STEP Method: deploy AI in your SME in 2025

Artificial intelligence is no longer the preserve of large corporations and technology giants. In 2025, it has become a strategic lever available to all organizations, whatever their size. Yet many SME managers are still hesitant to take the plunge, faced with a multitude of questions:

Where to start? What are the best use cases? How to mobilize teams? How to measure return on investment?

The answer to these questions lies not in technology alone, but in a structured approach that places people at the heart of the transformation. This is precisely what is proposed by the STEP method (Segmentation, Transition, Education, Performance), developed by researchers at the University of California and applied in the Pando Studio methodology.

In this article, we'll explore how this method can help you identify and deploy high-ROI AI projects in your company, while mobilizing your teams around a shared, sustainable vision.

AI challenges for SMEs in 2025

Before diving into the STEP method, it's essential to understand why AI adoption represents a strategic issue for SMEs in 2025.

Increase productivity and competitiveness

While large companies are investing heavily in AI, SMEs also need to evolve to maintain their competitiveness. Automating repetitive tasks frees up time for higher value-added activities. Employees using AI gain an average of 5 hours a week, precious time that can be reallocated to innovation or improving customer relations.

Bridging the skills gap

Faced with recruitment difficulties in many sectors, AI offers an alternative to absorb part of the workload and optimize the allocation of existing human resources. Without replacing employees, it enables them to concentrate on tasks where their expertise is truly indispensable.

Improving the customer experience

Customer expectations are constantly evolving towards greater personalization, responsiveness and availability. AI makes it possible to meet these expectations without necessarily increasing headcount, by offering personalized recommendations, 24/7 customer service or predictive analysis of needs, for example.

Stimulating innovation and differentiation

AI opens up new possibilities in terms of products, services or business models, enabling agile SMEs to differentiate themselves in their markets. It also makes it easier to explore new niches or territories, thanks to a better understanding of opportunities.

Enhancing data assets

Many SMEs have a wealth of under-exploited data. AI can transform this data into actionable insights and competitive advantage. Whether it's sales history, customer data or operational information, these resources often conceal unsuspected value.

The STEP method: a progressive, human-centered approach

The STEP method, developed following a three-year research project involving 10 knowledge-intensive companies, has established itself as a reference framework for successful AI transformation. It is based on four complementary pillars that form a logical, step-by-step process.

S for Segmentation: identifying where AI really creates value

The first and arguably most crucial step is to identify and categorize the company's tasks and processes according to their potential for automation or augmentation by AI.

As the research points out, "no AI can do everything a person does in a professional role" - so it's essential to determine precisely where AI will create significant value.

This segmentation is based on several criteria:

• Volume and repetitiveness of tasks
• Data availability and quality
• The relationship between cognitive complexity and value creation
• Time and reactivity constraints
• The need for large-scale customization
• Impact on employee satisfaction

The aim is not to automate as many processes as possible, but to identify those where AI will bring the greatest benefit, both for the company and for the teams.

T for Transition: planning and supporting change

Once the opportunities have been identified, the next step is to plan and support the transition from traditional operating modes to AI-enabled processes.

This transition is not limited to technical aspects, but takes into account all human and organizational issues. Case studies show that the most successful companies generally choose to deepen or enrich existing roles rather than downsize.

The transition involves :

• Redefine certain processes and responsibilities
• Adapt interfaces between teams or departments
• Plan the necessary resources (human, technical, financial)
• Draw up a realistic timetable with milestones
• Anticipating and managing resistance to change

E for Education: developing the necessary skills

The third pillar consists of training employees and making them aware of new tools and working methods, developing new skills and encouraging continuous learning.

Research shows that companies that invest in continuous training see a 30% reduction in the attrition rate of their employees, representing significant savings on recruitment and integration costs.

Education covers several dimensions:

• Raising awareness of the possibilities and limits of AI
• Technical training in tool use
• The development of new skills (data analysis, supervision of AI systems, etc.).
• Sharing knowledge and best practices between teams

P for Performance: measure impact and adjust strategy

The final pillar concerns performance assessment and the ongoing adaptation of the AI adoption strategy. This involves rethinking appraisal metrics to reflect new skills, learning and mutual support between employees, while adopting shorter, more dynamic appraisal cycles.

This approach to performance involves :

• Define relevant indicators (operational, business impact and adoption metrics)
• Implement continuous learning cycles (measure, analyze, learn, adjust)
• Promoting collaboration and knowledge sharing
• Communicate regularly on successes and learnings

Implementing the STEP method: a practical guide for SME managers

How can SME managers apply this method to their organization? Here's a practical guide to organizing and running an AI diagnostic workshop based on the STEP method.

Workshop preparation

Participants to be involved :

• Executive management (for strategic vision)
• Department managers (for their process knowledge)
• Representatives of operational teams (for their business expertise)
• Possibly, depending on the context: CIO, innovation manager, HR

The ideal is to form a group of 5 to 10 people, sufficiently diverse to cover the different aspects of the business, but small enough to enable effective exchanges.

Format and duration :

• A full day (8am-5.45pm) to cover all the necessary aspects
• Alternating presentations, discussions and group work
• Facilitated by a STEP-trained facilitator

Workshop sequence

Morning:

• Introduction and awareness of AI capabilities in 2025
• Presentation of concrete examples of the use of AI in your sector
• Identify relevant contexts for AI in your company
• Discussion of the legal framework and potential risks

Afternoon:

• Introduction to STEP with focus on segmentation
• Task segmentation exercise by department
• Assessing the potential ROI of identified initiatives
• Project prioritization and roadmap development

Key success factors

To maximize the effectiveness of this approach, several factors need to be taken into account:

Start with "quick wins": organizations that start with quick wins (high-impact, low-effort initiatives) establish positive momentum that facilitates broader AI adoption. These quick wins enable :

• Demonstrating the value of AI to the most skeptical
• Developing confidence in the approach
• Acquire valuable initial skills
• Fuelling team enthusiasm

Adopt an iterative approach: Rather than a rigid plan, design your roadmap as a living document that will be regularly updated according to :

• Results of the first initiatives
• The evolution of available technologies
• Changes in organizational priorities
• Learning and feedback

Measure impact at multiple levels: To properly assess the effectiveness of your AI initiatives, integrate three levels of metrics:

• Operational metrics (use, accuracy, processing time)
• Business impact metrics (costs, productivity, quality, customer satisfaction)
• Organizational adoption metrics (skills, commitment, innovation)

Actively involve employees: Research clearly shows that team involvement right from the design phase is a key success factor. This involvement makes it possible to :

• Benefit from their irreplaceable business expertise
• Reducing resistance to change
• Identify opportunities that management may have overlooked
• Promote adoption of deployed solutions

Conclusion: AI as a lever for sustainable transformation

The adoption of AI should not be perceived as a simple technological evolution, but as a transformation that affects the entire organization. The STEP method offers a structured framework to guide this transformation, placing people at the heart of the process.

By following this progressive approach - Segmentation, Transition, Education, Performance - SMEs can identify and deploy high-ROI AI projects, while mobilizing their teams around a shared, sustainable vision.

The challenge is not to adopt AI to do what everyone else is doing, but to integrate it strategically where it creates real value for your company and your employees.

Technology is evolving rapidly, but the fundamental principles of the STEP method will remain relevant:

identify concrete opportunities, support change, develop skills and measure impact.

In 2025, the question is no longer whether AI will transform your industry, but how you will use this transformation to strengthen your competitive edge and create new opportunities for growth.

FAQ : Frequently asked questions by SME managers