Summary of Key Points
The growth of AI products going global is shifting from being "traffic-driven" to being "value and mindset-driven." The traditional methods of rapidly acquiring new users through standardized channels (such as SEO and KOL reviews) are becoming less effective due to the soaring cost of customer acquisition, severe product homogenization, and low user retention and willingness to pay. The panelists unanimously agreed that the prerequisite for growth is to first validate the Product-Market Fit (PMF), focusing on solving real needs in specific vertical scenarios; secondly, it's essential to establish a clear brand identity that helps users understand "who you are and what you can do for them"; and finally, it's necessary to distinguish between "value functions" (that retain users) and "growth functions" (that attract traffic), while exploring unexploited, non-standardized channels (such as niche communities and targeted marketing).
1. The increase in the cost of traffic is just a symptom; the real issue is that products don't clearly communicate what problems they can solve
Many AI teams are complaining about rising customer acquisition costs—SEO links have increased from $200 to $1000, and the prices of YouTube review videos have multiplied several times, with larger companies occupying high-value keywords. However, this is just a surface issue: the real problem is product homogenization. Users don't see the difference between your product and others, and even if they are attracted, they won't stay. For example, some AI tools have numerous features, but users don't know how to use them; or they are touted as all-powerful, yet their performance is unreliable in practice. Traffic can bring people in, but if you can't retain them, even the cheapest traffic is wasted.
2. PMF is the key to growth; don't rush into spending on advertising
The panelists repeatedly emphasized that without a clear PMF (a product that meets user needs), no amount of sophisticated operations will be effective. A common mistake early-stage teams make is to spend money on advertising and seek out KOLs without first understanding whether users really need their product. For instance, Cheng Mengqi's team realized that users couldn't articulate the value of their product three weeks after its launch, so they paused growth and used Reddit to gather genuine feedback—users directly told them that they needed a certain feature for a specific use case. The correct order is to first validate the product's value through user interviews and data analysis, and then use operations to drive growth.
3. User mindset is more important than features: make users remember your role in their lives
When AI products have similar capabilities, why should users choose yours? The answer lies in their "mindset"—they should think of you first when a particular need arises. For example, a brand that initially targeted business elites later realized that mothers were using it for breakfast preparation for their children and needed a sense of peace that their children would arrive at school without hunger. By adjusting its marketing strategy, the brand resonated with users. AI products should do the same: focus on a specific vertical niche (such as SoloEnt.ai, which specializes in long-form novel writing) and make users think of you when they need to write novels.
4. Look for untapped growth opportunities in non-standardized channels
Channels that everyone is competing for (such as top KOLs and high-traffic keywords) are expensive. Small teams should look for areas that others have overlooked. For example:
- Niche communities: Real user discussions on platforms like Reddit may not generate much traffic, but they provide valuable feedback.
- Targeted marketing: During Meta's layoffs, a company rented trucks near its headquarters to promote AI entrepreneurship tools to the unemployed.
- Segmented content strategies: Inviting LGBTQ+ creators to review Labubu and using their style to highlight the product's uniqueness. These approaches require patience and industry understanding; they can't be replicated by simply throwing money at them.
5. Don't mistake short-term popularity for long-term growth: distinguish between value functions and growth functions
Many teams get excited when a feature becomes popular (e.g., receives millions of views), but this might just be a "growth function" that attracts traffic, not a "value function" that retains users. For example:
- Value functions: Solve user problems consistently (e.g., AI writing tools that produce high-quality novels).
- Growth functions: Create engaging content or leverage trends (e.g., AI-generated funny images) to generate buzz, but they may not retain users. Teams need to evaluate these functions separately: don't neglect value functions just because growth functions are popular, and don't focus solely on them without traffic. The key is to ensure that traffic serves a purpose—use growth functions to bring in users and then use value functions to keep them.
In conclusion
The essence of AI product growth has shifted from "grabbing traffic" to "convincing users that you can solve their problems." First, make your product right; then target the right audience and tell the right story, and growth will follow naturally.