Introduction
DFINITY is exploring integrating an online analytics solution to monitor errors and aggregate usage data for the Internet Identity and NNS dapp.
This initiative aims to improve user experience and proactively address issues while maintaining the dapps core privacy and security principles.
The Need for Analytics
The lack of data makes it difficult to:
- Detect regressions or issues introduced by new updates.
- Prioritize bug fixes or feature development based on user impact.
- Understand how users interact with the applications across different browsers, operating systems, and devices.
Currently, feedback is limited to support tickets and forum posts, representing only the most vocal users. This creates a biased understanding of user experiences and leaves many issues unaddressed.
Additionally, Passkeys and WebAuthn are still evolving, leading to inconsistent behaviors across platforms. Without data, it’s challenging to identify and resolve these inconsistencies effectively.
Privacy and Data Handling Requirements
To ensure the applications remain privacy-focused, any analytics tool must adhere to strict requirements:
- No Cookies.
- No Personal Data:
- We’ll store UTM codes, referrer, country, device type, browser version, and operating system.
- Click and load events will be used to monitor the usage of features.
- No Individual data: The tool won’t collect data tied to any individual across sessions.
- Data Location: All data will remain within the EU, where privacy laws like GDPR provide strong protections.
- Compliance: The tool will comply with all relevant privacy regulations, including GDPR, PECR, and CCPA.
- No Sale of Data: Data will not be sold to third parties.
- Public Access: All analytics data will be made available to the community for transparency.
Community Feedback
So far, we have identified Matomo and Simple Analytics as potential solutions that meet the above requirements.
DFINITY values community input on this proposal. Your feedback will help determine which tool to use and how best to integrate it into the applications without compromising privacy.
Please join the discussion by sharing your thoughts.
Thanks!