Discover valuable insights into Google’s leaked ranking factors and learn how to leverage them to enhance your SEO experiments and strategies.
Over the past week, I’ve encountered numerous arguments against delving into the 2,596 pages. However, the primary question we should ask ourselves is, “How can I test and extract as much knowledge as possible from these documents?”
SEO is an applied science where theory serves not as the end goal but as the experiment’s foundation.
14,000 Test Ideas
You couldn’t ask for a better breeding ground for test ideas. However, not all factors can be tested the same way. They vary by type (number/integer: range, Boolean: yes/no, string: word/list) and reaction time (the speed at which they affect organic rank).
Therefore, we can A/B test fast and active factors and use them before/after testing for slow, passive ones. To test ranking factors systematically, follow these steps:
- Select a Ranking Factor
- Choose the Impacted (Success) Metric
- Define Your Test Location
- Decide on the Test Type
Ranking Factors
Most ranking factors in the leak are integers, operating on a spectrum, but some Boolean factors are straightforward to test:
- Image compression: Yes/No?
- Intrusive interstitials: Yes/No?
- Core Web Vitals: Yes/No?
Factors You Can Directly Control:
- UX: Navigation, font size, line spacing, image quality.
- Content: Fresh, optimized titles, non-duplicative, rich in relevant entities, focused on one user intent, high effort, crediting sources, using canonical forms of words instead of slang, high-quality UGC, expert authors.
- User Engagement: High rate of task completion.
- Demoting (Negative) Factors: Links from low-quality pages and domains, aggressive anchor text (unless you have a strong link profile), poor navigation, and poor user signals.
Factors You Can Only Influence Passively:
- Title Match and Relevance: Between source and linked document.
- Link Clicks
- Links from New and Trusted Pages
- Domain Authority
- Brand Mentions
- Homepage PageRank
Starting Point: Assess your current performance in the area you want to test. For instance, start with Core Web Vitals.
Metrics
Choose the right metric for each factor based on the leaked document or your understanding of its potential impact:
- Crawl Rate
- Indexing (Yes/No)
- Rank (for main keyword)
- Click-Through Rate (CTR)
- Engagement
- Keywords a Page Ranks For
- Organic Clicks
- Impressions
- Rich Snippets
Where to Test
Identify the best location for testing:
- Skeptical? Use a country-specific domain or a low-risk site. If your site is multilingual, implement changes based on leaks in one country and compare performance with your core country.
- Isolate Tests: Limit one-page type or subdirectory to isolate impact better. Focus on pages targeting specific keywords (e.g., “Best X”) or user intent (e.g., “Read reviews”).
Considerations
Ranking factors can interact, enhancing or diminishing their effects, forming part of a complex equation.
Humans often need help intuitively grasping functions with many variables, leading to underestimating the complexity involved in achieving high-rank scores. Despite the high complexity, don’t shy away from experimentation.
Testing Dynamics:
- Aggregators find testing easier because they have more comparable pages, leading to significant outcomes.
- Integrators face challenges due to page-specific content differences, which dilute test results.
My Favorite Test: One of the best ways to deepen your SEO understanding is to rate ranking factors by your perception of their importance, then systematically challenge and test your assumptions. Create a spreadsheet with each ranking factor, rate it from zero to one based on its perceived importance, and multiply all factors.
Monitoring Systems
Testing provides initial insights into the importance of ranking factors, but monitoring is essential for measuring these relationships over time and reaching more robust conclusions.
The goal is to track metrics that reflect ranking factors, such as using CTR to gauge title optimization. We can determine if our optimizations are effective by charting these metrics over time. This approach is similar to regular (or what should be regular) monitoring but includes new metrics. You can build monitoring systems using tools like:
- Looker
- Amplitude
- Mixpanel
- Tableau
- Domo
- Geckoboard
- GoodData
- Power BI
The tool is less critical than selecting the right metrics and URL paths.
A Leak of Their Own
Google’s ranking factor leak isn’t the first instance of a significant platform’s algorithm becoming public. Here are a few notable examples:
- January 2023 – Yandex Leak: Yandex’s ranking factors were leaked, revealing many similarities to those found in the recent Google leak. The lackluster reaction at the time was as surprising as it is now.
- March 2023 – Twitter Algorithm: Twitter published most parts of its algorithm. While it lacked “context” between the factors, it still provided valuable insights.
- March 2023 – Instagram Insights: Instagram’s chief, Adam Mosseri, shared an in-depth post on how the platform ranks content across different sections of its product.
Despite these leaks, no known instances of users or brands ethically hacking these platforms exist. The more a platform’s algorithm rewards engagement, the harder it is to game. Google’s algorithm leak is particularly intriguing because it’s an intent-driven platform where users express their interests through searches rather than behavior.
Understanding the ingredients for success is a significant step forward, even without knowing the precise quantities.
It’s puzzling why Google has been so secretive about its ranking factors. While I’m not suggesting they should have disclosed them to the extent seen in the leak, transparency could have promoted a better web environment with fast, easy-to-navigate, visually appealing, and informative sites. Instead, the secrecy led to excessive speculation, resulting in poor content and algorithm updates that significantly cost many businesses.
If you need help, check out our monthly SEO packages and let the experts help you.
