Recap: My "AI in Action" Talk at Singtel's FY26 Townhall

3 AI tools. 1 keynote. 2 retail wins.

I recently had the opportunity to speak at Singtel’s FY26 Partner Townhall on 18 July 2025 - a closed-door event for the operators of Singtel’s authorized retail stores across Singapore.

My session was titled “AI in Action,” and it focused on two things:

First, to show what’s possible. I walked through how AI is already transforming areas like marketing, store intelligence, and frontline enablement. The goal was to spark ideas and show how fast the landscape is shifting.

Second, to bring it down to the ground. I visited one of Singtel’s retail stores ahead of the townhall, mapped the real workflows, and shared two practical AI ideas that could be implemented immediately.

Let me quickly recap a few of the more interesting things I showed during the session.

1. AI Avatar of Singtel’s Business Director (HeyGen + ElevenLabs)

To kick things off, I wanted to show just how far AI realism has come.

So we created a full AI avatar of Singtel’s Business Director, Daniel Soh and embedded it directly into the Singtel’s Townhall deck.

What I used:

• A photo of Daniel

• A 2-minute voice recording of him speaking in English

What we got:

• A hyper-realistic video of Daniel presenting and it’s fully lip-synced (created via HeyGen)

• His actual voice, cloned using ElevenLabs

• And the surprise: it translated into Mandarin too with synced lip movement

It was a live example of how AI can localize messaging, at scale, in seconds.

2. Built the Entire Deck in Gamma (AI-Powered Slides)

I built the entire AI portion of the Singtel’s Townhall deck myself using Gamma.

Gamma is an AI-powered tool that helps turn rough outlines into clean, structured slides fast.

What it helped me do:

• Turn talking points into full slide flows

• Apply clean, consistent layouts without manual formatting

• Move fast on feedback from the Singtel team

The result?

A polished, professional deck which was created end-to-end in under 60 minutes, not days.

3. How US Retail Chains Use AI to Pick Store Locations (Placer.ai)

I also shared a quick case study from the U.S. where retail chains are using geolocation AI tools like Placer.ai to decide exactly where to open new stores.

Instead of relying on gut feel or old traffic reports, these tools use:

• Real-time foot traffic data

• Mobile location signals

• Demographic movement trends

The result?

More accurate store placements, stronger retail performance, and faster decision-making all powered by location intelligence.

4. Shared Actual Live AI Workflows

I shared a few real AI marketing systems I’ve built that are already running inside live campaigns.

• One helped a company save over $12,000 by automating a process that used to take hours — built in under 10 minutes

• Another enabled a team to scale their ad spend from $30K/month to $250K/month by speeding up creative testing with AI-generated ad visuals and copy

I showed how tools like OpenAI, image generation, and simple automation flows (like the one below) can remove bottlenecks without adding headcount.

5. Mapped a Real Retail Workflow → Found a Quick-Win AI Use Case

Before the Townhall, I visited one of Singtel’s retail stores to observe frontline operations firsthand.

I spent time speaking with the senior sales associate on duty and also asked one simple question:

“What are the tasks you repeat every day?”

The goal was to spot areas where AI could step in to augment speed, accuracy, and customer experience.

One quick-win opportunity emerged clearly: AI Wi-Fi Consultancy.

Here’s the gap:

• Sales associates often spend 5–10 minutes per customer advising on optimal mesh Wi-Fi placement based on home layout, ONT location, and signal strength.

Here’s the fix:

• Use tools like ChatGPT to auto-analyze floorplans and output mesh router guidance instantly using Singtel’s own recommendations.

The outcome?

• Faster consults

• More accurate recommendations

• And a scalable process that doesn’t require more headcount

Wrap Up

Getting to walk the store floor, speak with the frontline team, and then share live demos with the wider group made it feel less like a keynote… and more like a working session that mattered.

Grateful for the opportunity and honestly, I’d love to do more of these.

Helping teams turn ideas into action is where AI gets interesting!

I test AI workflows on real marketing problems.

And turn the useful ones into 10x Playbooks.

So business owners and solo marketers doing it all themselves can skip the guesswork… and copy what actually works.

→ I drop one of these 10xPlaybooks 🚀 every week. Don’t miss the next.

John