The hunt for unique advantages
Networks, data, podcasts, and legacy media - an outlook on the forces that are set to shape 2025 and dominate the headlines in entertainment, tech, and business.
Dear Readers,
Happy New Year and welcome back! I hope you had a wonderful time off and are ready for what is already shaping up to be an eventful, and potentially crazy, new year.
The world’s second largest gaming company Tencent being designated a military company by the US military?
Mark Zuckerberg dressing like a Kendrick Lamar fanboy confusing content moderation with censorship and essentially taking off all the guardrails on social media that will make the cesspool even more hazardous?
It’s only January 9.
356 days to go.
What could possibly go wrong…
There’s not a single newsletter or blog out there that hasn’t already released or is writing about their predictions for 2025. Technically Entertaining wouldn’t be doing right by you if we didn’t equip you with our view on the next 12 months. To be fair, we didn’t have Tencent being labelled as a military company on our bingo card. But predictions aren’t about being right. They’re about inciting a meaningful conversation by offering a unique perspective. Aside from that, the downside of getting it wrong is, well, almost nonexistent.
As the late Nobel Prize winner Daniel Kahneman said: “Most successful pundits are selected for being opinionated, because it's interesting, and the penalties for incorrect predictions are negligible. You can make predictions, and a year later people won't remember them.”
With that, here are the forces set to shape technology, entertainment, and business in 2025 (and beyond) - let’s see if we’ll be right about these. If not, let’s hope Kahneman is right and we’ll all just forget about it ;)
The battle for network efficiency
The marketing and advertising environment to reach consumers will be increasingly challenging due to strengthened privacy protections and the full body combat between the different channels and platforms to increase their share of advertising dollars spent will be increasingly violent. It all comes down to this: In a near cookieless world, who can target the best and offer brands the most effective and efficient access to the best possible consumers?
In order for online advertising to maintain and increase its scale, advertisers need to make their money back and be able to grow their services (e.g. their games, apps, you name it) efficiently. Part of the onus here is on the advertiser himself to, simply put, make better ads that consumers actually care about. The remaining responsibility lies with the ad network itself to become smarter and understand which ad should be shown to which person, when (resonance trumps relevance).
AppLovin has been on a bender in 2024 (its stock grew by almost 800%) and rumor has it that it is mulling over multiple acquisitions this year. Moloco is a fan favorite and appears to be growing nicely on the back of what industry insiders are calling superior machine learning algorithms to facilitate the ad to user matching.
Chartboost looks like it will be spun out this year by Zynga’s parent company Take-Two and either be taken over by a PE firm or acquired by another ad network. Look for Roblox to potentially make a move here to accelerate the development of its own advertising platform.
TikTok has been demonstrating great results for advertisers, but what happens if the app gets banned on January 19?
Meta is a force simply due to its sheer size and with Threads crossing 270m users and set to overtake X (the platform fka Twitter), it’s giving advertisers new and more reasons to spend with them.
What do all of these have in common? They need to find, and internalize, new and unique data sources that hold novel signals their machine learning algorithms can leverage to increase the efficiency in their networks. Ideally these are data sets that let the networks understand and identify a consumer more precisely without needing to understand their real life identity (privacy regulation won’t go away).
The network that can give advertisers the most curated reach to high quality users, at scale, at the best prices, wins. That’s why network efficiency is critical. It’s all in the margins.
AI needs different data to overcome its plateau
With all the hype and promise around AI, the profound advancements of the last 24 months are taking a little bit of a breather. There are three core drivers, and limitations, to the continued progress of AI:
1. Energy consumption (it’s why nuclear energy is suddenly en vogue, with companies like Meta releasing RFPs to find a partner to build Meta its own nuclear power plant).
2. Computing power (the chip race, and who can make better chips, faster, cheaper, and get them deployed at scale. This is Nvidia).
3. Data (why pretty much all AI companies signed licensing deals to pay millions of dollars to scrape the content of Reddit, the Wall Street Journal, and many more).
The first two are constraints to AI’s scale. The third one is a constraint on its intelligence, and hence the value it can deliver to users and organizations.
After a flurry of licensing deals from the likes of OpenAI or Anthropic (see above), the search is on for unique and proprietary first party data sets they can feed into their models. Only these unique data sources can give an AI system differentiated intelligence and a lasting advantage over its competitors. Adding the next online newspaper won’t do the trick because it’s essentially more of the same type of data they have ingested already, and the AI systems all are essentially trained on the same type of information.
Until now, most Fortune 1000 CEOs and CFOs are still a bit disillusioned by the impact of AI on its organizations. The promise is there. The tangible results are not. Which is why most AI efforts within organizations remain focused on delivering efficiencies (read: unlocking cost savings) rather than generating revenue. At least for now. 2025 will be the year where AI companies make a strong push to find and secure novel data sources to put them ahead of the pack and deliver more value.
We’re entering the age of podcasting
The recent US elections have shown one thing very clearly: traditional media is on the way out. New media formats - podcasting - are in. When over $1 billion in campaign funds lose to $600 million, it sends a clear message.
Podcasts were already popular before the election, but their influence in the political sphere as well as their ability to give advertisers access to highly curated audiences makes them incredibly attractive. The average age of a podcast listener is 34 years. The average age of a cable news TV viewer is 70. Who’s the more attractive consumer for brands? Your call.
In 2025, watch for new streams of money to flow into podcasting, to advertise as well as develop new content formats. As a part of this, video plays a huge role. And it’s not audio or video podcasting - it’s both. Joe Rogan’s interview with president-elect Trump generated 15 million downloads on Spotify and Co. It generated over 40 million views on YouTube. This is also why Spotify, again, is doubling down on its podcasting business, including video podcasting.
This sets up an interesting battle between the Swedish streaming giants, YouTube, and Netflix as it pushes more and more into live events and broadcasting. We should also keep an eye on SiriusXM, which bought Spotify’s second biggest podcast of 2023 “Call Her Daddy” for $100 million last year and will begin the exclusive distribution of the podcast this year.
The ongoing consolidation of legacy media
This an easy one, because Comcast paved the way in 2024 already and Warner Bros. Discovery restructured its internal operations to do the same - shed its growing businesses from the declining legacy business that is cable TV networks. Keep an eye on Disney as it could spin off its cable TV assets (e.g. ABC, ESPN, National Geographic) and look for a big private equity form to come in and consolidate the different assets and reap the rewards of a high cash flow generating business (MSNBC and CNN could have the one, instead of two separate, newsrooms - margins go up immediately).
We’ll take stock of how accurate we were at the end of this year. So if you haven’t already, please consider subscribing to be sure you don’t miss out on our scorecard.
Until then - 356 days to go. 2025 will be a wild one.