Operating Model
Prediction Is Halfway
Knowing what a customer will probably do is real progress. Knowing why, and acting on it in time, is where the value actually lands.
June 17, 2026 · 3 min read
Essays on AI & the operating models that work
The Cover Story
AI capability is easy. The returns are the hard part.
The technology works. The returns come from redesigning the operating model underneath it — the slower, harder work that turns capability into results.
5
Countries, GE capability-center buildout
GE (GECIS)
80+
Transformation programs directed
Moody's
40%
Average efficiency gains
Deloitte
1,000+
Global engineer network
Code and Theory
25 years inside
GE CapitalMoody'sDeloitteCode and TheorySigmaArc
01
Perspective
In 1999, on a GE Capital operations floor in Gurgaon, I took 80% of the cycle time out of a billing process. We didn't put a clever tool on top of the old way — we took the work apart so most of the steps no longer existed. The savings held because there was nothing left to drift back to.
Twenty-five years later the lesson hasn't moved. A copilot license tells people their jobs are getting easier. Redesigning the operating model tells them their jobs are changing. One is a subscription; the other reaches the P&L. That's the work I do.
02
On Building
View allField notes from twenty-five years of making transformation land — and making AI reach the P&L.
Operating Model
Execution Is a Structure Problem
A good strategy that never moves didn't fail on the thinking. It failed because no one built the structure to carry it.
June 12, 2026 · 4 min read
Global Capability Centers
The GCC Is Where Transformation Happens — Not Where It Gets Outsourced
As AI automates the work GCCs were built to do, the captive center isn't the casualty of transformation — it's where transformation actually lands.
June 15, 2026 · 5 min read
Operating Model
Why AI Capability Is Easy and AI Returns Are Hard
The technology works. The returns come from the operating-model work underneath it.
June 1, 2026 · 5 min read
Workforce & Talent
The Layoff Mirage
Cut-first AI transformation is a cost program wearing a strategy costume.
May 15, 2026 · 5 min read
Subscribe
Follow the writing. New pieces, occasionally. No noise.
03
Work
In-house leadership · GECIS, India (later Genpact) · GE Capital – Consumer Banking (US)
Building GE's capability centers — at the birth of the category
- Outcome
- Attrition cut from ~30% to under 12%. A capability-center model established across five countries — the operating model the industry would later call the GCC.
- Challenge
- In the earliest days of global offshoring, GE was extending analytical and financial work from GE Capital – Consumer Banking in the US to India and beyond, with no template for how to build a durable capability center. The hardest problem wasn't standing up the work — it was holding the people: attrition ran near 30%, and every departure took hard-won knowledge with it.
- Approach
- Worked across GE Capital – Consumer Banking in the US and GE Capital International Services (GECIS) in India — one of the world's first global capability centers, later spun out as Genpact. From 2002 to 2005, built out GE's capability-center model across five countries: India, China, Hungary, Mexico, and Canada. Attacked retention directly through university partnerships — including the Madras School of Economics and Dalian University of Technology — that let employees pursue advanced degrees while staying with the company.
I sent Raj to India to set up and manage the GE Money Global Analytics Center… still running strong today.
In-house leadership · Code and Theory (Stagwell-backed)
Modernizing an acquired capability center for the AI era
- Outcome
- Integration delivered across all four dimensions, and two new revenue-generating practices stood up inside the portfolio company — cost taken out, performance instrumented, new EBITDA lines created.
- Challenge
- A newly acquired 200+ person engineering center in India sat outside the agency's delivery model — separated by culture, structure, process, and practice — while the network needed revenue lines it couldn't yet serve.
- Approach
- As Global Chief Technology Officer, modernized the acquired India capability center through post-acquisition integration across all four dimensions, stood up the agency's federal government and Adobe practices from zero, and built AI-enabled performance management and analytics across a 300-person direct organization — within a global network of 1,000+ engineers.
…the rare combination of curiosity and vision when leading teams.
External engagement · Deloitte Consulting
Rebuilding sales and service operations for global financial institutions
- Outcome
- 40% efficiency gains in insurance operations, across distribution, contact center, and onboarding.
- Challenge
- Banking, insurance, and asset management clients ran distribution, contact center, and onboarding on fragmented processes and aging systems — slow to serve, expensive to run, resistant to change.
- Approach
- Led 10+ digital sales and operating-model programs across these institutions, directing teams of 100+. Reset the operating models, instrumented the work, and sequenced delivery so the change held after the program closed.
…deep financial services expertise… played a pivotal role in developing our go-to-market strategy.
In-house leadership · Moody's
Building a global enablement function from zero
- Outcome
- 60% improvement in client onboarding.
- Challenge
- Commercial teams across 15+ countries worked from inconsistent processes, duplicated reporting, and a sprawl of legacy systems that obscured the client relationship and slowed onboarding.
- Approach
- Launched the firm's first global sales enablement function and led the global Salesforce implementation across every region. Built the commercial-side capabilities — commercial operations, marketing operations, and program management — consolidated under the company's global capability center, distinct from ratings. Retired 25+ legacy systems and 700+ reports onto a single operating spine.
Engagements described reflect work conducted as an external consultant (where labeled) or during prior in-house employment. Client and employer details are anonymized or generalized to protect confidentiality. Past results do not guarantee future outcomes.
04
About
25 years inside financial services and enterprise technology, building and rebuilding the functions these institutions run on. One thread runs through it: I've built or extended capability centers across three companies. At GE — joining GECIS in 1999, one of the world's first global capability centers, later spun out as Genpact — I built out the capability-center model across five countries before the industry had a name for it. At Moody's, I built the commercial-side capabilities — commercial operations, marketing operations, and program management — consolidated under the company's global capability center, distinct from ratings, and ran change management for a multi-year transformation of 80+ programs reaching 2,000+ employees across 12 countries. At Deloitte, I led financial services transformation as a senior growth leader. Most recently, as Global Chief Technology Officer at Code and Theory — a 2,000-person Stagwell agency — I modernized an acquired India capability center for the AI era, inside a sponsor-backed business: post-acquisition integration, cost and operating-model improvement, and new revenue lines built against an investment thesis.
I'm now founder of SigmaArc, a decision-intelligence platform for financial services, and I advise firms on integration, operating-model change, and making AI reach the P&L.
05
Engagements
I work with a small number of firms each year. Whether an engagement runs as advisory, interim, or fractional follows the problem.
The work spans financial institutions, private equity firms and their portfolio companies, and the firms that serve them — wherever a decided change has to land: an integration, a capability build, an operating-model reset, an AI program that has to reach the P&L.
06
If this resonates, get in touch.
A direct conversation about what you're working through. No pitch.
Contact card
Raj Bhatia
Founder, SigmaArc
- raj@rajbhatia.com
- linkedin.com/in/rajbhatia
Scan the code or add the card to keep my details on hand for when the timing is right.
Scan to save
Rather talk live? Find a time