Google URL Encoding Update: How Bracket Handling Affects SEO

Technical SEO often hinges on the smallest details of how a search engine interprets a string of characters. A recent clarification from Google has highlighted a critical nuance in URL processing that could lead to unexpected duplication issues for site owners. John Mueller, Google Search Advocate, recently confirmed that raw square brackets and their percent-encoded counterparts are not viewed as the same URL by Google's crawling systems.

For many developers and digital marketers, the assumption was that ?filter[]=123 and ?filter%5B%5D=123 were functionally identical. However, the reality is that these two strings are treated as separate entities. This distinction is not an arbitrary Google preference but is rooted in the fundamental architecture of the internet.

The technical logic behind RFC 3986

The divergence in how Google handles these characters stems from RFC 3986, the official internet standard for Uniform Resource Identifier (URI) syntax. According to this standard, brackets and plus signs are categorized as reserved characters. Section 2.2 of the RFC explicitly states that URIs that swap a reserved character for its percent-encoded version are not equivalent.

Mueller noted that raw brackets in a query string are technically invalid under these rules. In the strict eyes of the RFC 3986 standard, brackets are reserved specifically for the hostname to wrap raw IPv6 addresses. When they appear elsewhere, such as in a query parameter, they should be encoded. Despite this technical invalidity, Google still crawls raw brackets, which creates a scenario where two different versions of the same page can exist simultaneously in the index.

How Google manages duplicate bracket URLs

When Google encounters both the raw and encoded versions of a URL, it treats them as two distinct pages. If the content on both pages is identical, Google's algorithms will attempt to eliminate the duplication by selecting one version to be the primary representative in the search results. This process of canonicalization, however, is not always predictable.

If a website links to the raw version in some places and the encoded version in others, it leaves the decision of which URL to keep up to Google. This can lead to inconsistencies in how traffic is tracked and how link equity is distributed across the site. Mueller admitted that he had likely provided the wrong answer on this specific topic in the past, emphasizing that the JavaScript checks now confirm these URLs return a false result when compared for equality.

Consistency across the SEO stack

To avoid the risks associated with duplicate content and fragmented indexing, the recommendation is absolute consistency. It is not enough to simply pick a format for the visible links on a page; the choice must be mirrored across the entire technical infrastructure of the site.

Site owners should ensure that the chosen format—whether raw or encoded—is used uniformly in the following areas:

• Internal hyperlinks throughout the website architecture.
• Canonical tags defined in the HTML head.
• XML sitemaps submitted to Google Search Console.
• Structured data and schema markups.

It is worth noting that while the case of the encoding (e.g., %5B versus %5b) does not make a difference, the presence of the encoding itself does. Hex characters are treated as equivalent regardless of case, adhering to the rules set forth by the RFC standard.

Broadening the scope to other reserved characters

The bracket issue is a symptom of a broader rule regarding how Google handles URI syntax. The same logic applies to other common characters that often cause confusion for developers. Specifically, the rules for duplicate formats apply equally to spaces, plus signs, and the %20 encoding.

When a system generates URLs dynamically, it often alternates between these formats based on the library or language used. For instance, some systems might convert a space to a plus sign, while others use percent-encoding. If these variations are not standardized, Google may see them as separate URLs, triggering the same duplication and selection process described with the brackets.

The intersection with AI-driven search results

This technical precision in URL handling arrives at a time when Google is fundamentally changing how it presents information via AI Overviews. While the bracket issue affects the crawling phase, the way links are presented in AI-generated answers affects the clicking phase. Recent observations indicate that not all links within AI Overviews lead to external websites; some merely trigger new Google queries or open AI Mode conversations.

As detailed in reports from Search Engine Journal, the distinction between a click to a website and a click to an AI-driven follow-up is critical for measuring ROI. For businesses, this means that technical SEO—such as fixing bracket encoding—is more important than ever. If Google's AI selects a non-canonical or incorrectly encoded URL to feature in an AI Overview, it could lead to tracking discrepancies or inefficient crawling of the source material.

Global implications for international enterprises

For businesses operating in the USA, UK, and global markets, this update underscores the necessity of a centralized technical SEO governance model. In large-scale e-commerce environments—common in the US and UK markets—where filters and facets are heavily used in URLs (often utilizing brackets for arrays), the risk of creating millions of duplicate URLs is high.

From a regulatory and operational standpoint, while there is no specific US or UK law governing URL encoding, the impact on data integrity and analytics is significant. For enterprises managing multi-regional sites, ensuring that URL structures are standardized across different locales prevents the dilution of page authority. In the context of the global digital economy, where AI-driven discovery is replacing traditional keyword searching, the technical cleanliness of a URL is no longer just a best practice; it is a requirement for visibility. Companies should audit their URL generation logic to ensure that RFC 3986 standards are met, thereby removing any ambiguity for Google's crawlers and AI systems.

FAQ

Sources: Searchenginejournal (2), Semasocial ·