On August 10th, the Indian media Analytics India Magazine published a sensational “prediction”: OpenAI may be forced to file for bankruptcy by the end of 2024.
Meanwhile, according to SimilarWeb’s statistics, ChatGPT’s global website traffic in June 2023 decreased by 9.7% month-over-month, and the average time users spent on the website also dropped by 8.5%. When it came to July, the downward trend of ChatGPT’s daily active users still failed to be alleviated, with the website’s user base decreasing by 12% from 1.7 billion users to 1.5 billion.
According to the Ad Intelligence of SocialPeta, although 【ChatGPT】 had already launched on the App Store in May 2023, it still adopted a relatively cautious and conservative promotion strategy in terms of advertising. The data shows that 【ChatGPT】 has only placed 98 creatives in total on iOS in the past 90 days.
It’s not hard to see that the major paid AI models on the market today are facing declining user engagement, attention and revenue across the board. In contrast, the app market has seen a surge of hit AI apps, forming a stark comparison between the stagnation of B2B and prosperity of B2C.
For this reason, the SocialPeta Research Institute will conduct case studies on AI apps based on their functional features, use cases, monetization challenges, and promotional creatives.
An Overview of Practical Applications of AI Mobile Apps
Based on the output types of AI models, they can be divided into 4 categories: text, image, audio and video. This corresponds to specific tools as follows:
[ChatGPT], which specializes in text output
[WOMBO Dream], which specializes in image output
[Otter], which specializes in audio output
[Typpo], which specializes in video output
Based on the application platforms of AI mobile apps, they can be divided into 3 categories: Discord community, web version and app version.
Based on the use cases of AI mobile apps, they can be divided into 8 categories: search, image/video processing, social/chat, education, office, audio, health, and general.
- AI + Search Engine and Information Retrieval:【 Microsoft Edge】,【 Bing: Chat with Al】
- AI technology is used to improve search engines and information retrieval, providing more accurate and personalized search results.
- AI + Image and Video Processing:【 WOMBO Dream】, 【RunwayML】
- Includes AI + image editing, AI + video editing, and AI + image generation, involving processing, editing, and generating images and videos.
- AI + Social/Chat: 【Chat with Ask Al】,【Al Chatbot】
- Includes AI chatbots, AI+writing, and AI+audio transcription, utilizing NLP technology to enable automated conversations and text generation.
- AI + Education: 【Carneige Learning】,【Blackboard】
- AI applied in education, including personalized learning, intelligent tutoring, and education content generation.
- AI + Office and Business: 【Notion】,【 Taskade】
- Covers AI applications in office automation, business analytics, and decision support.
- AI + Music and Audio Processing: 【Spotify】, 【Podcastle】
- Includes AI + music apps like music recommendation and generation, as well as audio to text transcription.
- AI + Health and Medical: 【CareGPT】,【0xmd】
- AI used for medical diagnosis, health monitoring, and analysis of medical data.
- AI + General: 【Josh.Al】, 【Artifact】
- Includes other AI applications like smart home, traffic management, environmental monitoring.
The 8 Major Monetization Pain Points Faced by AI Mobile Apps
Facing such a promising yet highly competitive AI mobile app market, startups and smaller companies have encountered endless difficulties in monetization. Judging purely based on usage data, AI apps focusing on any application scenario have considerable numbers of users. However, paid conversion rates remain unsatisfactory.
Some AI app professionals have raised “heart-wrenching” questions to me:
Company A: We make male-oriented AI chatbot companions. We have lots of users but very few paid subscriptions.
Company B: We make female-oriented AI chatbots. Users always complain that the avatar designs are too “generic” with an artificial vibe and there are too few male character options.
Company C: We use the same AI models as our competitors, and our subscription fees are even half of theirs. But our market awareness is still inferior…
Employee D: Our company purchased several internal creative AI models. To be honest, they don’t seem to differentiate much in actual use and have a hard time maintaining consistent character depictions.
The capabilities appear incredible in promotions, but all kinds of problems arise when the apps launch. Based on user feedback, the SocialPeta Research Institute has summarized 8 major pain points for AI mobile app monetization:
- Technology Innovation and Selection:
- How to integrate AI technology into the app to meet user needs while ensuring reliability and performance.
- Whether to use existing AI models or develop proprietary ones, considering complexity, costs and competitiveness.
- Market Positioning and User Base:
- Identify the target market, whether B2B or B2C, domestic or overseas.
- Understand users’ needs, preferences and behaviors to tailor the app design.
- Business Model and Monetization:
- Build a sustainable monetization model, how to generate revenue from the app e.g. subscriptions, ads, in-app purchases.
- Balance profits and user experience for long-term viability.
- Technical Feasibility and UX:
- Assess feasibility of implementation and ensure AI meets user expectations.
- Provide smooth, intuitive and valuable user experience to engage and retain users.
- Competition and Differentiation:
- Understand competitors, analyze their strengths and weaknesses, determine how to establish competitive differentiation.
- Offer unique functions, UX or positioning compared to rivals.
- Data Privacy and Compliance:
- Handle user data privacy, comply with regulations and standards, prevent leaks and abuse.
- Establish transparent privacy policies to increase user trust and satisfaction.
- Market Changes and Scalability:
- Consider market trends and tech developments, ensure adaptability to evolving conditions.
- Business model and architecture should be flexible enough to meet future challenges and opportunities.
- Regulations and Legal Liabilities:
- Anticipate regulatory challenges, build compliance processes, ensure legal app operation.
- Address legal issues like user generated content, privacy and IP rights.
Currently, AI app developers are in a “red ocean” phase for user acquisition. Sustained growth surely depends on mobile marketing beyond core functionality development. In addition to development, app providers should also focus on marketing, since subtle model differences can hardly sway users’ final payment choices. Meanwhile, the intensity of marketing campaigns and exploration of high-volume advertising scenarios can often become the last straw guiding users to download and pay.
To effectively improve product conversion rates and impression of creatives, SocialPeta leverages a library of 13 billion mobile creatives and 100 million daily updates to track the latest market trends in real-time. With data-driven and targeted approaches, our Creative Index, Delivery Index, Reservation Index and App Index precisely break down hit products, assisting developers to closely follow trending apps for successful benchmarking and refinement.
Volume Buying Guide for AI Apps: Precise Brings Profits
Next Level AI Writing App 【TypeAI】
As a young AI writing app launched on May 30, 2023 (same day as ChatGPT), [TypeAI – AI Keyboard Extension] demonstrates its functional attributes fully through its app title, description, and 5 app store images. Specifically, it is a keyboard extension that enables seamless AI writing without copying or switching pages.
【TypeAI】Introduction to store promotion
Ad Copy: This app corrects all your mistakes
Content Framework: Lead capture headline + Context setup + App demo
The creatives first set up a common input scenario with many typos to highlight the app’s utility. Then, the ease of use is demonstrated through screen recordings of 【TypeAI】 operations. In particular, rapidly generating long passages of text from a single keyword with strong before/after contrast can often spur user interest in downloading and trying the app. The quality, speed, and length of generated text have become the most important components of AI generative app creatives. Whether the context is relatable determines if the creative can resonate with more potential users.
Background Removal Specialist 【PhotoRoom AI Photo Editor】
As a veteran photo editing app launched in March 2019, 【PhotoRoom AI Photo Editor 】successfully jumped on the AI bandwagon through a simple renaming operation. (According to SocialPeta App Intelligence, 【PhotoRoom: Studio Photo Editor 】added “AI” to its name on March 27, 2023).
Ad Copy: Best photo editing app
Content Framework: Couple interaction storyline + AI background removal demo
This creative was published on TikTok, using the background removal and couple interaction hook to grab user attention. It demonstrates quickly generating png images with transparent backgrounds via AI, which can then be freely combined with diverse backgrounds provided by the platform. This highlights the app’s two signature features.
Like the AI writing apps, the AI computation time here is also amazingly fast, delivering some contemporary AI speed shocks. (Some images are from online sources)
In general, AI+ apps are like gusts of demonic wind, loud but insubstantial. As consumers, we don’t care how many times you’ve iterated your model, how much data it consumed, what percentage of human intelligence it has, or how much computing power it packs.
We only care about the quality and speed of content generation, one-time purchase or subscription, expensive or budget friendly, and diversity of use cases.
If AI models are ingredients, then companies are master chefs, while customers only care about the color, aroma, and taste of the cuisine, not the storms behind it.