Many AI pilots don't make it — here’s how your business can beat the odds
Sharon Prosser
Vice President, SMB Sales and Scaled Acquisition
AI offers small and medium businesses huge opportunities. But it can feel challenging for lean teams to capitalize on new tech. Our guidebook shows how.
Every generation of business technology is measured against very human ambitions: Does it remove friction? Make work easier? Create actual value?
Agentic AI is no exception to those rules.
AI has already shown it can summarize, synthesize, and accelerate work. In the past two years, there’s been a lot of exploration with it through pilots and demos. For mid-sized businesses across Europe, the Middle East, and Africa, there’s now a bigger opportunity at play — moving AI out of one-off pilots and into the real workflows, decisions, and systems you use every day.
That shift matters. Because, in the many conversations I’ve had with growing companies, they’ve often told me how their teams don’t have time for endless experiments to see what sticks or add governance after the fact. They need AI that works within their business — built to support their people, grounded in their specific context, and ready to scale securely.
I see this moment as a real turning point. Not because every organisation needs to transform overnight, that’s unrealistic, but because the path to operational value via agentic AI can now be built deliberately.
Perhaps best of all for small and mid-size organizations is that with AI, you don’t need a massive budget or specialised team to get started or scale.
AI is already delivering
According to our latest AI ROI report surveying thousands of business and technology leaders, 84% of organisations using generative AI are already reporting consistent, with 26% experiencing accelerating returns ROI.
These numbers are encouraging. But value does not appear just because a business has adopted AI. This can be especially true for small and midsize firms — previous research from Google Cloud found that only 12% of AI pilots were making it into production, in part because teams were ill-equipped to use these new tools, and partly because the tools were not fine-tuned to their needs.
A real return comes when AI is designed around the work your people actually do. It’s asking the question, “Where can AI make the biggest operational difference right now?” Answering that requires a more deliberate approach.
The goal is not to automate everything all at once but instead create repeatable progress, one workflow at a time.
At Google Cloud, one of the strategies we recommend to our SMB customers is called the 5D Framework. It’s a structured, five-step path to help you move from AI pilot to tangible business value. Detailed in our new playbook, Agentic AI for Lean Teams, the 5D Framework follows these steps:
- Diagnose where pilots fail, so you can avoid dead-ends and wasted resources.
- Define what better looks like for you.
- Design how AI fits into real workflows .
- Demonstrate a 30/60/90-day path forward with measurable results.
- Deploy securely, with governance and sovereignty.
This is how AI becomes more than just another productivity tool. It becomes a way to connect trusted context with action. So your teams can retrieve the right information, coordinate multi-step processes, and focus more of their time on higher-value work instead of repetitive tasks.
The goal is not to automate everything all at once but instead create repeatable progress, one workflow at a time. That’s where agentic value really begins.
Real-world competitive advantage
Let’s take an example of this framework in action. SYS-DAT Group — a 600-person Italian ICT company — built 12 agents that are transforming how they deliver software to clients across more than 15 industries.
The team started by looking at software implementation. Each time, they were producing around 20 documents, from business process analyses to technical specifications — all manual, with inconsistent quality and limited knowledge capture.
To change that, SYS-DAT Group chose Google Cloud's AI platform for its enterprise-grade AI capabilities, interoperability with their existing stack, and governance fit for GDPR and NIS2 requirements. Now, each of the 12 agents align with a stage of SYS-DAT’s standardized delivery methodology. The results speak for themselves:
- Productivity gains of up to 300%.
- Consistent, high-quality documentation regardless of who’s producing it.
- An expected 15%-20% productivity increase across delivery.
You don’t have to be a software company to reap the rewards of agentic systems. Take Genscom, a print company that transformed its business model by automating a crucial business workflow.
Manually designing personalised print campaigns with hundreds of variations used to require days of layout work. But by automating the entire workflow, Genscom was able to cut campaign creation to just 10 minutes, saving 340 hours of manual work every week.
The enhanced workflow — built using Gemini, Cloud Functions, BigQuery, and Cloud Run — has helped to triple revenue. The company ultimately doubled its headcount to meet growing demand.
Embedding AI into the firm’s core operations has been transformative.
“We're no longer a traditional printing company competing on volume,” said Stefaan Vanysacker, CEO and founder of Genscom. “We're an agile software provider that helps businesses cut through the noise and build real, lasting relationships with clients.”
A big moment for small and medium firms
For growing organisations, becoming an agentic enterprise means starting small, building with purpose, and scaling with confidence. Then repeating this across the next workflow, the next team, the next function. That’s how you ensure AI makes people’s work better in just 90 days.
To explore the full 5D Framework and practical steps for turning AI pilots into operational value, download our new playbook, Agentic AI for Lean Teams.
