Referer Reality
This post dives into the often-misunderstood world of web traffic attribution, contrasting the unreliable Referer header with the author's preferred method of custom query strings. It highlights the practical implications for site operators and offers a nuanced perspective on digital etiquette. Hacker News readers appreciate this blend of technical insight and opinionated web development practices.
The Lowdown
The article explores the intricacies of tracking web traffic sources, sparked by another developer's decision to outright reject URLs containing query strings. This developer, Chris Morgan, asserts a preference for relying solely on the Referer header for source identification.
- The author counters that
Refererheaders are frequently absent from traffic originating from emails or mobile apps (e.g., Instagram, Substack), leading to a significant portion of traffic being mislabeled as 'Direct' or 'Unknown'. - To address this, the author employs a custom query string,
utm_source=Robin_Sloan_sent_me, for all outgoing links shared from their channels, particularly email newsletters. - This practice is framed not as an analytical tool for the author (who claims not to collect analytics) but as a form of 'digital etiquette,' providing downstream sites with clear attribution for incoming traffic.
- The author notes that while this approach is generally beneficial, some major platforms, including YouTube, may not handle unexpected query strings gracefully, necessitating a list of exceptions for such cases.
The piece ultimately advocates for a more transparent and informative approach to traffic attribution through custom query strings, emphasizing the courtesy it extends to receiving websites, despite acknowledging that not all systems are built to accommodate them.