Why did my "nonsense" on Threads go viral? Unveiling the "traffic dividend" and distribution logic of Threads and X.
Editor's note:
In the game of social media traffic, we may be entering an era that defies common sense.
Have you ever experienced something similar? On X (formerly Twitter), meticulously crafting a profound insight results in only a few hundred views; yet on Threads, casually posting "I'm so tired today" or sharing an Instagram photo brings likes and comments flooding in.
Is this phenomenon—ordinary content receiving absurdly high visibility—pure luck, or is there a hidden social code behind the algorithm?
Recently, through several "accidental" dissemination experiments, I reviewed the latest empirical papers from 2025, attempting to reconstruct the underlying logic of this distribution mechanism and revealing the power map hidden within this algorithmic black box.
Accidental "Cold Start":Why can a new account go viral overnight?
The story began with an unintentional action. I casually synchronized a post from Instagram to Threads. At the time, my Threads account was still a "new account," with almost no follower base beyond friends migrated from IG.
Surprisingly, this content—extremely ordinary in nature—received unexpectedly high exposure.Later, I tried leaving a "meme-style" comment under someone else's popular post, and surprisingly, this comment achieved "parasitic distribution," triggering an avalanche of likes and significantly boosting the visit count to my personal profile.
Why did my "newbie account" accidentally go viral? This was thanks to Meta's "nepotism.
In computer propagation science, this is known as "cross-platform weight inheritance." In other words, Threads and IG share underlying data—essentially brothers wearing the same pair of pants. When the algorithm checks your ID, it recognizes you as an established, trustworthy user on IG and identifies the "initial fingerprint" of your previously posted content.When you sync content, the platform grants this content an "initial credit," much like entering a beginner's village—Threads gives new users a "welcome gift package." Combined with the algorithm's inherent preference for visual imagery (Viewport), even starting with zero followers, you can still receive that first wave of traffic injection.
Debate with the Professor: Is the Algorithm a "Deterministic Pattern" or a "Probabilistic Lottery"?
When I shared these "patterns" with my boss—a professor in the field of science communication—he challenged me:
"I've also tried syncing content, but for identical posts, one got 3,000 views while another only received over 100."This doesn't look like a replicable pattern at all—it's more like a lottery.
To respond to this challenge, let's start from the perspective of data mining and break down several core metrics that are often misunderstood by the general public.
1. Metric Definition: How meaningful is the number you're focusing on?
In data analysis, the way we count determines our value judgment. We’re often easily impressed by numbers like tens of thousands of "impressions," but in reality, many such figures are merely "distraction tactics" used by social platforms to create an illusion of false prosperity. Sometimes, the vague term "impressions" alone fails to reflect the actual reach of a piece of content. Here, let’s clarify three key metrics: Impressions , Views, and Engagement .。
Impressions：Think of it as "foot traffic at a subway station." Any time a post appears in the user's scrolling path, even if they quickly scroll past (scrolled past), as long as the content loads, it counts as one impression. This is typically the part with the most data inflation。
Views：Similar to people who walk into a store and look at the menu. It usually requires a certain amount of dwell time from the user, or them actively clicking "see more."
Engagement：These are the ones who actually entered the store and "bought coffee."Not only includes clicking into posts to view them, but also subsequent actions such as liking, replying, and sharing or reposting. Compared to simply passing by and briefly looking at a menu, actually spending money to buy a cup of coffee clearly requires more time and effort. Engagement is also a core signal that algorithms use to assess content quality.
2. Horizontal Comparison: Why Do Data Perceptions Differ So Drastically Across Platforms?
Later, while using Threads, I really enjoyed opening the "Insights" page—it felt like my personal data analytics dashboard. There, I could check metrics from the past 90 days, including post views, interaction counts (such as likes, replies, and shares), and follower numbers.
Image source:Threads
But when using different social platforms, we notice that each platform defines the "content value" in slightly different ways, which directly determines their counting logic and distribution incentives.
YouTube（extremely conservative）：To protect advertisers' interests, YouTube defines a view extremely sparingly. A valid view typically requires users to actively click or watch for over 30 seconds. Under this mechanism, every frame of data is audited, giving it strong perceived value and high intrinsic worth.
X/Twitter（data inflation）：Musk brought the previously backend-hidden impression count (Impressions) forward and labeled it as "views," essentially creating an illusion of platform prosperity.This calculation method greatly amplifies the weight of users "quickly swiping," so even if you merely scroll through a post at 100 rpm, the number increases by 1, causing the metric to be significantly inflated.
Threads（velocity-first）： Its logic is nearly identical to X's, primarily emphasizing a post's "exposure" rather than actual engagement. It focuses on engagement velocity—the initial speed of interaction. During the current bonus phase, as soon as your content enters a user’s viewport, the algorithm quickly records it and seeks the next distribution opportunity, aiming to rapidly ignite social activity.
TikTok（silent counting）：Its For You Page (FYP) uses an auto-play mechanism with logic similar to X—viewing counts as engagement.As long as a video appears on the user's screen, the view count increases by one, even if the user swipes away without watching it. This low-threshold counting mechanism is designed to feed its powerful recommendation engine, using massive "passive samples" to filter out the rare viral hits that can stop people in their tracks.
3. Algorithmic Underlying Mechanism: Exploration Budget
The professor's mention of "the same content, different fates" leads us to the core of our discussion—the algorithm's random experimentation. Academia refers to this concept as the "Exploration Budget," driven by the "Diversity Explorer."When you publish a new piece of content, the system allocates a "discovery budget":
Random Sampling：The post is shown to a small group (e.g., 50–100 people) of randomly selected users.
Stress Testing：The system monitors real-time data from this group—whether they quickly swipe past (ineffective impressions) or pause to click "Read More" (effective views).
Amplification Effect：If these "seed users" respond positively, the system determines the content has "cross-boundary potential," triggering secondary or even tertiary cascading distribution.
The data variation mentioned by Professor essentially stems from "random sampling error." The same piece of content might sink quickly if the first batch of 50 random passersby are busy; however, if the second batch just happens to be highly interested, engagement is sparked, activating the algorithm’s large-scale distribution.This is not mysticism, but the brute-force application of probability theory in social networks.
Threads' Vibe：Why are "nonsense" posts more popular than "insightful" ones?
Why does the same text become "resonance" on Threads, while on X it's labeled as "noise"? This stems from algorithmic weighting based on "reply-ability."
X's logic: prioritizes information density and confrontational content. The barrier to posting is high, so users tend to "read without replying."
Threads' logic: emphasizes sociability and emotional value, with algorithms seeking "extremely low-barrier entry points."
A single phrase like "This coffee is really terrible" or "I'm so tired" is considered extremely valuable content by algorithms. Because it triggers countless people to reply with "Me too." Such "high-frequency, short-response" conversation chains are perceived by algorithms as high-intensity user engagement (Engagement). Just like the previously viral post about Hong Kong's "Da Bao Ice Room meat pie rice," which seemed absurdly repetitive at first glance, yet gained algorithmic favor due to its easy-to-join nature, becoming a trendy writing style that swept through Threads.
Empirical Exploration:The Truth Behind the "Bonus Period" Revealed by Data
To substantiate this intuitive sense with hard data, I reviewed the empirical paper published in《Journalism and Media》in March 2025 titled《Digital Political Communication in the European Parliament: A Comparative Analysis of Threads and X During the 2024 Elections》。
Image source: Digital Political Communication in the European Parliament: A Comparative Analysis of Threads and X During the 2024 Elections
Researchers compared the performance of official accounts from the European Parliament (EP) on two platforms, revealing highly intriguing differences (Velasco Molpeceres et al., 2025):
1.The "miracle" of participation:
Paper data shows that although the European Parliament (EP) official account on X has ten times more followers than its Threads counterpart, during the 2024 election period, Threads achieved an engagement rate of 0.09%, significantly higher than X's 0.05%. This confirms that Threads' algorithm is more inclined toward "inclusive" distribution of ordinary content.
Image source:Digital Political Communication in the European Parliament: A Comparative Analysis of Threads and X During the 2024 Elections
2.A complete "decoupling" of narrative form:
Threads is a paradise for thread-based content: in the research sample, 65.67% of posts on Threads were published in "thread" format, compared to only 7.69% on X. Threads users are more accustomed to reading content through continuous conversational structures.
Misalignment of visual resources: The original post ratio on Threads is 100%, and its usage rate of infographics (11.94%) is higher than that on X (8.65%).
3.Conversation bundling mechanism： The paper's discussion found that Threads bundles "high-quality replies + the original posts being replied to" and pushes them to strangers.This explains why your replies under trending topics can "parasitize" the trending traffic to break through boundaries — because the system treats your reply as an indispensable extension of that conversation block.
After reading the paper, we can vividly imagine the following two scenarios:
X (Twitter) is like a highly hierarchical plaza. Only those well-equipped and loud voices (blue-check verified, seasoned politicians, super celebrities) are heard by more people. In this plaza, information flows top-down, and ordinary people's voices easily get drowned out by the grand narratives of the elite class.
Threads is more like a decentralized bar. People gather around different round tables, and the algorithm no longer focuses solely on your title—it pays attention to who’s saying something that truly resonates. Even if you’re an unknown nobody, just tell a joke so sharp it compels others to respond, and the algorithm instantly expands your table, making you the social center of this moment.
This paper also reveals an interesting phenomenon: even official accounts with a serious nature, such as the European Parliament (EP), tend to put aside their elitist lecturing demeanor on Threads.
This strategic shift demonstrates that institutional accounts are actively leveraging Threads’ “phatic value.”By leveraging "threaded posts" and "continuous conversations," users are integrated into community life—transforming the experience from one-way broadcasting into more resilient, interactive dialogues. This approach helps accumulate warmer social capital (Social Capital) for political propaganda, ultimately converting this "algorithmic resonance" into voters' democratic identification and increased electoral participation.
Bonus from "Interest Circles":Decoding the Morse Code of "Shared Concerns"
The final traffic key lies in "Interest Circles"—a system based on interest graphs.
The post I unintentionally shared on Threads was a summary after attending a concert. The text mentioned keywords such as my favorite singer, "Hong Kong Coliseum," and "Cantopop," accompanied by a photo I took of the live performance venue.Such visually and textually rich posts trigger the "community effect" associated with keyword matching. Within the subcultural circles, content doesn't need profound insights—just a pun only insiders (same-fan) understand, a string of emojis or memes (meme images) representing a specific inside joke, or even a seemingly empty emotional musing can instantly spark engagement from fellow insiders (same-fan).
Since the algorithm determines you belong to a certain topic cluster, it prioritizes showing your post to those who “get it.”
In summary,
Ultimately, Threads’ algorithm isn’t searching for “greatness,” but for “conversation.”
It breaks the long-standing hierarchical structure on social media, dominated by blue-checks or individual influence, allowing every ordinary person’s fragmented emotions to have the chance to become a center of resonance.When posting next time, try letting go of your burdens and share a simple "I'm so tired from work today." You might discover that in this algorithm-driven world, nonsense is humanity's last kindness.
Author:
Zou Yanping
Sentiment Insight Researcher, BoYi Data
References: [1] Velasco Molpeceres, A., Miranda-Galbe, J., & Prieto Muñiz, M. (2025). Digital Political Communication in the European Parliament: A Comparative Analysis of Threads and X During the 2024 Elections. Journalism and Media, 6(1), 42. https://doi.org/10.3390/journalmedia6010042 [2] Chi Wei Lin. (2025). Threads 經營攻略解析：如何利用同好圈提升互動？. 數位時代. https://www.bnext.com.tw/article/81784/threads-engage-strategy [3] Rem Chiu. (2025). 2025年最新香港品牌Threads行銷案例：解構如何打造高互動率貼文. More Digital. https://moredigital.com.hk/2025/02/18/threads-marketing-cases-hk/ [4] Akira. (2025). 港品牌藉Threads吸客 打造「貼地」形象. 大公文匯. https://www.tkww.hk/a/202512/17/AP69425f2ee4b032040a15615a.html