What Drives ByteDance’s Success? A Deep Dive into Its Product Matrix

People love perfect stories, but reality is often imperfect; optimism and pessimism are amplified, and labeling oversimplifies truth. In the internet world, ByteDance is praised almost hyperbolically, seen as unstoppable, with media exaggerating its speed, algorithms, strategies, and expansion.

ByteDance’s success is evident in several facts:

Products: Today’s Headlines and Douyin dominate their respective markets.

International expansion: TikTok is the most successful Chinese internet product abroad.

Revenue: 2019 revenue estimated at 140 billion RMB, making it one of China’s highest‑earning internet companies.

Usage time: ByteDance accounts for 15.4% of China’s mobile internet usage, second only to Tencent.

To understand this success, the article examines ByteDance’s products. The official “Our Products” page lists eleven apps: Today’s Headlines, Douyin, Douyin Volcano, Xigua Video, Dongchedi, Gogokids, Pipi Shrimp, Feishu, and Tomato Free Novel, with FaceU and Light Camera mentioned but without icons.

The products are classified along two dimensions:

Internal attribute: "Locomotive" apps (e.g., Today’s Headlines, Douyin) that drive other apps, and "Carriage" apps that rely on the locomotives for growth.

External attribute: Apps that define industry rules versus those that compete on the same level.

Mapping these dimensions creates a four‑quadrant matrix (see image below).

Most of ByteDance’s products fall into the first and third quadrants. The first quadrant contains the “locomotive” apps, which have high commercialisation, dominate revenue, and empower the third‑quadrant “carriage” apps. The carriage apps are generally cost centres with lower commercialisation and often do not yet generate profit.

ByteDance’s strong commercialisation enables rapid large‑scale experimentation. Its advertising is highly efficient, driven by algorithmic relevance (eCPM). Effect ads, powered by sophisticated recommendation algorithms, achieve higher user tolerance than traditional banner ads.

ByteDance’s algorithm excels with “fast‑decaying” content—low‑cost, short‑lifecycle media such as news, jokes, short videos—because large supply and fragmented consumption generate abundant feedback data. High‑budget, high‑granularity content like movies receives less algorithmic benefit.

The company’s recent expansion includes:

Education: Significant investment in K‑12 and early‑child platforms (e.g., Gogokids, DaLi Classroom).

Gaming: Comprehensive coverage from casual to large‑scale games.

Connected cars: A dedicated team developing in‑car infotainment solutions.

Online literature: Tomato Novel and related acquisitions.

Online office: Lark/Feishu competing with DingTalk and WeChat Work.

Online healthcare: “Xiao He” brand entering the medical app market.

However, the synergy between these new ventures and ByteDance’s core strengths is decreasing. Unlike early expansions from news to video, newer domains (education, literature, car‑internet, healthcare) have weaker ties to the recommendation engine. Moreover, the expansion remains primarily consumer‑facing (C‑side), with limited B‑side impact.

Growth is uneven: Tomato Novel quickly captured a top position in free novels, while Feishu still trails behind DingTalk and WeChat Work. Some products like WuKong Q&A and Micro‑Headlines have faded despite heavy investment, illustrating that mismatched strategies lead to decline.

In summary, ByteDance should be viewed as an incubator that provides talent, technology, traffic, resources, and capital. Entrepreneurs should avoid idolising the company and instead focus on rational, industry‑specific paths.