Other entertainment subscriptions fell more among Twitch lapsers
Decline in the average number of other active entertainment or gaming subscriptions per subscriber from Q3 2025 to Q2 2026
Decline in other active subscriptions
Source: YipitData U.S. e-receipt panel. Retained Twitch subscribers n=1,987; lapsed subscribers n=1,045. Both cohorts were established panelists with a Twitch subscription receipt in Q3 2025. The measure covers 12 named services and excludes Twitch.
Chart data
| Series | Decline in other active subscriptions |
|---|---|
| Lapsed Twitch subscribers | 14.2% |
| Retained Twitch subscribers | 3.9% |
How many other subscriptions did each Twitch group hold?
Lapsers averaged 0.76 other subscriptions in Q3 2025 and 0.65 in Q2 2026; retained subscribers moved from 0.87 to 0.83. Lapsers began with a slightly smaller wallet, so the comparison uses each group’s change and does not show that the Twitch lapse caused the wider decline.
Did overall receipt activity fall by the same amount?
No. Non-Twitch receipt activity fell 4.0% among lapsers while rising 4.4% among retained subscribers, and lapsers still generated about 90 non-Twitch receipts per quarter afterward. Because their other receipt activity remained high, the steeper 14.2% decline is unlikely to reflect subscribers simply dropping out of the panel.
Did observed patronage shift to Patreon?
Not in the measured receipts. A Patreon payment receipt appeared for 1.2% of lapsers both before and after the Twitch lapse; the retained group moved from 1.4% to 1.7%. Kick and most other creator-tip services were not meaningfully captured, so this shows no observed shift to Patreon rather than no competitive switching at all.
Assumptions & Methodologies
We analyzed 3,032 established U.S. e-receipt panelists with a Twitch subscription receipt in Q3 2025. Retained subscribers (n=1,987) had another Twitch receipt in Q2 2026; lapsed subscribers (n=1,045) did not. We counted distinct active subscriptions across Netflix, Spotify, Hulu, YouTube, Disney, Peacock, HBO Max, ESPN, PlayStation, Nintendo, Discord, and Crunchyroll, and checked non-Twitch receipts for continued observability. Apple and Amazon intermediary billing was excluded. Kick was not meaningfully captured. Starting wallet levels differed, and email receipts may miss services. Results show co-movement, not sequence or causality.