Too soon to measure Generative AI ROI for M&E?

Media and Entertainment companies are launching Generative AI initiatives and some are even showing positive ROI within the first year, according to a Google survey on Generative AI in M&E done with National Research Group.

The report found 64% of the 263 M&E respondents (representing large media and entertainment companies with over 100 full-time employees and revenue of at least $10 million) have Generative AI use cases in production, 24% are still evaluating use cases and 12% have not yet started evaluation.

These typical use cases include customer service, digital commerce and enhanced experiences, individual productivity, sales and marketing, back-office processes, engineering productivity, new products and production processes. Most GenAI projects were launched in a 3-6-month period (47%), and 83% of M&E companies can go from a GenAI idea to implementation within six months, while 17% reported taking longer than 6 months.

Google GenAI M&E use cases graph
M&E organizations reporting genAI use cases in production. (Google ROI of GenAI in M&E survey)

There's a famous book called Crossing the Chasm written by Geoffrey A. Moore about how new technology advances. There are a number of distinct stages; stage one is innovators, then adopters, early majority, late majority and finally laggards. GenAI for M&E may still only be at stage one, maybe at the most stage two, but Google says those early movers are seeing benefits. 

"Our global survey reveals a media and entertainment landscape where early adopters are already reaping significant rewards," wrote Albert Lai, global director of Media & Entertainment at Google.

What does this look like for M&E?

The report states: "Well-resourced studios and their commitment to innovation are likely at the forefront of this trend, while others with a focus on stability may be adopting these changes at a more measured pace. However, the pressure to innovate is only going to intensify. The need to engage consumers (personalization), the rising cost of content, and the need to evolve new monetization models are compelling media and entertainment organizations to embrace gen AI."

Adoption of GenAI so far has been lower in production workflows compared to customer service or marketing because integrating this may require greater technical investment, a higher tolerance for disruption, and more planning, commitment and budget, per the report. Responsibility for driving GenAI strategy development is evolving with 72% saying the objective is shared, those driving strategy at the senior level range from CTO (65%), CEO (49%), CIO (46%), CFO (31%), CSO (26%), line of business management (24%) to CMO (22%).

Positive ROI?

Reached by StreamTV Insider, Google cited confidentiality and would not provide details on any actual GenAI projects, aside from two brief endorsements included below. 

“As one of the leading global media brands, our highest priority is to always provide an informative and engaging experience for our audience,” said Burhan Hamid, CTO of publisher TIME. “Generative AI has allowed us to create new experiences for our readers, making it easier for them to find the content they want to consume.”

Another comment: "Gen AI is not just a technological advancement; it's a strategic imperative for us. We're actively exploring and implementing gen AI use cases across various areas of our business, from marketing and ticketing to fan engagement and player performance analysis," Daniel Brusilovsky, Vice President, Technology, Golden State Warriors. "We're confident that gen AI will play a crucial role in driving our future success and shaping the future of sports entertainment."

Of respondents,106 M&E organizations currently utilizing genAI reported a meaningful impact on business growth from the technology. Of those using GenAI and reporting business growth attributed to the technology, 66% reported experiencing increased conversion rates, 65% created new products and services, 63% increased revenue and 63% improved leads and new customer acquisition.

And of the 169 M&E respondents currently using genAI, 72% are seeing ROI on at least one use case. According to the survey, 30% of M&E companies using GenAI and leveraging or planning to use the tech for the digital commerce use case reported seeing ROI for that application now and another 34% expect to see ROI in the next year. GenAI for new products and services initiatives is delivering ROI now for 24% of M&E companies using or planning to implement the tech for that use case, while 36% expect ROI next year.

Here’s the timeline on which M&E companies leveraging genAI are seeing or expect to see ROI from the technology across various use cases where it's currently or planned to be used.

Google GenAI ROI M&E graph
ROI timeline for genAI use cases by those leveraging or planning to use the technology by specific use case.  (Google ROI of GenAI in M&E survey)

Its always the output which matters in Generative AI. When it comes to how businesses' plan to use the gains from their Generative AI initiatives, 43% said they want to identify new revenue streams, 43% will enhance competitive differentiation and 41% to improve operating profit margins.

The Pace of Media

It is because of the newness of everything that companies are experimenting but may be holding out for some time to be able to do projects at scale, said Simon Solotko, senior analyst at Tirias Research.

Tirias’ research areas include Generative AI, 5G and wireless communications, AR/VR/XR, artificial intelligence (AI) and machine learning (ML), autonomous vehicles, cloud and networking, electronics systems, gaming, PCs, semiconductors, and smartphones and other mobile devices.

"The weakness today is that while AI can be employed to help at various steps of content creation, an expert is still required to check quality and course correct at each step, and manage the handoff between steps,” Solokto said. “Unlocking productivity will require the automation of these workflows... [like] pre-visualization, VFX, scene construction, storyline development, and marketing content."

Richard Sussman, an early-stage venture capitalist noted a slow start but cited expectations for quicker adoption ahead. 

"I think industry as a whole, as they leave the research phase of the process and going into deployments, is moving very slow. Employees have to make decisions that could impact businesses in a huge way. I think people just don't know what they don't know and they're reluctant, but then you do have outliers that are moving at warp speed," said Sussman, former co-founder of Rebel Fund and Nordic Eye VC. "I think you're going to see a faster adoption rate. I expect that to happen in probably the next six to maybe 18 months."

Solotko, meanwhile, pointed to experimentation.

"For the media industry, Generative AI is the change maker in technology. There is low hanging fruit - AI for search, customer engagement, data analysis, and content creation," said Solotko."The temptation is to look for fast ROI, however I would suggest letting your teams experiment and explore. [Then] invest in the longer term, bigger payoff opportunities - reinventing workflows and media itself."

While this reinvention is coming at a speed no one has experienced before the media industry is especially impacted by what Generative AI can do to their IP, using content these companies have heavily invested to siphon off their revenue. 

"I think overall media and entertainment need to move quicker because they're one of the industries that has the most to lose with all their assets," said Sussman.

While DRM (digital rights management) was an easy choice for studios and rights holders to get behind to protect their copyrighted works from piracy, the possibility of what Generative AI is capable of can be horrifying to creative professionals. Who wants to see a movie recut, or worse, have their creative property used to train an AI model to create similar content?  Since much of M&E’s value is based on creatives’ content, infringement on copyright is a direct stab to the heart of their businesses. 

"When you're building out your M and E practice you're trying to not only protect your assets, which obviously is a whole other conversation, but also making sure that the assets are being used in the right way,” Sussman continued. 

Some of the ways genAI is having an initial impact on media and entertainment, according to the Google research, include algorithms for targeted content recommendations, streamlining production processes and optimizing ad strategies with AI-powered platforms that help targeting and personalization of ads.

And 64% of M&E organizations that use genAI currently are reporting productivity gains, with 34% of that group claiming employee productivity has at least doubled. 

Overall, this research sounds promising, but it’s still the early days of genAI initiatives. The research was based on interviews with 263 M&E leaders of organizations with at least $10 million in revenue. The M&E research is a subset of a larger study of multiple industry verticals and 2,843 business leaders.

Article updated with information on Richard Sussman's role.