Why 70% of your AI success has nothing to do with AI

TL;DR Summary

Over 80% of AI projects fail, not because of weak algorithms, but because organizations neglect people and processes. The 70-20-10 rule reframes AI success: 70% depends on culture, leadership, and workflows; 20% on technology foundations; and just 10% on algorithms. Real transformation starts with the business problem, not the tool. Case studies in consumer goods and education prove measurable ROI when organizations prioritize people and process change first.

If AI is your strategy, you’ve already failed.

The real differentiator isn’t the tech, it’s whether your organization can adapt fast enough to use it.

Right now, companies are pouring billions into AI. In 2024 alone, global investment hit $252B and adoption surged to 78% (Standford). And yet, over 80% of AI initiatives still fail to scale or deliver outcomes (Rand).

Why?

Most leaders still treat AI as an IT project, not a transformation project. The algorithms aren’t broken. The organizations are. Misaligned leadership, outdated processes, and unprepared teams sink more AI initiatives than weak models ever will.

At ADAPTOVATE, we see this time and time again. Teams start with AI and then go looking for problems it might solve. The result? Disconnected pilots, shallow wins, and expensive technology stuck in “pilot purgatory”.

The question that should come first is always the same: What problem are we trying to solve?

Whether it’s reducing onboarding time, improving forecast accuracy, or automating manual work, the business problem needs to be specific. Only then can you assess the best solution. Sometimes that’s AI. Sometimes it’s not.

When you start with a clear business problem, you can judge whether AI is the right engine and how to deploy it in a way that works.

The 10-20-70 rule: A practical framework

AI delivery relies on far more than just model performance. The 10-20-70 rule reframes success:

70% People and Processes

The driver, the road, and the rules of the journey. This is where success is made or lost.

Redesign workflows, align leadership, reskill teams, and embed new behaviors. Without this, even the best AI sits idle.

20% Technology

The foundation and fuel. Infrastructure matters but isn’t the differentiator.

Good tech keeps the engine running, but it won’t win the race.

10% Algorithms

The engine itself. Strong models are important, but they won’t deliver alone.

A powerful engine without a road, driver, or destination is just noise.

Scaling AI requires redesigning workflows, building capability, aligning leadership, and enabling new behaviors. The whole vehicle must be road-ready, not just the engine.

• Shiny toy syndrome: chasing the latest model with no roadmap
• Pilot purgatory: promising initiatives that never scale
• Misaligned culture: 85% of employees say their organization has no clear AI strategy (Gallop)

• Business objectives drive use cases
• The right people define the problem before selecting solutions
• Reskilling employees is non-negotiable
• Failure is treated as learning
• Workflows and roles evolve alongside technology

• Commercial impact first: AI aligned to growth strategy and performance, not hype
• Behavioral change at scale: Mobile-first training to uplift capability across all levels of the business
• Proven transformation experience: Governance, operating model, and culture tailored for scale.

At ADAPTOVATE, we partnered with a Higher Education Provider that serves 30,000+ students to embed Gen AI across leadership and frontline staff, reducing student wait times by 50%.

We have proven experience in helping organizations turn AI ambition into impact

Case study: Global CPG company – driving AI implementation and upskilling

The Problem:

Despite strong interest in GenAI, a leading consumer packaged goods (CPG) company faced several challenges:

• Digitization and innovation: Struggling to keep pace with fast-changing consumer tastes, and innovative interaction models
• Operational inefficiency: Slow, manual processes in functions such as marketing, sales and logistics
• Competitive Differentiation: Without adopting emerging technologies, the business risked falling behind competitors who were already at the forefront of technology and innovation
• Employee Development: Staff were eager but under-skilled in applying AI to daily work

Leadership recognized that without a structured approach, GenAI risked being a buzzword rather than a value driver.

What we did:

ADAPTOVATE partnered with leadership to deliver a holistic AI implementation program.

• People & Processes: We worked with senior leaders to align on priorities and remove bottlenecks, while empowering “project champions” to drive cultural and behavioral change. A six-module Copilot training program was rolled out, ensuring teams had the skills and confidence to integrate AI into daily workflows.
• Technology: In a time-boxed 3-week discovery sprint, we assessed 115+ use cases across seven functional areas, prioritized the top five, and co-designed a delivery roadmap. Microsoft Copilot was deployed to enable practical automation and productivity gains across HR, Marketing, and E-commerce.
• Algorithms: Iterative testing of AI-driven use cases allowed the organization to experiment with new approaches to campaign design, communication drafting, and customer insight generation.

Impact:

• HR: 2+ hours saved daily drafting and editing internal communications
• Digital Marketing: 1+ hour saved daily on campaign content and visuals
• E-commerce: 3+ hours saved weekly generating insights from customer datasets
• Overall: measurable productivity gains, higher adoption of AI across functions, and participants using Copilot 3+ times per week post-training

The investment in people and processes (training, leadership alignment, and role redesign) created the conditions for change. The technology foundation (Copilot, infrastructure, prioritized use cases) enabled scaling, while the experimentation with algorithms allowed the company to test and refine new approaches. Together, this holistic balance unlocked sustainable value and a shift in ways of working.

Case Study: Higher education provider – enhancing student & staff experience with GenAI

The problem:

• Over 4,800 student queries a month, overwhelming staff capacity
• Manual processes slowed response times, with course information hard to access
• Students experienced long wait times, reducing satisfaction
• Staff lacked confidence and skills to use AI effectively

Leadership saw that without change, both staff productivity and student experience would continue to suffer.

• People & Processes: Partnered with leadership and frontline staff to identify priorities and bottlenecks. Trained staff to work with AI tools and empowered them through capability uplift programs. Focus was on embedding new behaviors by using AI daily to handle routine queries while freeing staff capacity for higher-value student support
• Technology: Designed, launched, and integrated a GenAI solution which allows staff to intuitively search their course information that was not easily accessible, to respond to queries faster capable of responding to student and applicant queries
• Algorithms: Experimented with different AI use cases (16 prioritized), testing the model’s ability to source correct information, generate accurate responses, and adapt to different contexts. This iterative testing refined the solution and built trust among staff

• 10x faster query resolution
• 50% reduction in wait time for students
• Staff empowered to source and deliver information more intuitively
• Frontline staff capability uplifted, with AI now part of their everyday workflows

The focus on people and process generated strong success from redesigning workflows, aligning leadership, and reskilling staff to embed AI in practice. The technology provided the infrastructure to scale support, while the algorithmic experimentation let the organization safely test and refine new AI use cases before wider adoption.

Final Thought

AI matters, but without strategy, culture, and execution, it won’t scale.

If your AI program doesn’t start with a business problem, it’s not a strategy, it’s a science experiment.

To deliver real impact:

• Start with the business challenge
• Build the culture and capabilities to solve it
• Then, and only then, bring in AI as the enabler

Do that, and AI will stop being the shiny toy stuck in pilot mode, and start being the engine of real transformation.

Is your organization positioned to scale AI beyond pilots and deliver measurable impact?

ADAPTOVATE helps organizations embed the 70-20-10 rule – aligning people, processes, and culture ahead of technology – to unlock sustainable value from AI.