How to choose the right AI platform

A simplified guide for finding the right
AI studio / AI platform / AI software

06/2024

Choosing the right platform, tool, or software for a company’s AI efforts can seem incredibly challenging – perhaps even almost impossible – especially for those who are not well-versed in the AI industry.

The vast array of options and the complexity of AI technology can be daunting. However, we believe that it doesn’t have to be this difficult. In this guide, we will explain how to relatively easily identify a category of appropriate AI software. By breaking down the selection process into manageable steps, we aim to simplify your decision-making and help you find the right AI solution for your business needs.

The simple-to-answer criteria

When selecting the right AI software or platform, several key factors related to your company’s specific circumstances should be considered. Let’s start with some simple-to-answer criteria.

Company size:

How many employees do you have? Determining whether you consider yourself an SMB (small or medium-sized business) or a large corporation can significantly influences your choice of AI tools.

Team constellation:

Next, evaluate your team’s constellation. Do you have a technical team with several IT employees, or are your key competencies in other areas? The presence of a dedicated technical team can affect the complexity of the AI solutions you can effectively implement and maintain.

Industry:

Consider the industry in which you operate. Are you active in a regulated industry? Do you have above-average demands concerning data security or compliance? These factors are crucial in selecting AI software that meets industry-specific regulations and standards.

Geography:

Finally, think about your geographic location. Where do you reside? For instance, if you are within the EU, you need to ensure that the AI tools you choose are GDPR compliant. Local regulations can have a significant impact on the software and platforms suitable for your business.

Pre-defined decisions

After having evaluated some simple criteria, let us assist you now with the next set of criteria, many of which are essentially pre-defined for every company. Having consulted > 100 companies from small to large in the last three years about AI-related projects, the following aspects of this chapter are in our opinion must-haves. We therefore make these choices for you very easy.

Strategic value:

Simply assume that the strategic value of AI is high. AI technology will significantly shape every industry, as it is a general-purpose technology akin to computers and the internet. Just as every industry relies on these fundamental technologies, AI will become an integral part of business operations across various sectors.

Data:

Every company needs external data. AI allows you to process public data that you would normally consume manually, now in an automatic way. Every business relies on external data for making informed decisions, so you will want a tool that can efficiently collect and process this data for you. Additionally, every company collects some internal data. Utilizing your internal data is another obvious functionality, which your desired AI studio needs to provide. It’s essential to choose a solution that can handle both external and internal data effectively.

Financial capabilities:

If you are a small or medium-sized business (SMB), it can be assumed that you have rather limited financial capabilities, and costs will play a significant role in your decision-making process. On the other hand, if you are a large enterprise, it can be assumed that there is a certain budget allocated to the advancement of your AI activities. While costs are also a consideration for large enterprises, they may be less crucial than for SMBs. Larger companies often have the option to engage consultants, external experts, or invest in custom developments.

AI functionality:

Some AI functionalities are simply non-negotiable for effective business usage. AI chats are the most flexible way to interact with AI models and are excellent for real-time interaction, making them great for certain tasks. However, most business processes are too complex to be efficiently solved through a chat interface alone. For these use cases, which we estimate to be around 80%, business users need and desire a workflow architecture. Therefore, it is essential to ensure that your AI solution provides both AI chat capabilities and a robust workflow architecture to meet all your business needs.

Make sure your AI solution provides both.

Subjective criteria

After having discussed some simple criteria regarding the choice of your AI platform, let’s take a look at three criteria, which are rather subjective.

Company-wide vs. department-specific solutions:

When selecting an AI platform, consider whether your company prefers a single tool for the entire organization or multiple specialized tools for each department.

Pros of a company-wide tool: Opting for a company-wide tool simplifies the overall approach, reducing complexity and avoiding silo solutions. This can lead to lower costs and improved collaboration between departments. Additionally, it facilitates better data collection and integration across the organization, enhancing acceptance for the AI tool.

Pros of multiple department-specific tools: On the other hand, using multiple department-specific tools can provide more specialized functionality tailored to the unique needs of each department. This can result in more effective and efficient workflows within each area of your business.

Empty tool vs. prepared templates:

Another crucial consideration is whether you prefer an “empty” workflow tool, where you can build your own solutions from the ground up, or prepared templates for common business tasks that are ready to use.

Pros of an “empty tool”: An “empty” tool allows for complete customization, enabling you to create solutions tailored precisely to your specific needs and preferences.

Pros of prepared templates: Prepared templates, however, require no technical skills for creating workflows, making them faster to implement and start generating business value. Since many business tasks are universal and require little customization, templates provide a good entry point into AI and can significantly speed up adoption and effectiveness.

AI model flexibility:

Consider whether you need the flexibility to choose between different AI models or if you are concerned about being locked into a specific model provider.

Pros of AI model flexibility: Choosing a solution that offers AI model flexibility prevents the lock-in effect, allowing you to switch models as needed to best suit your evolving requirements and to always switch to the best performing model.

Pros of no flexibility: While there are no outlined advantages of having no flexibility in the model choice, it can also work out to simply trust the solution provider to choose the best suited model.

Results

Having discussed 11 criteria for finding the right AI platform, let’s summarize with some practical options based on your company’s profile.

Option 1: You are an IT-savvy company (either small or large):

If your company is IT-savvy, whether small or large, you have the advantage of leveraging your developers to build workflows tailored to your specific needs. This approach allows for a high degree of customization and flexibility. You can obtain an API of your desired AI model, such as GPT-4, GPT-4o, or GPT-3.5, and utilize it to power your AI functionalities. By using prompt orchestration frameworks like LangChain, you can combine model prompts and streamline your processes effectively.

Furthermore, integrating a tool like LangChain with a workflow tool such as n8n enables seamless integration of various tools and communication with APIs. This combination allows you to build complex, customized workflows that cater to your unique requirements. While this approach does not provide ready-made templates or the convenience and speed of an out-of-the-box solution, it offers flexibility and customization. The mentioned developer tools can be recommended.

Option 2: You are a large enterprise with a suitable IT department

If you are a large enterprise with a dedicated IT department, compliance is a major prerequisite. Therefore, it is advisable to use an enterprise ecosystem optimized for large businesses, providing compliance certificates and related features. Prominent options include Google Cloud Platform, AWS, Microsoft Azure, or other enterprise cloud platforms. Within these extensive platforms, your IT department can access a vast amount of functionalities and third-party vendors. While these platforms have a steep learning curve and require some time to become familiar with, your IT department will soon be able to build AI solutions in-house based on an enterprise platform.

For each major cloud platform, there are consultants and specialists available for hire to speed up the onboarding or development process. Utilizing such a complex and powerful platform may not offer the most comfort or cost-optimized solution and can lead to a certain degree of lock-in, as the costs of switching to another cloud (i.e. through retraining your IT department) can be significant. However, this approach ensures compliance and scalability, accommodating the needs of potentially thousands of employees.

Option 3: You are a non-technical, small or medium-sized business (SMB)

If you are a non-technical, small or medium-sized business, the best approach is to select an out-of-the-box tool. Opt for a company-wide solution, as you can then avoid the complexities of handling multiple department-specific AI tools, and avoid the complexity of an enterprise ecosystem. Instead, choose a tool that includes data collection capabilities, ensuring you can gather and utilize the necessary data without extensive technical expertise.

Avoid empty workflow software, as lacking the skills to build workflows from scratch can lead to frustration, low usage, and slow progress. Instead, seek a solution that offers ready-made templates for common business tasks. These templates will enable you to quickly get up and running, achieving business value without the need for technical skills. By choosing a tool with these characteristics, such as the AI studio ValueFlow, you can effectively leverage AI in your operations while minimizing complexity and ensuring ease of use, making the adoption process smooth and efficient for your business.

Conclusion

After discussing 11 criteria and presenting 3 recommendations, we conclude that breaking down the decision-making process for buying an AI solution reveals that many sub-decisions are quite straightforward.

By considering your company’s characteristics, such as team composition and company size, many options can be easily eliminated. Certain criteria are simple to address, while others are essential must-haves for effectively utilizing AI in your business processes (such as data collection or AI workflow functionality). Once these fundamental criteria are met, you can focus on the remaining subjective factors. For the subjective criteria, companies can adopt an experimentation approach. By trying multiple tools for a short period, you can gather firsthand experiences and make a more informed decision.

When you are down to choosing between tools that vary only in subjective criteria, you are already very close to finding a suitable solution. In the end, there is no one perfect tool; the ideal choice greatly depends on the specific needs of your company. However, it is far more important to simply get started with any suitable tool rather than seeking the perfect one. Taking the first step is key, as it enables you to start reaping the benefits of AI for your business processes.

We wish a successful and rewarding journey.