What 25 Years of Transformation Failure Taught Me About AI Readiness
Created on 2026-02-06 08:40
Published on 2026-02-10 09:00
The career that became a laboratory for understanding why change succeeds and fails
I have been fired twice.
I have lost $250,000 on a business that failed.
I have been publicly humiliated in front of my entire executive team.
I have watched transformation initiatives I led collapse despite doing everything the textbooks said to do.
And I have succeeded at scale, leading change that reached 33,000 employees with 93% awareness and 72% participation.
Twenty-five years across consulting, executive roles, and business building in APAC, UK, US, and Australia taught me something that most AI experts miss entirely:
Transformation is not a technology problem. It is a human problem. And humans are irrational.
This is the story of how I learned that lesson, and why it matters for every organization trying to figure out AI.
The First Failure: Electrolux
In my late twenties, I was promoted from Marketing Manager to General Manager of Blondal (Electrolux Direct), responsible for their water filtration division. Thirty days after joining.
I was young, ambitious, and absolutely certain that results were all that mattered.
Over the next twelve months, I delivered 241% sales growth. KPMG audited the numbers. They were real.
Then I was fired.
My Chairman, Gunnar Broberg, looked at me and said: “Indhran, you are brilliant. But you are not ready for management.”
Here is what 241% growth actually looked like from the inside:
The sales team resigned. Not individuals. The team.
The company was still losing money despite the revenue growth.
Every relationship I had with peers was destroyed. Other department heads avoided me. I was toxic in the hallway.
I had optimized for one variable: the number on the spreadsheet. I moved fast. I hit targets. I ignored everything else.
The bridges I burned. The trust I destroyed. The people I treated as obstacles rather than partners.
I was a Destroyer. That had been my childhood nickname. I was the kid who took everything apart. It followed me into adulthood.
At Electrolux, I dismantled the organization while appearing to save it.
Gunnar saw what I could not see. My results were real, but they were unsustainable. I was extracting value from the organization faster than I was building it. Keeping me would have cost more than losing me.
The lesson: Results without relationships is destruction.
It took me years to understand what that meant. Now I see it everywhere in AI transformation.
Organizations deploy AI to optimize metrics. They move fast. They hit targets. They ignore the human systems that determine whether those results can be sustained.
AI accelerates whatever is already there. If you are a Destroyer, AI makes you destroy faster.
The Second Failure: Enterprise AI
Years after Electrolux, I became CMO at an enterprise AI platform in Houston, Texas.
I thought I had learned my lesson. I was working sixty-hour weeks. My calendar was packed. I was present in every meeting.
Then my CEO said something in front of the entire executive team:
“Indhran is the laziest person I have ever met.”
Sixty hours a week. Visible effort everywhere. And he called me lazy.
He was right.
I was using busyness as camouflage.
Every time technical discussions started, I deferred. “That is for the product team.” Every time AI architecture came up, I nodded along without understanding. “I trust the engineers.”
I was leading the marketing of an AI platform I did not understand.
I had once confused NLP, neuro-linguistic programming, the communication technique, with NLP, natural language processing, the AI technology. I never corrected the gap. I let my ignorance compound.
I was a spectator pretending to be a participant.
That public humiliation gave me a choice: transform in thirty days or be fired again.
I chose transformation.
I cancelled half my meetings. I stopped performing busyness. I started learning. I spent evenings understanding our platform, our architecture, what made it actually work.
Within three months, I could write every Gartner newsletter draft myself. Not because I had to. Because I finally could.
The rescue of an Indonesia mining company deal happened because I had developed real fluency. I understood what we sold well enough to solve problems I would have previously delegated.
The lesson: Busyness is the enemy of depth.
In the AI era, this lesson matters more than ever.
AI can generate ten defensible answers before you finish your coffee. It can fill your calendar with outputs. It can make you feel productive while you are actually just present.
The value has shifted. From creating to evaluating. From producing work to judging whether the work is right.
Leaders who stay busy without going deep will be exposed. AI will not cover for their lack of understanding. It will amplify it.
The Third Failure: The Chocolate Business
Between corporate roles, I built a chocolate business with Pierre Ledent, a Master Chocolatier.
Premium Belgian chocolate. Handcrafted. Beautiful. The kind of product that makes you close your eyes when you taste it.
I was brilliant at marketing. The packaging was gorgeous. The brand story was compelling. The photos made people want to reach through the screen.
I was terrible at business model.
$250,000 later, I understood the difference.
Hidden costs everywhere:
Import duties that eroded margins before we sold a single bar.
Cold chain logistics in tropical Singapore. Chocolate melts. Keeping it from melting is expensive.
Retail margin expectations that squeezed us from every direction.
Shelf life constraints that turned unsold inventory into expensive waste.
Every cost I had not modeled compounded. Every assumption I had not tested became a leak. The business bled money in a dozen places I had not anticipated.
I confused great marketing with a great business model. The photos were beautiful. The financial model was broken.
The lesson: The costs you do not see are the ones that kill you.
I see this constantly in AI adoption.
The visible costs are obvious. Licenses. Implementation. Training. These appear in the business case.
The hidden costs compound silently:
Opportunity cost. What you are not doing while the initiative stalls.
Credibility cost. The organizational scar tissue from failed transformations. The cynicism that makes the next initiative harder.
Culture cost. The fear that spreads when people see AI as threat rather than tool.
Competitive cost. The gap that opens while you are stuck in pilot purgatory.
The chocolate business taught me to look for the costs that do not appear in the spreadsheet. They are usually the ones that matter most.
The Pattern Recognition
Three major failures. Each teaching a different lesson.
Electrolux: Results without relationships is destruction.
Enterprise AI: Busyness is the enemy of depth.
Chocolate: The costs you do not see are the ones that kill you.
But there was a pattern underneath all three.
In each case, I had focused on what was visible while ignoring what was invisible.
At Electrolux, I focused on the visible metrics while ignoring the invisible relationships.
In Houston, I focused on visible busyness while ignoring invisible understanding.
At the chocolate business, I focused on visible marketing while ignoring invisible unit economics.
The invisible always won.
This pattern appears everywhere in AI transformation.
Organizations focus on visible technology while ignoring invisible readiness.
They focus on visible pilots while ignoring invisible adoption.
They focus on visible announcements while ignoring invisible capability gaps.
And then they wonder why 95% of AI initiatives produce zero return.
The invisible determines the visible. Always.
The Success: HSBC
Not everything I touched failed.
At HSBC, I was one of five managers selected globally to deliver a transformation program reaching 33,000 employees serving 10 million customers across Asia Pacific.
The results: 93% awareness. 72% active participation.
Here is what we did differently.
We did not launch to everyone simultaneously. We did not mandate adoption. We did not rely on corporate communications to create change.
Instead, we found the Sparks.
In every function, there were opinion leaders. People others watched. People whose adoption would signal that this was worth paying attention to.
We identified them. We equipped them first. We supported them intensively. We helped them succeed visibly.
Then we let them pull their peers along.
We did not push the transformation onto the organization. We built capacity for the organization to pull the transformation toward itself.
By the end, 93% of the organization knew about the transformation. 72% had actively participated.
Not because we mandated it. Because we made it worth participating in.
The lesson: Find the Sparks. Equip the willing. Let them pull the rest.
This is exactly how AI adoption should work.
MIT’s research found that over 90% of workers are already using personal AI tools for work tasks. There is a Shadow AI Economy operating in every organization.
These shadow users are your Sparks. They have already adopted. They have already figured out what works. They are already getting value.
Find them. Learn from them. Equip them. Let them pull their peers.
The 5% who succeed with AI understand this. The 95% who fail keep trying to push adoption onto people who are not ready.
The Crucible: Kilimanjaro
In 1993, I became the first Malaysian to summit Mount Kilimanjaro.
The final push started at 2am. The scree slope. Loose volcanic gravel that slides beneath your boots with every step. Two steps forward, one step sliding back. Frozen darkness. Air so thin every breath felt insufficient.
By dawn, we reached Gilman’s Point. 5,681 meters.
The view was spectacular. The sense of accomplishment was real. You can get a certificate at Gilman’s Point that says you summited Kilimanjaro.
Five of my companions stopped there.
They took their photos. They got their certificates. They celebrated their achievement. They started the descent.
But Gilman’s Point is not the summit.
Uhuru Peak, the true summit, is another 200 meters up. Another hour of climbing when every cell in your body wants to stop.
96% of the climb is complete at Gilman’s Point. The certificate is in hand. The photos look great.
But you did not reach the top.
I continued alone with my guide.
That last hour was the hardest. Every step was a decision. The altitude made thinking slow. My body wanted to stop.
And then: Uhuru Peak. The true summit. Sunrise over Africa spread below. The glacier. The sky. The silence.
They got the certificate. I got the summit.
The lesson: The last 4% is where the transformation happens.
I see this everywhere in AI transformation.
Organizations do the assessment. They run the pilot. They get some adoption. They announce success. They have their Gilman’s Point certificate.
But they stop at 96%.
The last 4%, scaling beyond early adopters, building the Context Graph, sustaining momentum past the first quarter, that is where transformation actually happens.
95% of AI initiatives fail. Most of them stopped at Gilman’s Point.
The Synthesis: Why AIR APAC
These experiences across twenty-five years led me to found AIR APAC, the Center for AI Readiness in Asia Pacific.
Not because I wanted to start another consulting firm. There are enough of those.
Because I kept seeing the same pattern.
Organizations failing at AI transformation for the same reasons I failed at Electrolux. Focusing on results while ignoring relationships. Focusing on the visible while ignoring the invisible.
Executives repeating my earlier mistake. Busy without depth. Present without understanding. Delegating AI to IT without ever grasping what AI would change about their business.
Companies bleeding money like my chocolate business. Hidden costs compounding. Assumptions unexamined. The visible looking good while the invisible collapsed.
And everywhere, organizations stopping at Gilman’s Point. Satisfied with the pilot. Celebrating the certificate. Never reaching the summit.
MIT’s research confirmed what my experience had taught me.
95% of organizations are getting zero return from AI investments. The divide is not driven by technology. It is driven by approach.
The approach that works is what I now call the Human Layer.
Leadership. Data. Capability. Process. Governance. Culture.
The invisible 90% that determines whether the visible 10% succeeds.
Why Mid-Market APAC
I focus on mid-market APAC for specific reasons.
Mid-market organizations have an advantage most do not recognize. MIT found that mid-market companies implement AI in 90 days while enterprises take 9 months or longer.
Fewer layers. Faster decisions. The ability to align in days rather than months.
But mid-market organizations also face a gap. They are too mature for the generic AI courses flooding the market. They are too lean for the McKinsey engagement that costs more than their annual IT budget.
They need something different. Expertise without the enterprise price tag. Frameworks designed for their context. Support that understands their constraints.
APAC adds another dimension.
What works in Silicon Valley does not work in Singapore, Seoul, or Kuala Lumpur. Cultural factors differ. Trust dynamics differ. Regulatory environments differ.
I have lived and worked across this region for most of my career. I understand that context is not a limitation. It is a moat.
Organizations that build AI readiness with APAC context in mind create advantages that generic global solutions cannot replicate.
The 18-Month Window
There is urgency to this work.
MIT’s interviews with 17 procurement and IT leaders established consensus: a strategic positioning window is closing between mid-2026 and early-2027.
Organizations that build AI readiness now create compound advantages:
Data advantages. Every interaction with AI systems generates context. Organizations that start now will have years of accumulated learning that cannot be bought later.
Capability advantages. People who have worked with AI for two years develop judgment that newcomers lack.
Switching costs. Once systems are trained on your workflows, moving becomes prohibitively expensive.
The organizations that delay are not just missing current opportunities. They are ceding structural advantages that will define competition for the next decade.
This is why I do this work. Not because AI is interesting technology. Because the window is closing and most organizations are not ready.
The Path Forward
My failures taught me what my successes confirmed.
Transformation is not about technology. It is about humans.
Humans are irrational. They act on identity rather than information. They resist change even when change benefits them. They trust relationships more than data.
Any AI transformation that ignores these realities will fail. The technology will work. The organization will not change.
The Human Layer is my answer to this challenge.
Six dimensions. Deliberately designed. Built before the technology accelerates.
Leadership that understands enough to lead.
Data that flows where it needs to go.
Capability to judge, not just use.
Processes redesigned, not just automated.
Governance that enables, not just restricts.
Culture that supports experimentation.
This is what the 5% build. This is what the 95% skip.
Twenty-five years of failure taught me what success requires.
It requires the invisible work. The work that does not get applause. The work that happens before the technology deploys.
It requires building the Human Layer.
The question is whether you will learn from my failures or repeat them.
What experiences shaped your understanding of transformation? What failures taught you the most?
If you want to assess your organization’s Human Layer, comment “SCORECARD” below. I will send you the assessment I built for mid-market APAC leaders. It maps your readiness across all six dimensions and shows you exactly where the gaps are.
The window is closing. The question is whether you will be ready when it does.