain and keep it on our servers. No one else has access there, just us. On your website, it is going to look exactly the same.

We offer the possibility to deploy our solution to any domain - If your company has a website with domain "example.com", we can deploy as "e.example.com" (just an example - you can choose the subdomain).

The effect can be seen right away.

An Example of How We Do It

Look at the screenshots here - They demonstrate that we are collecting events from our web to our Meiro Events instance in Brave (privacy focused browser), in Chrome with Do not track enabled, in Firefox with strict tracking prevention enabled, in Safari with their newest ITP2.2. All those browsers have uBlock Origin, AdBlock and Ghostery installed (where available).

And how it goes? All greens - All web events collected. And it is because we are collecting data for ourselves, as a part of our application. We are not sharing the data with anyone and it is for our own purposes. And now you can do it too!

(https://res.cloudinary.com/dnxlpdalu/image/upload/v1687201809/data_collection4_7c81a1c1bb.png)

Examples demonstrating the successful data collection in different browsers. Disclaimer: This is kind of user tracking doesn’t have the ambition to compete with your general purpose web analytics. It is not built to replace this function. It really has a completely different purpose. Here you are tracking an identified individual known only to you instead of ‘an audience’. Simple example is demographic data that tool like GA has available. With the amount of data GA processes across all websites where it runs, it is possible to use data science to attribute to different users things like gender and age group based on exhibited behavior such as types of pages visited.

Why Should You Even Care?

Customer analytics, or to be precise, well executed strategic customer analytics in mature markets is the differentiator between the brands that make it and those who don’t. Not just the analytics but the execution and activation of it of course. In contrast to some popular views, we are seeing the slow transformation of this practice in Asia, where large, slow-moving behemoths in denial of change are naturally being disrupted by the agile innovators. Just look at the payments market in Indonesia.

An additional factor here is that in the fast-growing nature of South East Asian markets, most companies are still concerned with growth and acquisition and are still neglecting customer experience and retention.

The fuel that analytics run on is data. Unless you have a customer data that is well structured, connected and accessible, you can have the best tools, best tech stack, and flashiest dashboards, but it won’t add up. Because as always, GIGO!