Reference:
Article : Wu, M. W., & Ham, S. H. (2026). More Ads, More Viewers? Analyzing Behavioral Shifts from Advertising Permissions to Live Streaming Consumption. Journal of Marketing, 90(2), 96–114. https://doi-org.devinci.idm.oclc.org/10.1177/00222429251363144
Key words: Live streaming, difference-in-differences, midroll advertisement,machine learning,causal inference,digital marketing.
Summary:
This article examines how Midroll advertisement might affect live streaming viewer’s consumption behaviour.
It further analyzes how advertising permission can affect:
The number of views
The duration
Audience engagement
The study is based on panel data from a live streaming platform .
They applied DID methodology which means difference in differences that helps estimating the causal effect of advertising intervention.
The main objective is to determine whether adding ads would discourage viewers or increase their engagement.
Development:
Live streaming which is considered a type of user generated content has shown rapid growth becoming a crucial digital format for marketing and entertainment. (Fader and Winer 2012, Lamberton and Stephen 2016).
Some platforms allow monetization from live streaming through donations, subscriptions and even advertisement.
Streamers can display ads through sponsorship support that requires a verification status (e.g;Grayson 2022).
A streamer’s channel can be partnered to possess the ability to display ads
Midroll Advertisement (MRA) is a video format ad that is usually displayed for around 5 to 30 seconds. They are relatively invasive. Other sources of money for streamers include sponsorships and donations (D’Anastasio 2022; Grayson 2021:Lu et al. 2021).
Audiences can watch and interact through text and donate money to the streamer through subscription or direct donations.
There exist 6 sociomotivational reasons for live streaming consumption (Hilvert-bruce et al. 2018).
-social interaction
-sense of community
-meeting new people
-entertainment etc.
However these ads can interrupt user experience and generate negative reactions (Edward,Li,& Lee, 2002).
Intrusive ads like MRA can increase avoidance (goodrich, 2011).
Research Questions:
How might MRA increase or decrease consumption?
Do streamers strategically adjust their behavior after receiving ad permissions?
Are effects heterogeneous depending on streamer size or content type?
Methodology:
The authors of this article used Panel data from a live streaming platform for 84 consecutive days and applied the DID model (Angrist & Pischke, 2009) to estimate the causal impact of midroll advertising permissions.
It defines two groups :
1-Treatment group : the streamers who receive advertising permission.
2- control group : streamers who did not receive advertising permission.
To compare the outcomes before and after the intervention?
They then conducted parallel trend tests , event study analysis and robustness checks. To strengthen the causality validity.
Main findings:
There is evidence of positive effects of MRA abilities on live streaming consumption behaviour even though it shows that ads are usually not desired.
MRA significantly increases average viewership and hours watched and duality for partnered channels.
Conclusion:
The study finds that advertising does not reduce consumption as predicted by ad irritation theory (Edwards et al., 2002). Instead, the revenue incentive encourages streamers to improve content quality.
The effects are stronger for smaller streamers and social content channels.
This aligns with research showing that financial incentives influence content strategies (Chen & Xie, 2008).
Limitation :
The secondary data does not include streamers usage of MRAs.
The study does not include nuances such as targeting and or time of the day and their influence on the results.