Rethinking Analytics & Decision-Making
AI doesn't make better decisions — skillful people, working with AI, do. We help you build the culture that turns analytics and generative AI into better decisions.
Edgius in a Nutshell
Building an Analytics-Driven Decision-Making Culture
What brings you here?
Most people arrive with a problem, not a product in mind. Start where it hurts.
From Data-Driven
to Analytics-Driven
Data doesn't help make better decisions.
Analytics leverages data through skillful people to bring the insights needed to make better decisions. Much more than data goes into good analytics and good decision-making.
The Elephants Story
To get better at analytics-driven decision-making, we need to address the elephants in the room.
-
Elephant 1 · Munger
50–80%
is the average time analysts and data scientists spend on data preparation before they can start analyzing.
-
Elephant 2 · Manual
500–1000
hours per year is the time analysts spend on repetitive manual work that could be automated.
-
Elephant 3 · Duplicate
+25%
is the average estimated wasted time on duplicated or highly similar analytics in large organizations.
-
Elephant 4 · Quality
+20%
is the loss in productivity related to data and analytics quality and noisy decision-making.
-
Elephant 5 · Question & Problem
85%
of surveyed CEOs say their organizations are bad at problem diagnosis — not answering the right questions.
-
Elephant 6 · Bias
180+
is the number of cognitive biases negatively affecting analytics and decision-making.
-
Elephant 7 · Diversity
+20%
increase in innovation and profitability are some of the benefits of people diversity. Do you leverage it?
The second act
A new herd has moved in
The seven elephants above are about analytics, and they have not gone anywhere. But generative AI arrived fast, and it brought a herd of its own — the things everyone in the organization can already see, and nobody puts on the agenda.
42%
of companies abandoned most of their AI initiatives in 2025, against 17% the year before — at a reported average of $7.2M sunk per abandoned initiative.
S&P Global, 2025
20%
of organizations reported a breach involving shadow AI — tools nobody approved — adding $670K to the average cost of the breach.
IBM, Cost of a Data Breach Report 2025
45%
of AI-generated code samples carried a security flaw, across 80 coding tasks and more than 100 models. Syntax pass rates rose from about 50% in 2023 to about 95%; security pass rates never moved.
Veracode, 2025 GenAI Code Security Report
−7.2%
in delivery stability for every 25% increase in AI adoption, across roughly 3,000 practitioners. DORA's 2025 follow-up found the throughput penalty had gone; the stability cost had not.
DORA, Accelerate State of DevOps 2024, and the 2025 follow-up
The quietest cost never reaches a dashboard: the more people trust an answer, the less they check it. Microsoft and Carnegie Mellon put that to 319 knowledge workers in 2025 and found exactly the trade — confidence in the tool, against critical thinking. None of this is an argument against AI. It is an argument for naming the herd out loud, which is the only way anyone has ever moved an elephant.
Microsoft Research & Carnegie Mellon University, 2025
“Elephants belong in the wild, not in your office!”
Sylvain Dufour
Founder, Edgius
Can you spot elephant #8?
Elephant 8: The Skills Gap
In the last 2 years, how much did your organization spend on technology for analytics, data warehouse, big data, BI and AI?
Now, how much have you invested in upskilling and empowering your people?
The Skills Gap
A significant challenge to the success of businesses and organizations.
59%
of the global workforce will need training by 2030.
39%
of workers' core skills are expected to change by 2030.
63%
of employers see the skills gap as the biggest barrier to business transformation.
From Technology
to Skills Mastery
The model is the smallest part of the problem.
BCG's 10-20-70 rule puts roughly 10% of what makes an AI initiative succeed on the algorithms, 20% on the technology and data around them, and 70% on people and processes (BCG, Closing the AI Impact Gap, 2025). That 70% is the work, and it is the part most programs skip. Analytics automation, BI and generative AI all pay off the same way — when people with both technical and power skills know what to ask of them, and what to check.
Top 10 Core Skills in 2030
World Economic Forum, Future of Jobs Report 2025
- 1 Analytical thinking
- 2 Resilience, flexibility & agility
- 3 Leadership & social influence
- 4 Creative thinking
- 5 Motivation & self-awareness
- 6 Technological literacy
- 7 Empathy & active listening
- 8 Curiosity & lifelong learning
- 9 Talent management
- 10 Service orientation & customer service
- Critical Thinking & Problem-Solving · Power Skills
- Self-Management · Power Skills
- People Network · Power Skills
- Technology Use & Development · Technical Skills
From an Organization of Individuals
to Collective Intelligence
Organizations are more than the sum of individuals.
The most valuable asset of an organization is its people and how they work together: the collective intelligence. But most organizations don’t know:
- How do people work and collaborate together?
- How does knowledge flow between people?
- How diverse is their network to help generate innovation and de-bias analytics and decisions?
Understanding, nurturing and leveraging that collective intelligence is the key to unlocking an organization’s true potential.
How we engage
Assess first.
Pilot small, then scale what earns its keep.
Most generative-AI programs stall before they ever reach the P&L — MIT's 2025 study of enterprise pilots put it at 95%. The cause is rarely the model. It is the distance between a tool and the way an organization actually works. So that is where we start, and we prove the value on something small before anyone commits to something large.
MIT NANDA, “The GenAI Divide: State of AI in Business”, 2025.
The assessment looks at three things at once
People
An organizational network analysis shows how work really flows: who connects teams, where knowledge stops moving, and which people others actually turn to.
Process
We follow the work itself — the steps, the handoffs, and the workarounds your teams built because the system never quite fit. That is where automation earns its keep.
Technology
We take stock of what you already own and what it is ready for: data, platforms, governance, and the AI tools already in use, sanctioned or not.
01
Assess
A few weeks, not a few quarters. People, process and technology examined together, ending in a clear picture of where you stand and a short, ranked list of what is worth doing.
02
Pilot
One real problem, one workflow, one measurable outcome. Small enough to finish, concrete enough to prove — or to disprove, which is worth just as much.
03
Scale
Whatever earns its keep gets extended, with your own people able to run it, govern it and change it. Nothing important should depend on us being in the room.
No long discovery phase, and no platform decision before there is evidence to justify one.
We address these challenges with Coaching · Technology · People Network
We see analytics & decision-making holistically
Technology
to enable automation, citizen development and generative AI
Coaching
to upskill and empower people
People Network
to leverage the collective intelligence
Ready to build an analytics-driven decision-making culture?
Empower your people and get fast ROI.
Use Case
Combining Coaching, Technology and People-Network Analytics
- Leveraging Organizational Network Analysis to optimize coaching cohorts and increase people-network diversification.
- Measuring the benefits of the Analytics-Maker Coaching Program combined with automation technology.