With the introduction of Chrome’s new user-agent string structure, web analytics solutions need to be prepared for future updates.
Google is changing its user-agent string to provide a better browsing experience for users. The change will affect all browsers on Android devices.
The User-agent string is sent with every HTTP request to the server. This string can be used to fingerprint users. This means you can use it to identify these people even when they’re using private browsing modes or incognito windows.
There are many reasons why an attacker might want to know what browser you’re using. On one end, they could be trying to determine if you’re using an insecure connection (e.g. HTTP instead of HTTPS). Moreover, they might be able to take advantage of security flaws in certain browsers (e.g. Opera).
Before we get into the details of how you can make sure your analytics tool can keep up with the upcoming browser changes, let’s see exactly how user-agent reduction works.
What is user-agent reduction?
User-agent reduction is the process of removing unnecessary information from your website’s user-agent string. The user-agent string is a part of every HTTP request that your browser makes to a web server. It’s used by servers to identify browsers and sometimes even versions of browsers. This way, they can provide different content or block certain features.
What this means for marketers?
In practice, this change will impact various areas of your marketing campaigns, including
- Anonymized online user activity within web browsers (and consequently improving the end-users privacy). The user-agent string is a piece of data that web servers, browsers, and operating systems send as part of HTTP requests. It contains information about the device and software being used by the end-user. This lets you identify individual users and track their browsing habits. The idea behind user-agent
reduction is to anonymize this data by reducing its size or removing some of its fields altogether, so it cannot be used for tracking purposes.
- Websites will no longer have the ability to adapt to different devices. They’ll still be able to do it to some extent but will need to constantly adapt to new ways of extracting information. On top of this, it will be harder to distinguish between Chromium-based browsers and Chrome in your analytics, as they might appear as a single traffic source.
- Conversion modeling might require collecting consent (yay to banners!) to collect different kinds of information. You’ll have to re-think your consent banners and make sure you’re covering ALL data you’re tracking right now.
How to get prepared for the upcoming browser changes?
In light of the future browser detection and user-agent reduction changes, we’ve prepared a list of steps you can go through now to ensure you’ll be tracking data the right way:
1. Make a plan
Create an actionable plan and timeline for preparing for the changes. You’ll want to make sure you’ve got dedicated team members prepared to take on this challenge whenever needed, as well as dev and web analytics tools to stay ahead of new standards.
2. Be mindful of the timeline
Keep an eye on the official Chromium timeline updates to never miss a change. New updates are added along with each phase rollout as more users will be impacted.
3. Have your developers review the current website code
Take a look at how you’re currently implementing web analytics. Which tools are you using today? Do they support the new user-agent reduction change? Have your engineering team review the official documentation and ensure your website’s code is up-to-date. This is the first step to ensuring you know exactly where and why user-agent data is being used on your website.
4. Test the updates to make sure you’re getting the right data
5. Optimize critical hints
Client hints are a new feature in HTTP/2 that allow servers to tell browsers what they want them to do. There are two main kinds of client hints: high entropy and low entropy. High entropy client hints require the browser to have some knowledge about the server’s capabilities before making
a decision, while low entropy client hints assume the least amount of knowledge.
Two key rules here:
- Focus on low entropy client hints. For instance, low entropy client hints might provide some general information, like the fact the browser is on a mobile device. Based on this information, you can opt to show users a pop-up suggestion to prompt them to switch to your mobile app.
- Avoid high entropy hints. These might not be shared by browsers in every case. On top of this, their fingerprinting capabilities can turn into personal data, which will inevitably require consent. High entropy hints assume that the browser knows a lot about your site. For example, you’ll know the exact browser version and every minor detail. In turn, this can support rather accurate fingerprinting.
6. Switch your analytics solution
Whether you want to get accurate data or personalize the experience you deliver to your customers, you’ll need a web analytics solution you can trust. Keep in mind, not every solution is ready to support the change.
Using reliable web analytics tools like Wide Angle Analytics that can support upcoming changes lets you stay in control, so you’ll fully understand your audience and drive better conversions.
7. Engage with other developers and marketers to get help from the community
The change is still in its earliest stages, so there’s a lot of uncertainty as to how things should be done and what impact they’ll have. Make sure you stay connected to other developers and marketers who are actively testing this on their websites. There are already loads of threads and support repositories you and your team can access to clarify what changes you need to make.
What to do next to future-proof your marketing strategy?
If you want your marketing strategy to be future-proof, there are some things you should do right away. One of them is to start handling ads and other marketing campaigns in a privacy-oriented environment.
This is important because many of today’s online users are becoming more aware of the issues around privacy and data protection. They’re starting to act on their concerns, which means they’re using ad blockers more often, and they’re spending less time looking at ads.
Not all web analytics offer this level of support today. Switch to a web analytics solution like Wide Angle Analytics that’s prepared to handle upcoming changes.
Finally, revisit your current strategy to use client hints appropriately. Do you rely on obscure details of your users’ devices? If yes, stop, and refocus. Not getting accurate data will only result in a bad user experience. Instead, we should focus on what is best for the user and make sure that they can access the content they want on any device they use.