X Puts $1.05M into Long-Form Content Contest, Winners Revealed

Original Author: David, Deep Tide TechFlow

In mid-January, X announced a 1 million dollar reward for the best long-form Article on the platform.

Elon Musk personally retweeted to confirm. The rules are also very simple: limited to U.S. users only, original English articles over 1,000 words, mainly ranked by exposure to U.S. paying users.

You should still remember that a few days before this content campaign was released, personal development blogger Dan Koe posted "How to fix your entire life in 1 day," which received 170 million impressions and became the best-performing Article in X's history.

X obviously saw the traffic potential of long articles and quickly followed up; they lowered the threshold for the Articles feature, adjusted the algorithm weights to prioritize long articles over short posts, and announced a million-dollar writing contest prize.

Tens of thousands of participants in a two-week competition period.

The results were announced on February 4, with a total prize money of 2.15 million US dollars, more than doubling the promised amount. The champion received 1 million, the runner-up 500,000, and there was also a 250,000 "Creator Choice" award and four 100,000 honor nominations.

The award-winning situation is roughly as follows:

You can see Dan Koe made the list again. However, his previous article about how to fix your life in one day had 170 million impressions, but this time the champion of the writing contest only had 45 million.

Blockbusters are still rare, but several award-winning articles are also worth analyzing.

The champion @beaverd's article title translates to "Deloitte, a 74 billion dollar toxic tumor spreading across the United States." It is about the well-known consulting company Deloitte.

This account currently has "only" 90,000 followers, and compared to the other award-winning individuals, it's considered a small account. Moreover, it has no media organization backing and no endorsements beyond the blue V verification.

The topics he wrote about don't touch any popular traffic keywords, but the issues he exposed are quite controversial, that is, how Deloitte obtained 74 billion dollars in contracts from the federal and state governments, and then messed up the projects.

PortalHere

Click in to see, and you'll find that this person really put in the effort.

He built his own website called somaliscan.com, collected millions of government invoice data, and cross-checked them one by one with audit reports and system failure records.

Then, using these first-hand data, it told a series of shocking stories: 32 billion dollars in California unemployment benefits were fraudulently taken, the Tennessee Medicaid system collapsed, causing 250,000 children to lose coverage, and a court informatization project wasted 1.9 billion dollars and ended in failure... In total, 25 states were covered.

He also uncovered the revolving door between Deloitte executives and government officials, specifically listing who jumped from Deloitte to which department, and which contracts were then awarded back, with names and amounts clearly listed.

A person created their own database and self-researched to earn $1 million.

The runner-up @KobeissiLetter is a familiar face in the macroeconomic and financial circles, with 700,000 followers, and has been following U.S. economic policies and market fluctuations for a long time.

What this article does is also straightforward: it breaks down Trump's repeated tariff tactics into a repeatable trading framework, titledTrump's Tariff Playbook: An Operational Guide.

Because Trump often acts unpredictably, likes to announce outlandish policies and threaten other countries, but doesn't necessarily fully follow through in the end, some people on Wall Street summarized this pattern as TACO, which stands for Trump Always Chickens Out (Trump will always back down).

TACO refers to a recurring pattern:

Trump announces harsh tariffs → market plummets → a few days later he backs off or delays → market rebounds.

Portal

What the article "KobeissiLetter" does is turn TACO from a joke into an operation manual with a timeline. He uses the tariff events of the past 12 months as a sample, and breaks out a complete cycle template for you to follow for trading according to the time periods.

For example, on the weekend, the White House side releases news to create panic, midweek bottom-fishing funds enter the market, the next weekend a signal of easing is released, and some kind of agreement is reached within 2 to 4 weeks. At the same time, he will continuously post updates at every step, telling you where it is now, more like a continuous drama of a preliminary research post.

He also provided a more practical method, which is to watch the yield on 10-year U.S. Treasury bonds. If this number breaks through 4.60%, Trump is likely to give in.

For X's paying users who focus on macro and trading, this kind of thing is right up their alley.

It doesn't discuss with you whether the tariff is good or bad, nor does it make moral judgments. It just tells you that next time when you come up with this plan again, what actions you should take at what time points in order to make money.

Dan Koe's contest entry "How to Enter an Extreme State of Focus Anytime" received 42,000 likes and 8,681 reposts, both figures being the highest among all contest entries. However, its exposure was only 11.04 million, less than a quarter of the champion's.

What X was given is strictly speaking not a third place, but a separate "Creator Choice" (official selection) award, worth 250,000 US dollars.

Actually, it's understandable. Dan Koe is the "person who inspired this match." His viral article in early January with 170 million exposures directly showed X how high the traffic ceiling for long-form articles could be.

Portal

I won't go into too much detail about the article itself, it's still the same old set of life growth methodologies. It's roughly about how to gain focus, and uses concepts like neuroscience and flow state to support and elaborate on the ideas.

Actually, this one has the best interactive data, but according to the core rule of the competition, "exposure to paying users in the US," it can't rank high.

Why do the articles with the best interaction have relatively low exposure? I will discuss this misalignment later.

Nick Shirley, Josh Wolfe, Kaizen Asiedu, and Ryan Hall each received 100,000 in incentives. Their accounts cover four areas: public policy, geopolitics, history, and public safety.

Among them, Josh Wolfe, co-founder of Lux Capital and a well-known venture capitalist, also announced that he would donate his prize money equally to four charitable organizations.

Since the original post did not list the specific articles of these four people, and due to limited time and energy, we did not conduct any further investigation. We also welcome everyone to supplement the information.

Some patterns that can be seen from the results of this competition are:

The article with the most likes has only a quarter of the exposure of the champion.

The most counterintuitive data from this match is definitely Dan Koe's.

42,000 likes, 8,681 forwards, 4,627 comments, the highest three interactive data in the entire event. But the exposure is only 11.04 million, less than a quarter of the champion @beaverd. And @beaverd's likes are 30,000, even less than Dan Koe.

If you have done social media operations, you would find this set of data awkward. According to general understanding, the higher the interaction, the more the algorithm is willing to promote, so the exposure should be greater.

But this time, X is not calculating total exposure; it's the "number of exposures on the home timeline for paid users in the U.S." This metric excludes all non-U.S. users, non-paid users, and visits from search and personal homepages.

Dan Koe writes about personal growth, which naturally has a more global audience, and has a large number of non-American fans. @beaverd writes about how Deloitte wastes American taxpayers' money, which naturally focuses its audience in the United States. Under the same algorithm recommendation mechanism, the "geographic concentration" of the content determines the level of this metric.

90,000 fans beat 900,000 fans, content scarcity > fan base size

Champion @beaverd had 90,000 followers before the competition. Runner-up @KobeissiLetter had 700,000 followers. Dan Koe had 900,000 followers.

If follower count could determine exposure, the rankings should be reversed. But the actual results indicate that in X's Articles recommendation logic, the weight of the follower base is far less significant than imagined.

@beaverd Whether he can win depends on whether he has something others don't have, or whether the scarcity of the content plays a role.

This is completely different from the traditional traffic logic. Big accounts rely on their existing fan base and posting frequency, but in an environment where distribution is dominated by algorithms, "whether you have exclusive content" is more important than "how many fans you have."

You have to build your own content "hardware"

Stepping back to look, the themes of these three award-winning articles are completely unrelated: one exposes government contracts, one teaches you how to trade tariff fluctuations, and one talks about how to concentrate.

Put into any content platform's categorization system, they wouldn't appear on the same list. But they have one thing in common: each has its own independent "hardware," in other words, you need a narrative framework.

@beaverd's hardware is a self-built database that scrapes government data; KobeissiLetter's hardware is a trading framework backtested over 12 months, and Dan Koe's hardware is a six-chapter methodology integrating neuroscience and psychology, which may seem profound but actually consists of common knowledge.

The award-winning ones are not purely opinion articles.They both require lengthy content to carry the information, which is exactly the reason the X Articles product form exists.

Another noteworthy fact is that,None of the eight winners were traditional media outlets.

All are independent creators. It's not that traditional media didn't participate, but under this competition format, individual accounts actually have an advantage.

The content of institutional media is usually posted on their own websites, with only a link and summary placed on social media. However, Articles requires the full content to be posted within the X platform, which is an awkward move for media accustomed to driving traffic outside the platform.

Back to the platform itself.

X initially promised a $1 million incentive, but finally issued 2.15 million. During the competition, a series of supporting actions were also taken: expanding the Articles feature from creator accounts to all paid users, adjusting the algorithm to increase the recommendation weight of long-form content, and changing the scoring method to "homepage exposure for paid users in the US."

It costs so much, and the most direct thing is definitely that X needs original long articles on the site.

In the past, long-form content on X mainly relied on external links, such as Substack, Medium, and personal blogs. When users clicked on them, they would leave the platform, and their reading time and interaction data would remain on other platforms. The goal of Articles is to keep this content within the platform, allowing users to read from beginning to end without leaving X.

Going even deeper, X has Grok. Training large language models requires high-quality long-form text data, but most content on X is short tweets of 280 characters. If Articles can consistently attract creators to produce in-depth long articles, this content would serve as training material for Grok.

Finally, paid user value.

The competition rules limit the metrics to "exposure on the home page for paid users in the US," which is equivalent to directly telling creators that your content should serve paid users.

This is using the creators' content to support the paid system, making paying users feel that "my money is worth it, because I can see in-depth content that I can't find elsewhere on the homepage."

From the perspective of content creators, we feel thatThe era of pure opinions may be coming to an end.

This trend also applies to creators in the crypto space. The crypto industry is not short of opinions; every day, countless people on X call for trades, predict prices, and comment on regulations.

But few can build their own on-chain data analysis tools like @beaverd, or break down market cycles into repeatable trading scripts like KobeissiLetter.

Maintaining scarcity and independence while consistently producing work is actually a very professional endeavor, and it's also a job with a great sense of achievement and positive feedback.

We also hope to see more content from the Chinese-speaking community that can appear on the charts in the future.