March 31, 2023
Adam Hoeksema
With the boom in AI startups happening now it feels like a gold rush, but it still feels a bit unclear how all of these new AI based startups will generate revenue. Given that our job at ProjectionHub is to help startups build financial models, I wanted to dive a bit deeper into all of the potential business models that are being used or could potentially be used for AI based companies.
Before I dive into 7 of the most popular business models, I just wanted to hit on some of the key data points in this fast growing industry.
AI Startup Industry Data
How Many AI Startups are there?
There are over 8,700 AI startups listed on Crunchbase.
Here is a more recent list of 100 of the top AI startups from 2023.
How Many AI Startups have been Funded?
There have been over 21,500 individual funding rounds for the 8,700+ AI startups according to Crunchbase. Total funding to AI startups is over $143 billion.
Who are the Top AI Investors?
There are over 5,200 investors listed in the Crunchbase database that have invested in AI startups.
Some of the top AI investors include Techstars and Y Combinator at the seed stage, and Softbank and General Catalyst at the Series A and beyond stage.
AI Hardware Startups
Although one might assume that AI startups are all software based startups, there are key hardware improvements that enable AI, and there are a number of AI Hardware based startups that are primarily focused on improving processing speeds. These startups are developing new chips that enable AI processing.
AI Software Startups
Most of the well known AI startups like OpenAI are developing AI software.
Since most AI startups are focused on software, and since hardware business models are pretty straightforward when you simply sell a chip or product to an end customer, I wanted to focus on the different business models that AI Software Startups are using or may use.
AI Startup Business Models
It is not completely clear what the primary or favorite business model will be for AI startups. Although a traditional SaaS model might make sense in some instances, some AI tools and use cases are providing such significant financial value that a simple monthly SaaS doesn’t always make sense. Let’s walk through 7 of the potential business models for AI companies.
AI SaaS
Software as a Service (SaaS): In this model, AI companies provide cloud-based AI software solutions to clients, who access the service via subscription fees. These subscriptions can be offered with tiered pricing, based on features or usage. OpenAI offered ChatGPT+ as a SaaS product for $20 per month.
Create financial projections for an AI SaaS business model.
AI Platform as a Service (PaaS)
Platform as a Service (PaaS): Similar to SaaS, PaaS offers a cloud-based platform where clients can develop, run, and manage their own AI applications. The AI company generates revenue by charging for access to the platform, tools, and resources. Amazon Sagemaker is an example of an AI Platform as a Service.
Licensing AI
Licensing: AI companies can develop proprietary algorithms, models, or software solutions and license them to other businesses. This can be done through one-time fees, ongoing royalties, or a combination of both.
AI Professional Services
Consulting and professional services: AI companies can offer their expertise in AI implementation, customization, and integration with clients' existing systems. This includes providing training, support, and maintenance services. Over the coming years, every business should be utilizing AI in their processes. There will be a big opportunity for professional service agency business models where an agency helps implement AI tools into business processes.
Create Financial Projections for your Professional Services Agency
AI Data Monetization
Data monetization: AI companies can generate revenue by collecting, analyzing, and selling valuable data insights to clients, particularly if the AI company specializes in data-driven AI applications.
AI Pay-as-you-go Model
Pay-as-you-go model: AI companies can offer on-demand access to AI resources or services, charging clients based on usage, rather than a fixed subscription fee. OpenAI uses a pay as you go model with much of its pricing where the user can pay per image generated or per minute of audio produced or per unit of processing used.
AI Partnerships and Joint Ventures
Partnerships and joint ventures: AI companies can collaborate with other businesses to develop new products or services, sharing revenue from the resulting products or solutions. An example of this model could be an AI company that partners with a pharmaceutical company. The AI company could develop models for the drug discovery process and if the drug is successful, the AI company could earn a share of revenue when the drug gets to market.
If you have a unique AI financial model and you need help building a custom financial model, we would love to help!
