Monetizing Generative AI: Strategies for SaaS Companies

Generative AI, a form of artificial intelligence that can create high-quality content like human speech, images, music or other types of output, holds enormous potential for software-as-a-service (SaaS) companies. Not only can generative AI be used to enhance existing solutions and develop innovative new products, but it also opens up unprecedented opportunities for SaaS companies to generate revenue from these new capabilities.

Alongside building these advanced solutions, organizations must thoughtfully consider the monetization models that will allow them to effectively capitalize on generative AI capabilities. This article will explore key considerations for monetizing generative AI use cases, including the strategic role of pricing, the optimal price metric and broader monetization methodologies. 

Strategic pricing considerations 

When evaluating pricing models for generative AI solutions, organizations face several strategic, foundational considerations. Most importantly, should you begin with a low price to drive adoption as the market scrambles for product leaders, or should you price high to set the customer perception of premium value and establish a baseline for future pricing? This trade-off is especially important for generative AI given its nascency. While both approaches have merit, it’s crucial to weigh the implications when choosing the right model. 

Promotional pricing

Driving initial adoption with a lower price point can fuel quick uptake but may limit future willingness to pay (and therefore limit the future total addressable market (TAM)). This can become an issue as the software achieves product-market fit and the path to long-term success becomes reliant on enterprise-level sales. Generative AI companies have followed the incumbent SaaS strategy of offering low initial prices to encourage user adoption, with the goal of upgrading users to enterprise-level offerings. 

Three-tiered pricing in generative AI: Will it work?

It is yet to be determined whether this standard approach will be viable for the new wave of generative AI products and if these new entrants are differentiated enough to compete on an enterprise product level.

However, this approach has been a common strategy for new entrants to the generative AI market during a period of rapid growth and cutthroat customer acquisition. An illustration of this can be observed in Copy.ai, a prominent player in the generative AI content-writing toolbox, which adheres to this pricing structure (see Figure 1).

Source: Copy.ai  

Value pricing 

On the flip side, marketing generative AI as a premium feature with a high initial cost may limit short-term gain but can expand the overall TAM by targeting customers willing to pay a premium for advanced capabilities.

Monetization models for generative AI  

For effective monetization of generative AI, organizations need to thoroughly consider how to integrate it into their offerings and design suitable monetization models. Here are two common approaches: 

1. Embedding generative AI to justify premium pricing 

Amazon Web Services (AWS) is leveraging generative AI in several ways to enhance its suite of cloud-based products. For example, through its newly launched Bedrock program, Amazon’s deep learning platform SageMaker now supports GANs (generative adversarial networks), a type of generative AI, allowing users to create new data that resembles a given data set. This feature can be useful across various industries, from designing virtual environments to developing new pharmaceuticals (see Figure 2). 

Source: aws.amazon.com 

Canva, a popular design platform, provides another excellent example. Canva has integrated generative AI into its platform with Magic Design, enabling users to generate text, brainstorm on a whiteboard, create and edit images, transform templates, and update brand guidelines automatically across the content organization. This adds considerable value to the pro offering and improves the overall user experience (see Figure 3). 

Figure 3

Canva’s magic design presentations demo 

Source: Canva.com 

2. Generative AI paid add-ons 

Some generative AI use cases may be of particularly high value and thus can be sold as stand-alone offerings or bundled into higher-value packages as paid add-ons. A notable example is GitHub Copilot, which offers software developers code suggestions to help bolster productivity in real time (see Figure 4). 

Figure 4

GitHub’s Copilot product 

Source: Tenor via GitHub.com

This type of approach allows organizations to monetize specific, targeted features while offering flexibility to customers who can add these advanced features as needed. 

Generative AI pricing strategies

Setting the price for generative AI solutions is not only crucial for effective monetization but also represents a new frontier in the AI industry, echoing the early days of pricing innovation in the cloud computing or SaaS industries. While the tried-and-true per-user pricing model remains popular, the evolving landscape of generative AI is birthing novel pricing strategies that are pushing the boundaries and redefining the economics of the AI market (see Figure 5).  

Source: L.E.K. research and analysis 

Per-word pricing 

A less prevalent but increasingly relevant approach is per-word pricing models. These are suitable for applications such as content generation, where the number of words produced significantly determines value. For instance, Writesonic, a generative AI tool for marketing copywriting, currently adopts this pricing model. 

Per-user pricing 

OpenAI offers its advanced natural language processing model, ChatGPT, at a flat monthly rate of $20, granting subscribers unlimited access to enhanced features like faster response times and priority access to new features and improvements. Bearly, a company known for its suite of generative AI tools, implements a per-user pricing strategy, charging $20/month for each user, which ensures that businesses of varying sizes can customize their expenditure based on their specific needs and user count (see Figure 6).  

Source: Jasper.ai 

Another application of per-user pricing is add-on pricing, where the add-on generative functionality is only available to existing users of a software product.  

Token-based pricing

This cutting-edge pricing approach involves billing customers based on the number of tokens processed by the AI, where each token represents a discrete unit of information, such as a word or a character in text, or a pixel in an image. This method can be particularly advantageous for applications like Midjourney, which generates imagery using commands, where the complexity and volume of information processed by the AI directly impact the derived value (see Figure 7).

Source: Reddit.com

Open source

OpenAI’s free user introduction of ChatGPT, the pioneering publicly available application of ChatGPT, serves as a prime example for this pricing approach. This free-to-all strategy resulted in a thriving user community that helped train the OpenAI model, elevating its quality for future users. The success of this approach is evident in ChatGPT’s unprecedented achievement of amassing 100 million monthly active users within two months of launch, solidifying its position as the fastest-growing consumer application in history. Most importantly, this model laid the foundation for future revenue streams such as ChatGPT’s API access. 
 
Facebook’s PyTorch is another example of this approach. Originally released as a free, open-source machine learning library, PyTorch has been embraced by a vast community of developers and researchers for its flexibility, robustness and ease of use. Facebook ultimately monetized this success by providing customer support and other enterprise-level offerings, demonstrating the viability of this approach for generative AI commercial success. 

Add-on feature pricing 

Companies like Notion and Microsoft have added AI functionality as a paid extra. Microsoft offers AI-powered features such as smart recommendations and automation in Microsoft 365, available as part of its premium plans, while Notion has integrated AI features like advanced text recognition and predictive typing into its paid subscriptions, enhancing productivity for professional users (see Figure 8). 

Source: Notion.ai

“The future of pricing in this space will likely be a mix of these models, customized to the value provided and the target customer segment."

— Rajiv Pratap, Head of Strategic Verticals, Glean

Broader commercial aspects 

Monetizing generative AI extends beyond pricing considerations. Organizations should also address broader commercial aspects to maximize the potential of their offerings: 

Defining your role in the generative AI ecosystem 

Identifying your organization’s role in the generative AI ecosystem is essential for strategic decision-making. This can be anchored in proprietary model development, leveraging proprietary data, or utilizing in-house expertise and customer insights. 

1. Proprietary model development: Companies like Google, Nvidia and Adept have excelled in developing their own LLMs, offering full control over model evolution and changing the cost structure of the business. This might be your path if your strength lies in AI research. 

2. Harnessing proprietary data: For organizations that possess proprietary data but lack the expertise to build models, collaborating with an AI model provider can be beneficial. Companies like Jasper, Glean and Tome all help enterprises turn proprietary data into powerful assets while contributing to the training and refinement of the AI model. 

3. Leveraging in-house expertise or customer base: Using domain knowledge or customer insights to train and fine-tune AI models can be effective. Collaboration with AI providers can enhance models with your unique industry insights. 

This analysis aligns with the “build vs. buy” decision-making process that often happens within enterprise environments. Your position in the ecosystem guides the commercial partnerships you should pursue, leading to more informed decisions and setting the stage for maximizing the potential of generative AI offerings. 

Commercial partnerships

Organizations can pursue partnerships with relevant platforms, such as document management system or project management solutions, to access a broader market and enhance the value proposition of generative AI solutions. Integration with complementary tools can create synergies and offer customers a more comprehensive solution. For example, a number of major enterprises, including Box, Salesforce and Canva, have announced plans to build products using Google Cloud’s generative AI functionalities.   

Emphasizing privacy, security and control 

Privacy protection is crucial in the generative AI domain. A notable example is IBM’s Watson, which faced scrutiny when questions arose about its handling of sensitive patient data during the development of its healthcare-focused AI solutions. This incident underscored the necessity for rigorous data protection measures when dealing with client data or sensitive intellectual property in generative AI applications. 

Focusing on customer outcomes 

Successful monetization requires an emphasis on customer outcomes. Creating helpful resources, such as buyer’s guides, tutorials and user testimonials, can assist customers in understanding the benefits and value of the solution, thus ensuring the “stickiness” of the product.

For instance, Adobe’s Sensei platform, an AI and machine learning technology, provides detailed tutorials and use-case demonstrations to potential users. This approach effectively illustrates how its AI-powered tools can enhance design workflows and result in better creative outcomes, thereby showcasing the tangible value of the solution to new customers (see Figure 9).  

Source: Adobe.com  

Evaluating generative AI pricing strategies 

Strategic pricing decisions play a pivotal role in driving adoption and communicating the value of the solutions. The choice between embedding generative AI or offering separate paid add-ons depends on the nature of the use cases and customer preferences. Pricing strategies should align with market trends and consider innovative approaches like per-word or token-based pricing. Additionally, organizations should focus on broader commercial aspects, including partnerships, privacy and security, and internal cost allocation.

Successfully monetizing generative AI not only yields financial benefits but also enables organizations to deliver advanced and competitive solutions to their customers. By striking the right balance between pricing, integrating generative AI and addressing broader commercial considerations, organizations can position themselves for long-term success in this rapidly evolving field.

If you’re interested in discussing how L.E.K. Consulting enables companies to maximize earnings in the competitive generative AI market with innovative pricing strategies, please contact us directly at strategy@lek.com. We would be happy to provide some initial guidance.

L.E.K. Consulting is a registered trademark of L.E.K. Consulting LLC. All other products and brands mentioned in this document are properties of their respective owners. © 2023 L.E.K. Consulting LLC