🎯 True Corp Interview Prep

"AI transformation fails when it lives only in the technology team. The real multiplier is when AI changes how the sales team qualifies leads, how customer care resolves issues, how operations predicts failures before they happen — and ultimately, how commercial outcomes improve.

At VGI Digital Lab, I experienced this directly. The platforms I managed touched every function — advertising, loyalty, payments, distribution. Technology decisions had direct commercial consequences across all of them. When our Rabbit LinePay platform had a processing issue, it wasn't an IT problem — it was a transaction failure affecting 8.6 million users and real-time revenue.

For True, the framework I'd use: identify the highest-frequency human decisions in each BU — things people do 50+ times a day — and ask whether AI should be making that decision instead. That's where you find the 3x value from AI in sales and marketing you've described. It's not the flashy use cases. It's the high-volume routine ones. Customer care resolution, lead scoring, enterprise onboarding, network incident triage — those are the targets."

"What you demonstrated at Telenor is that touch-free isn't a single project — it's an operating model shift. You need AI embedded in the process, not bolted on top. And you need the data foundation first — which True now has post-One Network integration.

I'd sequence it by volume and impact. First: customer care resolution — the highest frequency of routine decisions, directly tied to NPS and churn. Second: network incident response — Genie is already there, the question is how far to extend autonomy. Third: enterprise onboarding — the process that directly affects commercial velocity for the B2B segment.

The commercial case for touch-free is often framed as cost reduction. I'd frame it differently: a network issue resolved by Genie in minutes instead of hours is a customer retention event. A customer care issue resolved in 2 minutes versus 20 is an LTV event. Touch-free operations doesn't just reduce cost — it generates commercial value that appears in the revenue line, not just the cost line."

"The gap isn't a technology problem. True already has the infrastructure, BASIC5 framework, True AI Hub with 50+ models, Microsoft and SoftBank partnerships. It's a commercialization and adoption problem — which is exactly what this role exists to solve.

I'd attack it in three moves:

First — fix the GTM motion. True AI Hub currently offers 50+ models — that's a product catalog, not a commercial proposition. Thai enterprise buyers don't want to choose a model. They want to solve a business problem. I'd restructure the offering into 5-6 vertical-specific propositions: AI for retail, AI for manufacturing, AI for financial services, AI for healthcare, AI for logistics. Sector-specific outcomes close 40-60% faster than technology pitches.

Second — activate the channel. Direct enterprise sales is slow and expensive. True has relationships with SIs, ERP vendors, accounting software providers. Build a partner program that lets them bundle BASIC5 capabilities. This multiplies your effective sales force without proportional cost.

Third — exploit the data moat. True has network and customer data that no competitor — not AWS, not Google, not AIS — can replicate. Build 2-3 AI products that only True can offer. Retail footfall analytics built on True's mobile location data. Supply chain intelligence built on logistics partner data. These command premium pricing and are defensible.

Target: 11-12% enterprise revenue share in 24 months as a realistic milestone toward 15%. The 73/18 gap — 73% of Thai businesses plan AI adoption, only 18% have deployed — tells us demand is there. True needs to be the infrastructure that closes that gap."

"Resistance to transformation isn't about the technology — it's about identity and incentives. People resist when they think change threatens their role or when they're not measured on the new behavior.

At Tokio Marine Thailand, I was part of a major merger integration — two separate companies, two cultures, two legacy IT systems combining into one. You can't force that. The only approach that works: give people a role in designing the new model, not just receiving it. The most skeptical people are your best quality control — bring them inside the process.

For True specifically, with the DTAC cultural integration still ongoing AND a new AI-First mandate, you have dual change programs running simultaneously. The workforce AI upskilling program — 100% of employees by end 2026 — is exactly the right instinct. But it only works if people see AI as a tool that makes their job better, not a replacement for it. The commercial case for change management: organizations with high AI adoption rates among employees generate 2-3x more value from the same AI investments. That's not a soft benefit — it's a hard commercial multiplier."

"You raised this exactly at MWC — nearly every organization is still struggling with AI value realization. I believe the problem is structural: most companies measure AI inputs, not outcomes.

I use a three-layer framework:

Layer 1 — Activity metrics: use cases in production, models deployed, adoption rate. Necessary but not sufficient. Too many AI programs stop here and declare success.

Layer 2 — Outcome metrics: revenue generated per use case, cost eliminated, conversion rate improvement, churn reduction. This is where real value lives.

Layer 3 — Portfolio health: value realization rate (% of projected value actually delivered), time-to-value, ROI by BU. This tells you whether your AI investment allocation is correct — not just whether individual projects worked.

For True specifically, I'd add one enterprise-specific metric: AI-attached revenue — what percentage of new enterprise contracts include an AI or data component. That single metric drives the right commercial behaviors across the sales team and directly tracks the gap from 7-8% toward the 15% target."

"Days 1-30 — Listen and map. No big announcements. Meet every BU head, the CDAO João, key commercial leads at TrueBusiness, and the Microsoft/SoftBank/Thoughtworks partnership contacts. Map the current AI use case portfolio: what's in production, what's in pilot, what's stuck and why. Understand the True AI Hub commercial pipeline — how many enterprise clients, what sectors, what's the conversion rate from demo to contract.

Days 31-60 — Diagnose and prioritize. Produce an internal assessment: top 5 commercial opportunities, top 3 execution risks, quick wins available within 90 days. Present to Dr. Teeradet and relevant C-suite. The 73/18 gap — 73% AI intent, 18% deployment — tells me exactly where to look. I'd identify the 10 largest enterprise accounts that plan AI adoption but haven't deployed, and build a targeted closing plan for those specifically.

Days 61-90 — Execute first wins. One visible commercial win, one governance improvement, one cross-BU alignment unlock. Set the 12-month portfolio framework with committed commercial targets per quarter.

What I won't do in 90 days: launch a major reorganization, make large vendor commitments, or announce a new strategy before I've earned the trust of the people who'll execute it."

"Your MWC observation that sovereign AI is putting telcos back on the map is exactly right — and True is uniquely positioned to capitalize on it in Thailand.

Enterprises in banking, healthcare, and government cannot freely move their data to foreign hyperscalers. Data residency requirements, PDPA compliance, and national security considerations create real constraints. True has a sovereign AI advantage that no hyperscaler can replicate: locally-operated True IDC infrastructure, backed by Microsoft's technology through the CP Group partnership, with a Thai company's face and accountability.

This is a genuine commercial differentiator — not just for regulated sectors but for any Thai enterprise concerned about data sovereignty. I'd make this a primary positioning pillar for the enterprise AI commercial strategy: 'True-powered AI: world-class technology, Thai sovereign control.' That's a message that resonates in board rooms when CISOs are in the room.

And it aligns with the CP Group agenda — Ruza, your work connecting CP Group's multi-sector AI vision to True's infrastructure is the practical execution of exactly this. Retail data, agricultural data, fintech data — all of these benefit from sovereign AI infrastructure."

"Governance without teeth is just PowerPoint. Governance without speed is a competitive disadvantage. The answer is federated governance: central standards, decentralized execution.

True already has the foundation — AI Council, Responsible AI Policy, first in SEA. The execution layer needs to match that ambition. My model: the AI Council sets non-negotiables — risk thresholds, data usage standards, explainability requirements, responsible AI guardrails. Each BU has a designated AI program owner accountable for delivery within those standards.

The practical tool: a use case registry. Every AI initiative is logged with: business sponsor, expected commercial value, data sources, risk classification, and review date. Low-risk uses cases auto-approve. High-risk ones get council review. This creates speed for the 80% and oversight for the 20% that matters.

I track two health metrics: use case velocity (idea to production speed) and governance compliance rate. If velocity is high but compliance is low — risk problem. If compliance is high but velocity is low — bureaucracy problem. Both are failures."

"CP Group changes the scale of the opportunity dramatically. True alone has the telecom data asset. True within CP Group has telecom data + retail data + agricultural data + financial services data + logistics data. That's one of the most comprehensive multi-sector data ecosystems in Southeast Asia.

The AI opportunity this unlocks: cross-sector intelligence products that no standalone telco or retail company could build alone. A Thai SME that uses CP's 7-Eleven supply chain AND True's connectivity AND Krungthai Bank's financial services — there's an AI offering there that touches every part of their business.

For the enterprise commercial strategy, this means True AI Hub isn't just a telco product — it's an entry point to the CP Group AI ecosystem. I'd work with the CP Group AI function to develop cross-sector bundles that only the CP constellation can offer. That's a market positioning no competitor can match.

This is exactly the kind of platform thinking you've been building at the CP Group level — I'd see my role as the commercial execution arm of that vision within True."

"I operate at the practitioner level — I've personally built and run AI pipelines, LLM-orchestrated agentic workflows, RAG architectures, and automated data ingestion systems. I understand why data quality and MLOps matter more than model selection, and where generative AI genuinely adds value versus where it's hype.

That said, I deliberately don't position as an ML researcher — that's not the job at this level. The job is knowing enough to ask the right questions of technical teams, hold vendors accountable on their claims, and translate between AI capability and business outcome. In my experience, that translation capability is actually rarer at SVP level than pure technical depth.

The practical value: I can sit in a vendor demo and know when they're overselling. I can review an AI architecture and flag scalability or data governance risks before they become production problems. And I can take a technical team's output and immediately frame it in commercial terms for a board or executive sponsor. That's the bridge this role requires."

"Telco and financial services share the same fundamental DNA: massive regulated data assets, complex customer lifecycle management, real-time systems where failure has immediate commercial consequences, and the need to translate technical capability into commercial products in highly regulated environments.

The AI strategy patterns are identical. What's different is the asset class — True's advantage is network data and connectivity infrastructure, where my advantage was transaction data and financial infrastructure. Both are proprietary, real-time, behavioral datasets that third parties can't replicate.

But honestly, I'd turn the question around: True's '4 Big Moves' is a business transformation, not a network transformation. The critical skills for this role are commercial strategy, cross-BU execution, enterprise AI commercialization, and AI governance. Those are industry-agnostic.

And there's one specific advantage I bring as a non-telco native: I see True's assets the way an enterprise buyer does. I know what makes a technology partnership compelling because I've been on the buying side. That external perspective is rare in telco leadership."

"Two risks I'd watch closely.

First: execution speed versus governance quality. The '4 Big Moves' sets aggressive timelines — 100% AI upskilling by end 2026, major enterprise revenue growth. Moving that fast, with post-merger integration still ongoing, creates pressure to cut governance corners. If a high-profile AI deployment fails publicly — a biased decision, a data breach, a wrong output in a customer-facing system — it sets the entire AI-First program back by 12-18 months. True's Responsible AI policy and AI Council are the right foundation. The risk is if commercial pressure overrides governance discipline.

Second: internal efficiency versus external commercial traction. Organizations often optimize for internal AI adoption metrics — how many tools deployed, how many employees trained — while the external commercial pipeline stagnates. If True measures itself on internal AI maturity while the enterprise revenue gap stays at 7-8%, the commercial case for the program weakens with every board cycle.

My job in this role would be to hold both lines simultaneously — governance discipline AND commercial velocity. That tension is exactly what makes this role hard and interesting."

"I'll give you two connected stories that show the commercial thread.

At SCB Abacus, I was part of the founding operational team that built Thailand's first fully AI-powered digital lending platform — MoneyThunder. We deployed a machine learning credit scoring engine that approved loans in under 15 minutes with zero human intervention, using alternative data for borrowers that traditional banks had rejected. Loan volumes grew 3-4x annually. The platform went on to raise $32 million in VC funding and reached 20 million app downloads. This is AI value realization in practice — not a pilot, not a proof of concept, but a commercial engine that ran at scale.

Then at VGI Digital Lab, I've led IT and Operations across a ฿4 billion digital platform business — Rabbit Card at 14 million users, Rabbit LinePay at 8.6 million users, OOH advertising, O2O commerce. During my tenure, the company's revenue grew from ฿2.5 billion to ฿4.2 billion — 69% growth. Every technology decision I made had a direct commercial consequence. System availability was transaction revenue. Data quality was targeting accuracy. Integration speed was commercial launch speed.

The pattern across both: I operate where technology directly touches revenue. That's what I bring to True."

"Three things, honestly.

First — I'm a practitioner, not just a strategist. Most SVP-level AI candidates have directed AI programs. I've personally built and operated them — agentic AI systems, LLM pipelines, RAG architectures. That hands-on depth means I can hold technical teams accountable and detect vendor overselling. At SVP level, that's rare.

Second — I've operated where AI failure has immediate financial consequences. Capital markets. Digital lending. Insurance. Payment platforms. In those environments, every AI model is a live commercial decision. That's a more demanding proving ground than most enterprise AI programs. It means I have zero tolerance for AI that looks good in demos but fails in production.

Third — I understand the specific challenge you're trying to solve. The 73/18 gap — 73% of Thai businesses want AI, 18% have deployed it. True has the infrastructure, data, and partnerships to close that gap. What's missing is the commercial bridge between capability and revenue. That's not a technical problem. That's the exact problem I've been solving for 20 years across five industry sectors.

I don't build AI systems. I build AI systems that make money."

"What Telenor achieved is genuinely world-class — touch-free operations, cloud-first infrastructure, AI embedded across functions. The foundation True now has from that DNA is a massive advantage.

If I were thinking about where True's context differs: the commercial monetization layer. Telenor's transformation was largely internally focused — operational efficiency, network intelligence, customer experience. True's opportunity — given the CP Group constellation and the Thai enterprise market gap — is to also build an externally-facing commercial AI business.

True AI Hub is the right vehicle. But selling AI capability to Thai enterprises requires a different GTM motion than deploying AI internally. It requires sector-specific propositions, channel partners, outcome-based pricing, and a customer success function that ensures enterprises actually realize value — not just buy access.

That external commercial layer is where I'd focus that Telenor's transformation didn't need to. Not because Telenor did anything wrong — but because True has a commercial opportunity that Telenor as a pure telco didn't face in the same way."