The AI Wars Just Got Personal: Google’s AI Agents Are Now Running Your Ads

Google just changed the rules of digital marketing at I/O 2025 with the launch of AI agents in Google Ads. Discover what this means for growth marketers, how to adapt, and why working with AI—not against it—is your next competitive edge.

So this just happened earlier this week… The AI Wars Just Got Personal and They’re Inside Our Ads Account

We are now living in the AI Wars, and Google just sent in the ‘Terminators’.

At I/O 2025, the tech giant didn’t just unveil new features. It unleashed a battalion of AI agents — smart, tireless, and fully integrated into Google Ads. For growth marketers, this isn’t science fiction.

Forget faceless robots waging war in distant dystopias. These are inside your ad platform, rewriting headlines, adjusting bids, and optimising performance before you’ve had your morning coffee. If ChatGPT were the polite intern, this? This is Skynet learning to run media buying.

And here’s the twist: it’s not here to kill your job but to challenge it.

The bigger AI picture is becoming clearer: OpenAI leads in user adoption, Microsoft in enterprise productivity, but Google is coming for the growth stack. With unmatched access to user intent (hello, Search) and now Gemini-powered agents baked into every corner of its ecosystem, Google is rewriting what it means to do marketing in the AI era.

This isn’t about automation anymore. It’s about augmentation, and the marketers who know how to ride the wave instead of running from it will be the ones standing when the smoke clears.

If you thought performance marketing was already moving fast, buckle up. The new era isn’t just faster — it’s smarter, always on, and increasingly… not human.


Google I/O 2025: What Just Hit Us?

Google didn’t just update its product roadmap. It reprogrammed the marketing playbook.

At I/O 2025, it launched a suite of AI innovations that feel less like feature upgrades and more like an existential retooling. The most headline-worthy? AI agents are now fully embedded inside Google Ads. They don’t just help marketers. They do what we used to do — only faster, cheaper, and without needing coffee or a quarterly bonus.

But before we jump to “machines-are-taking-over” paranoia, let’s decode the actual announcements and what they mean for us on the front lines of growth.

🧠 AI Mode in Search: From Keywords to Conversations

Google’s new AI Mode turns traditional search into a full-on dialogue engine. You no longer get a list of links. You get synthesised answers, action steps, and the option to “keep going” with contextual follow-ups.

For growth marketers, this is both a dream and a nightmare. A dream because the customer journey becomes frictionless. A nightmare because we now need to optimise for conversations, not just clicks. Your SEO strategy just got an AI-shaped curveball.

🌊 Project Mariner: Your Agent Will Google That For You

Project Mariner is Google’s multitasking AI assistant. It doesn’t just respond — it acts. Think of it as the intern who not only researches the best CRM tools but also signs you up for trials, syncs your calendar, and sends a Slack update to your boss.

Implication? Expect a rise in fully automated conversion flows — all handled by AI. From a growth perspective, this means our new funnel touchpoints may no longer be human at all.

🧬 Gemini 2.5 Pro & Deep Think: Strategy as a Service

The brains behind the operation are Gemini 2.5 Pro, now with Deep Think mode. This isn’t your average autocomplete. It simulates layered reasoning, evaluating options before delivering an answer, like an analyst who’s actually good at their job.

This upgrade unlocks new possibilities in campaign planning, budget modelling, and even creative strategy. You’re not just delegating execution to AI — you’re increasingly delegating thinking.

AI Agents in Google Ads: Meet Your New Teammate (or Replacement?)

Google’s bet is clear: AI isn’t just a tool — it’s the new teammate. And these AI agents? They’re here to handle the grind so you can focus on the strategy.

⚙️ Functionality

These agents chew through your campaign data, generate creatives on the fly, optimise bids in real time, and even draft your performance wrap-up reports. They’re not perfect, but they’re relentless.

📈 Smart Bidding, Upgraded

AI-powered Smart Bidding Exploration takes historical data, cross-references with live signals, and calibrates for ROAS like a hedge fund algorithm. It’s not just about cost-per-click anymore; it’s about predictive profitability.

🎨 Creative Superpowers

Pair this with tools like Veo and Imagen, and you’ve got a creative engine that drafts high-quality ad visuals and videos at scale. We’re entering a world where every growth marketer is also a creative director, without needing to learn Photoshop.

What This Means for Growth Marketers?

Now let’s talk reality.

This isn’t just another shift in platform mechanics. It’s a redefinition of what “marketer” even means. AI won’t replace your job, but it will replace parts of it.

So the question isn’t “Will AI take my job?” It’s: “Will I know how to work with AI, or will I be replaced by someone who does?”

⏱️ Efficiency Gains: The End of Busywork

You’ll spend less time toggling through dashboards and more time making actual decisions. That’s a win. Campaign builds, creative iterations, and performance reviews are all streamlined.

🧭 Strategic Shifts: From Operator to Orchestrator

You’re not setting the dials anymore. You’re telling the system what outcomes matter and letting it figure out the rest. The value now lies in judgment, creativity, and context, not in button-clicking expertise.

🧠 Skillset Evolution: The 2025 Growth Stack

To stay ahead, you’ll need:

  • Comfort with prompting and AI workflows
  • Fluency in interpreting AI-generated insights
  • The ability to spot strategic angles machines still miss

This is your call to upskill — not just with courses, but with curiosity. Learn to speak AI as fluently as you speak ROAS.

Read more about how AI is impacting Performance Marketing here.

⚠️ Challenges Ahead: Not All Smooth Scaling

  • Control & Oversight: What happens when the AI makes decisions that don’t align with your brand voice or creative instinct?
  • Transparency: Can you explain to your client (or your boss) why the AI paused half the ad groups at 2 AM?
  • Adaptation Fatigue: Yes, it’s exhausting. But the 1% daily improvement mindset? That’s your edge. Don’t chase perfection. Compound progress.

Read more about the 1% compounding effect of improving your life here.

Final Thoughts: Embracing the AI-Driven Future

In 2023, we learned to prompt. In 2024, we started experimenting. In 2025? We partner with AI, or risk being left behind by those who do.

Google’s latest announcements don’t kill the role of the growth marketer. They kill the old definition of what a growth marketer is. What rises in its place is someone more strategic, more curious, and more adaptable.

So, my fellow comrades, the war is on. But you’re not being replaced by a robot.

You’re being upgraded.

🫶🏻 Thanks for reading till the end.

➡️ Follow Mervyn Chua and reshare to help others.

The Math Behind Going Viral: What Marketers Can Learn from Growth Loops, Network Effects, and Nerdy Equations

Discover the math behind going viral — from K-factors to growth loops, network effects, and Gladwell’s tipping point. Learn how to engineer shareable growth.

With the recent Meta-FTC trial dominating headlines, I found myself rewatching The Social Network — a film that, for all its dramatisation, still captures the raw energy of the early internet age. Somewhere between Jesse Eisenberg’s cold genius and that infamous “I’m CEO, bitch” line, something stirred. Maybe it was nostalgia. Or maybe it was the ghost of young Zuck whispering in my ear. Either way, my inner math nerd jolted awake, suddenly obsessed with one question: What’s the math behind going viral?

Because here’s the thing, in my line of work, this request is as common as coffee on a Monday: “Hey! I want to create this piece of content. Can you make sure it goes viral?”

As if virality is some checkbox you forgot to tick in the last campaign. (Spoiler: it’s not. But there is a formula. And yes, it involves math.)

In today’s landscape, growth marketers aren’t just creatives or copywriters. We’re system designers. We’re builders of loops, nudges, incentives, and networks, all engineered to nudge users into not just using a product, but spreading it. Because of the real growth? It happens when your audience becomes your distribution.

This post unpacks the real mechanics behind virality — from K-factors that measure how fast your user base grows, to network effects that make every new user more valuable than the last, to Malcolm Gladwell’s iconic “Tipping Point” and the underrated art of word-of-mouth.

Forget cat videos and dance trends for a second. Let’s talk about the equations (and human behaviours) that make ideas spread like wildfire. Because going viral isn’t magic. It’s math.

And it’s time we learn how to use it.


1. The Science of Spread: Understanding the Virality Coefficient

What is the Virality Coefficient (K-Factor)?

Let’s get to the math-y heart of virality: the K-factor. To put it simply, this number tells you how contagious your product or content is.

K = i × c

Where:

i = number of invites sent per user

c = conversion rate of those invites

So if each user sends 5 invites (i = 5), and 20% convert (c = 0.2), then:

K = 5 × 0.2 = 1

The Golden Rule:

  • If K > 1, your growth is exponential.
  • If K < 1, you’re leaking users faster than you can gain them.

Key takeaway:

If your K-factor isn’t above 1, your content isn’t viral — it’s just loud.

This isn’t just a SaaS metric or an investor buzzword. It’s a diagnostic for whether your product is self-propelling or reliant on paid crutches.

Where Network Effects Come In

The virality coefficient measures spread, but network effects determine value.

Here’s the upgrade in nerd math:

Metcalfe’s Law states that the value of a network is proportional to the square of the number of connected users.

Translation? If 10 people are using your product, it has 100 potential value points. With 100 people? That jumps to 10,000.

Why? Because each new user doesn’t just add value, they multiply it.

  • WhatsApp with just one user is useless.
  • With ten friends? It becomes essential.
  • With everyone? It becomes the default.

Network effects amplify virality. They don’t just help you grow, they make every new user more motivated to bring another.

2. Designing for Sharing: Growth Loops & Referral Engines

Growth Loops vs Funnels

Traditional marketers love funnels. But here’s the truth: Funnels are dead. Loops are alive.

❌ Funnels:

  • Linear. One-way.
  • Acquisition → Activation → Retention → Revenue → Referral

✅ Growth loops:

  • Circular. Self-sustaining.
  • Every user action feeds into acquiring the next.

Think of it like a flywheel: once you get the loop spinning, every bit of friction removed — and every ounce of value added — keeps the loop spinning faster.

Referral Incentives and Smart Nudges

Let’s talk behavioural economics meets viral design.

People don’t share just because you ask. They share because it:

  • Makes them look smart or generous
  • Gives them tangible value
  • Feels effortless

Great example:

Other success stories:

  • WhatsApp: seamless invite link in chats
  • Airbnb: $25 credit system that feels like gifting

Key takeaway:

The question isn’t “How do I make this go viral?”

It’s “How do I design this so the user wants to bring a friend?”

The Human Element: Word of Mouth and the Tipping Point

Gladwell’s Tipping Point and the Law of the Few

In his seminal book The Tipping Point, Malcolm Gladwell explained how ideas spread like epidemics but only when the right people are involved.

There are three types of viral agents:

  • Connectors — They know everyone. Social butterflies who can spread your idea across different groups.
  • Mavens — These are your product geeks. People who know all the details and love to educate others.
  • Salesmen — Persuasive personalities who can sell ice to an Eskimo.

Together, they form the perfect storm for virality. Without this mix? Even the best idea dies on arrival.

Why You Need the Right Influencers, Not the Biggest Ones

In the age of “#ad” fatigue and inflated follower counts, influence ≠ reach.

Real influence is about:

  • Trust
  • Relevance
  • Engagement

That’s why micro-influencers and community leaders often outperform celebrities. They speak directly to niche tribes, and those tribes listen.

Case study:

  • Glossier built an empire by sending products to 100 micro-influencers and superfans, not A-listers.

Word of Mouth is a Slow Burn That Becomes a Blaze

If you’re expecting overnight virality, you’re in the wrong game. Word of mouth is like compound interest — slow, steady, and eventually explosive.

  • One share leads to three…
  • Three to nine…
  • Nine to twenty-seven… until the curve bends, and momentum takes over.

But here’s the catch: This only works if your content is worth sharing. No nudge, network, or formula can fix boring.

Click here to read more about the effects of compounding in marketing and in life.

Key takeaway:

It’s not about mass media.

It’s about mass intimacy. One-to-one connections that scale, not one-to-many blasts that bounce.

Final Thoughts: Can You Engineer Virality?

Let’s get real: you can’t guarantee virality — not with the best creative team, not with the biggest ad budget, not even with a dance challenge that slaps.

But here’s the good news: You can architect for it.

Virality doesn’t happen by accident. It’s a function of intentional design, behaviour-driven nudges, and yes — a healthy dose of math. The most successful campaigns aren’t lucky; they’re engineered for shareability, built on systems that turn one user into many.

Key Takeaways for Every Marketer:

  • Know your K-factor — measure it, track it, and optimise it like your job depends on it (because it might).
  • 🔁 Build growth loops, not linear funnels. Great products don’t end at “conversion” — they feed themselves.
  • 🎁 Incentivise sharing — and more importantly, remove friction. Make the act of spreading feel like a reward.
  • 🎯 Find the right connectors — not just the loudest voices, but the ones who can truly move people.
  • 🧠 Create remarkable content — because if it’s not worth talking about, no formula in the world can save it.

At the intersection of psychology, design, and data lies the modern marketer’s greatest superpower: the ability to spark momentum that grows on its own.

Virality isn’t magic. It’s math, multiplied by human behaviour.

Master both — and maybe, just maybe, you’ll catch lightning in a bottle.

🫶🏻 Thanks for reading till the end.

➡️ Follow Mervyn Chua and reshare to help others.

📌 Save this post for future reference!⁣⁣⁣⁣

Growth Marketing ROI: Think Like a Portfolio Fund Manager

What if you managed your marketing channels like a financial portfolio? Discover how to drive smarter ROI by applying investment principles—diversify, monitor, and optimize like a fund manager.

So… with the recent market turmoil caused by a policy-tweeting President, I began to obsess over my personal investment portfolio. Mapping out my risk tolerance, allocating between equities, bonds, and even a cheeky play on thematic ETFs. Then it hit me, this is exactly how I’ve approached growth marketing for years.

Campaigns are like stocks. Channels are your asset classes. And ROI? That’s your return on capital. Whether it’s a bullish Meta campaign or a stable Google Search ad group, every move in marketing has its risk profile, expected return, and opportunity cost, just like managing a fund. It’s all the same game, just different dashboards.

And here’s where my finance training rears its analytical head: whether you’re investing $10K into ETFs or $1M into paid media, the fundamentals still hold. Know your objective. Balance risk and reward. Monitor religiously. Growth marketing isn’t about throwing money at shiny new platforms, it’s capital allocation with a creative twist.

So what happens when you start managing your campaigns like a portfolio manager? Let’s dive in. Your marketing mix may never look the same again.


1. Define Your Investment Mandate (a.k.a. Growth Objective)

Before a fund manager touches a single stock, they define their investment mandate. Income, growth, capital preservation — the goal shapes the portfolio.

Marketing isn’t any different. Are you acquiring new users? Building brand awareness? Maximising ROAS? Each objective requires a different mix of tactics, channels, and creative risk-taking.

And while I’d argue that every marketing dollar should ultimately ladder back to ROAS (let’s not pretend we’re a charity), the level of aggressiveness varies. Some brands want immediate cash flow; others play the long game with upper-funnel storytelling.

Your objective is your compass. It dictates how much you allocate to Google Search versus YouTube, branded versus performance, retention versus acquisition.

“Without a mandate, you’re just guessing with money.”

2. Diversify Campaign Bets (Like Building an Asset Allocation Strategy)

Enter Modern Portfolio Theory (MPT), the finance nerd’s holy grail. MPT says investors are naturally risk-averse and should diversify across assets to optimise returns while minimising volatility. The same logic applies to growth marketing.

Every channel has its own risk-return profile. Facebook might have a 10% ROAS upside, but it swings wildly depending on algorithm changes. Google Search? A reliable blue-chip with steadier returns. Influencer campaigns? The crypto of marketing: moonshot or dumpster fire.

Let’s break it down:

If you have four equally weighted channels with expected ROAS of 5%, 8%, 12%, and 18%, your portfolio’s expected ROAS is:

(5% x 25%) + (8% x 25%) + (12% x 25%) + (18% x 25%) = 10.8%

Not bad, but here’s the kicker: thanks to channel correlation (or lack thereof), your total portfolio risk could be lower than the sum of individual risks. That’s the magic of diversification. Facebook and TikTok may perform differently during the holiday season; one could tank, the other could thrive.

So design your stack accordingly:

  • Core Holdings: Your proven, high-conviction channels: Google Search, Meta, CRM.
  • Growth Picks: TikTok, UGC, influencer seeding, affiliate plays.

Base channel weightings on historical performance, CAC consistency, and campaign volatility. Measure each channel’s ‘variance’ using historical data, and track cross-channel correlations to understand how one campaign’s success (or failure) might affect others.

“Diversification is not just defense — it’s intelligent offense.”

3. Monitor Performance Like Stock Positions — and Cut the Losers, Double Down on the Winners

Fund managers have Bloomberg terminals. You have dashboards, or at least you should.

Every marketer should have a real-time view of channel spend, ROAS, CAC, CPM, and variance. Not just per channel, but in aggregate. Are you up overall? Or just being buoyed by one runaway success hiding a handful of underperformers?

But data’s only useful if you act on it. Set hypotheses for each campaign, like you would set target prices for stocks.

  • If Meta Ads doesn’t hit CPA targets in two weeks? Trim the position.
  • If YouTube starts outperforming by 20% with lower CPA? Scale it.

Here’s the tough love part: detach emotionally. The fact that “Instagram always worked before” doesn’t mean it’s a forever hold. In both markets and marketing, rational detachment is your moat.

“Be a performance pragmatist. Love the result, not the channel.”

4. Rebalance Regularly

The market changes. So should your marketing mix.

Just as fund managers rebalance portfolios quarterly, smart marketers reassess their stack frequently. What’s working? What’s stale? Are you overexposed to a declining channel? Is there a new ad format or beta worth testing?

Your media mix should reflect today’s consumer behaviour and not last quarter’s case study. Stay close to martech developments, algorithm shifts, and platform evolutions. What’s volatile today might be a stable performer next month (and vice versa).

Rebalancing isn’t just housekeeping. It’s strategic foresight.

“Past performance is not indicative of future results — in finance and in marketing.”

Final Thoughts: From Markets to Marketing – ROI Is All About Discipline

At the end of the day, growth marketing and investing share one simple truth: results come from discipline, not gut feel.

Thinking like a portfolio fund manager forces you to zoom out. It pushes you to look past the shiny new platform or viral ad and ask: Is this worth the risk? Does this align with my objective? How does this play with the rest of my mix?

It’s a mindset shift — from reactive tinkering to strategic capital allocation. It removes emotion, enforces structure, and most importantly, keeps your decisions grounded in risk-adjusted returns.

“The best marketers don’t just launch campaigns. They manage capital. Thoughtfully. Strategically. Relentlessly.”

So the next time you’re about to drop $100K on a campaign, pause. Take a breath. Ask yourself — would your portfolio manager do the same?

And now I want to hear from you.

👉 What’s your current “blue-chip” channel? The one you’d bet the farm on?

👉 And what’s your “high-risk, high-reward” play? The TikTok of your portfolio?

Drop them in the comments or DM me. Let’s compare portfolios. Who knows, your next winning bet might just come from someone else’s allocation.

🫶🏻 Thanks for reading till the end.

➡️ Follow Mervyn Chua and reshare to help others.

📌 Save this post for future reference!⁣⁣⁣⁣

Keep It Simple, Marketer: How to Evaluate Martech Tools Like an Investor

Discover a simple, profit-focused framework for evaluating martech tools like an investor. Learn how to cut through the noise, avoid shiny object syndrome, and make smarter marketing decisions that drive real business impact.

So this happened to me earlier this week at the MarTech Summit in Singapore… I had more than a few conversations that ended with the same question: “Is this tool really worth the investment?” 🤔

I’ll be honest. Being surrounded by some of the brightest marketers, coolest demos, and the latest marketing tech was super energizing. There’s a real buzz that comes from swapping ideas, learning about new platforms, and imagining the possibilities. ⚡ But that same buzz can quickly morph into overwhelm when you’re faced with a dizzying lineup of dashboards, AI-powered features, and bold promises that this is the tool that will change your life (and pipeline).

Sound familiar? You’re not alone.

In a sea of tools, demos, and jargon-filled pitches, how do you actually cut through the noise and decide whether to pull the trigger on a new martech investment?

Here’s a radical idea: Keep. It. Simple. 💡 Strip away the vanity metrics, the shiny features, and the FOMO. Because at the end of the day, if a tool doesn’t help your business make more profit — by increasing revenue or reducing costs — then it’s probably not worth your time (or budget). 📉💸

Let’s dive into how to make smarter, simpler, and more business-driven decisions when evaluating your next martech investment.

Why Marketers Overcomplicate ROI 🤯

Let’s face it. Marketers are notorious for falling in love with shiny new tools. It’s easy to get swept up by sleek demos, AI-powered this, machine-learning that, and dashboards that look like they belong on the Starship Enterprise. 🛸

But here’s the thing:

🔧 Features Over Functions

We often focus on what a tool can do rather than what it should do for our business. That leads to complexity over clarity and, ultimately, clutter in your stack.

📊 Too Many Metrics, Not Enough Meaning

With every tool claiming to give you “data-driven insights,” we end up swimming in KPIs but still struggle to make decisions that move the needle.

🧠 Analysis Paralysis

You’ve got dashboards for days but no clear next step. Sound familiar? When you try to track everything, you end up understanding nothing.

The Shift Needed

It’s time to stop asking, “Is this tool cool?” and start asking, “What’s the business impact?” The real question is: Does it move us closer to profit?

The KISS Framework for Martech Justification 💡

“Keep It Simple, Stupid (but Smart).” 😄

Let’s bring it back to the one question that really matters:

👉 Will this tool drive profit?

Everything else is noise. Strip away the fluff and focus on the two simple levers that drive profit: Revenue Growth and Cost Reduction.

💰 Revenue Growth

Invest in tools that earn their keep by helping you sell more or sell better:

  • Improve conversion rates: CRO tools, smarter funnels, or better lead scoring.
  • Enhance personalisation: More relevant emails or platform experiences = more engagement = more sales.
  • Boost retention: Loyalty platforms and CRM systems that drive repeat purchases = higher LTV.

💸 Cost Reduction

Save time, cut waste, and do more with less:

  • Automation: Think email workflows, content AI, and smart scheduling.
  • Reallocate manual effort: Free your team from grunt work so they can focus on strategy.
  • Better targeting: Stop burning ad dollars. Smarter targeting = less waste, more ROI.

A Finance-Informed Lens on Martech Investment 📊

Here’s where my background in finance and asset management comes in handy. Let’s take a step back and think like an investor.

🧮 Net Present Value (NPV) for Marketing Tools

Every martech tool is a business investment. And every good investor asks:

“Will the return outweigh the cost and is it better than using that money elsewhere?”

Break it down like this:

  • Upfront Cost: Licenses, integrations, onboarding, training. 💵
  • Benefits Over Time: Either in revenue gains or cost savings over 12 months (a typical contract period).
  • Opportunity Cost: What else could you do with that money?

So the mental model becomes:

“Will this tool deliver more value than its cost over a 12-month horizon?”

Even if you don’t build a full spreadsheet model, this mindset helps you make smarter, more grounded decisions.

Simple NPV Breakdown:

  • 🧾 Initial Cost = license + internal hours for implementation
  • 📈 Forecasted Impact = estimated uplift in conversions or time saved
  • 🔁 Discount Rate = your company’s risk appetite or benchmark ROI

No buzzwords. Just business thinking.

How to Build a Simple Business Case Without Drowning in Data 🛠️

Don’t worry, you don’t need to be a spreadsheet wizard to justify a tool. Just follow this simple, no-fluff approach:

1. Start with a Hypothesis

“If we implement this tool, we expect a 10% increase in lead conversion.”

2. Estimate the Dollar Impact

10% more leads converted × average revenue per lead = projected revenue gain

3. Add the Estimated Cost

Subscription fee + implementation hours (people x time)

4. Consider the Time to Impact

Will results show up in weeks or months? How fast can we get to value?

5. Align with Stakeholders

Finance and leadership don’t care about CTRs or impressions. They want to know if this makes or saves money. Talk in business terms and not just marketing jargon.

🚀 Final Thoughts: Be the Bridge Between Marketing & Business

At the end of the day, marketing isn’t just about crafting clever campaigns or plugging in the latest tools. It’s not just creative, and it’s not just technical — it’s a growth engine. 💡

And the most effective performance marketers? They think like investors. 📊 They know that every tool, every tactic, and every touchpoint needs to serve a higher purpose: profit.

When you lead with simplicity, profit-focused thinking, and business clarity, something powerful happens:

👉 You stop chasing shiny objects.

👉 You start making confident, data-informed decisions.

👉 And most importantly, you become a true strategic partner to the business. And not just the person who runs the ads or manages the tech stack.

So here’s your next move:

💬 If you’re currently evaluating a martech tool or stuck in one that isn’t performing, ask yourself: What’s the impact on profit?

If you can’t answer that clearly, maybe it’s time to go back to basics. Strip away the noise, follow the numbers, and focus on what actually moves the needle.

Because sometimes, the smartest strategy is the simplest one. 💥


Want more content like this? Follow along as I break down growth marketing with a business-first mindset — minus the fluff and with a healthy dose of real-world strategy. 👊

Boosting ROAS the Smart Way: The Math Behind More Profitable Ads

Learn how to optimize ROAS with smarter strategies. Discover the math behind Return on Ad Spend and how improving Average Basket Size, Conversion Rate, and Cost Per Click can drive better ad performance and profitability.

I’d be honest. When I first moved into digital marketing from a background in math and quantitative finance, I saw things a little differently. While most marketers focused on creative storytelling and audience psychology (which are undeniably important), I naturally gravitated toward the numbers. I wanted to break down ad performance the way I would analyze financial models—through data, formulas, and strategic optimizations.

One of the biggest misconceptions I’ve seen in performance marketing is the belief that cutting costs alone improves Return on Ad Spend (ROAS). Many marketers think that slashing CPC (Cost Per Click) or reducing spending on “underperforming” campaigns is the key to profitability. But that’s only part of the equation.

ROAS isn’t just about how much you spend, it’s about how much value you generate from every dollar. And to truly optimize ROAS, you need to understand its mathematical makeup. In this post, I’ll break ROAS down into its three core componentsAverage Basket Size (ABS), Conversion Rate (CVR), and Cost Per Click (CPC), and show you the strategic levers you can pull to drive better ad performance.

Let’s get into it. 🚀

What is ROAS? (Defining the Formula)

Let’s start with the basics. Return on Ad Spend (ROAS) is one of the most critical metrics in performance marketing. It tells you how much revenue you’re generating for every dollar spent on advertising.

ROAS Formula:

For example, if you spend $1,000 on ads and generate $3,000 in revenue, your ROAS is $3,000 / $1,000 = 3

This means you’re earning $3 for every $1 spent.

While a higher ROAS is ideal, blindly optimizing for it by just cutting costs can be misleading. To truly maximize ad performance, we need to break it down further.

Breaking Down ROAS: The 3 Key Components

ROAS is influenced by three key factors:

1. Average Basket Size (ABS)

What it means: The average amount a customer spends per purchase.

Why it matters: If customers spend more per transaction, your revenue increases without needing more conversions.

2. Conversion Rate (CVR)

What it means: The percentage of visitors who clicked on an ad and made a purchase.

Why it matters: Higher conversion rates mean you extract more value from the same traffic.

3. Cost Per Click (CPC)

What it means: The cost of acquiring each visitor to your site.

Why it matters: Lowering CPC without sacrificing traffic quality means getting more conversions for the same budget.

Rewriting ROAS using these components:

This equation makes it clear: To improve ROAS, you need to increase ABS, improve CVR, or lower CPC. Let’s dive into how you can optimize each.

Strategies to Improve ROAS via Each Component

A. Increasing Average Basket Size (ABS)

Let’s take an e-commerce brand as an example. Instead of focusing on acquiring more customers, they increased revenue by maximizing how much each customer spends.

Tactics to Increase ABS

✅ Upsells & Cross-sells: Recommend complementary products (e.g., “Frequently bought together” or post-purchase upsells).

✅ Bundling: Offer product bundles at a slight discount to encourage higher spending.

✅ Free Shipping Thresholds: Set free shipping at a slightly higher value than the average order to push customers to buy more.

🛍️ Case Study: Ravin Boosting Average Basket Size (ABS)

Ravin, an online fashion retail brand, implemented Wiser’s product recommendation engine to enhance customer engagement. By displaying related and frequently bought together items on product pages, they achieved a 30% increase in conversions and a 12% increase in sales.

B. Boosting Conversion Rate (CVR)

Let’s say you’re running a subscription app. You get clicks, but many users drop off before converting. Improving CVR means turning more of those clicks into paying customers.

Tactics to Improve CVR

✅ Landing Page Optimization: Make checkout seamless, ensure fast-loading pages, and optimize for mobile.

✅ A/B Testing Offers & Creatives: Experiment with different CTAs, ad visuals, and pricing models to see what converts best.

✅ Trust Signals & Social Proof: Showcase reviews, testimonials, and security badges to reduce buyer hesitation.

🥗 Case Study: FastEasy Boosting Conversion Rate (CVR)

FastEasy, a mobile fitness app, faced challenges in efficiently conducting A/B tests for their campaigns. By integrating Reteno’s AI-powered marketing automation platform, they were able to run multiple experiments simultaneously without relying heavily on developers. This approach led to a 29% boost in conversion-to-subscription rates.

C. Lowering Cost Per Click (CPC)

A mobile gaming app wanted to scale its paid campaigns while maintaining profitability. Lowering CPC without sacrificing quality was the key.

Tactics to Reduce CPC

✅ Better Targeting: Use lookalike audiences, retargeting, and negative exclusions to focus on high-intent users.

✅ Ad Quality Score: Platforms like Google Ads reward high-relevance ads with lower CPCs. Improve ad copy, CTR, and landing page experience.

✅ Bid Optimization: Adjust bids based on performance insights such as time of day, device, and geo-targeting can significantly impact CPC.

🕹️ Case Study: Voodoo Games Lowering Cost-per-Install

Voodoo Games, a French mobile game publisher, by actively testing more than 500 videos a week for their hit game Mob Control, reduced their CPI and surpassed $200M in revenue at an average of 150% ROAS.

Final Thoughts: The Smarter Approach to ROAS

Improving ROAS isn’t just about spending more, it’s about spending smarter. By understanding the math behind ROAS and optimizing its key components — Average Basket Size (ABS), Conversion Rate (CVR), and Cost Per Click (CPC), you can make strategic, high-impact changes that drive better profitability.

At the core of all successful ad campaigns is data-driven decision-making. Testing, iterating, and refining based on real numbers, not just gut feelings, ensures that every dollar you invest works harder for you.

So, what’s your biggest challenge with ROAS? Drop a comment below, and let’s discuss! 🚀

2023 Digital Marketing Predictions

At this point, doing a 2023 prediction now seems to be cheating. I admit that predictions are hard and it probably took me longer than I should to assemble my views. While better late than never, hopefully this will spark conversation and hold us accountable for our predictions. 

AI-Powered Digital Marketing

Let’s start with an easy and obvious one – something I have already wrote about previously. Whether we like it or not, the rise of AI in Digital Marketing is upon us. 

2023 is probably the first year we see the start of real competition to Google’s Search dominance in the form of Microsoft’s new AI-powered Bing. However, this hype about AI has already transcended Search marketing, and many Ad Tech businesses are eager to incorporate AI into their products. 

With AI providing efficiency, what this means for Digital Marketers is the need to go beyond building deep technical expertise and instead focus on soft skills like problem solving, strategic thinking and creativity.

Focus on Enhancing Customer Lifetime Value

From a macro economic standpoint, 2023 is set to continue the tailwinds of a turbulent 2022. Rising interest rates, inflation and a potential recession.

With such a gray backdrop, more companies will probably prefer to be conservative with their digital marketing budgets. As such, to obtain growth in revenue, companies will need to extract higher value per user. 

Companies should therefore focus on product and monetisation to enhance their customers’ LTV. Improving LTV will also reduce the opportunity cost caused by the rising interest rates.

Apple to Extend its Digital Advertising Dominance

Since Apple released iOS 14.5, the importance of Apple Search Ads to Digital Marketers has grown drastically. This has clearly revealed Apple’s ambition in the digital advertising space.

Apple’s strengths lies predominantly in its ecosystem. With full visibility of its audiences within the iOS ecosystem, Apple is in the best position to provide personalised ads and measure its effectiveness. 

All Apple needs now, is to build its own ad exchange and demand-side platform.

Privacy Forces Transition to Probablistic Tracking

Towards the end of 2023, Google is expected to finally release its Privacy Sandbox initiative where it will reduce cross-site and cross-app tracking. This is almost equivalent to Apple’s iOS 14.5.

So, Digital Marketers should prepare for a world without deterministic tracking such as device IDs or cookies. The broad solution to this is probabilistic tracking and it is likely that advertising platforms will resort to using this. 

Tiktok to Finally Overtake Meta and Google

Let’s face it. Attention spans are dropping globally. (If you made it to this point, kudos to you!) We have been saying it for years that video as a medium is the next big thing. Specifically in 2023, short-form videos will takeover the world. To combat Tiktok, Meta and Google have both released their own versions in the form of Reels and Shorts respectively. 

It is probably still a stretch that Tiktok may actually overtake the two behemoths in 2023. But with as the fastest-growing platform dominated by youths, it is clear that the future, for now, lies in Tiktok’s hands.

In all, 2023 will definitely be another interesting year for digital marketers.

What other futures do you see yourself in 2023?

Digital Marketers: 7 Skills to Have in 2023

Recently with a little more time on my hands, I was curious to find out what companies/recruiters are looking for in Digital Marketers. But going through the many job descriptions on LinkedIn, it seemed to be a mammoth of a task, an especially menial one.

So, I decided instead to use some simple text analytics to mine out the keywords used in these job descriptions. First, using “digital marketing” as a keyword on LinkedIn, I took 30 job descriptions as the raw data. After that, I used a text mining package on R to analyse it for keyword frequency and associations.

Here’s what I found out.

Taking together the keyword frequencies and associations, provided me with a list of skills I believe would be important for Digital Marketers to have now.

1. Teamwork and Team Management

While Digital Marketing may oftentimes be thought of as an individual contributor, this is hardly the case. To run an effective digital marketing campaign, multiple cross-functional teams need to work hand-in-hand.

I will say the critical teams here would be the Performance Marketing, Creative, Data and Product Teams. Effective digital campaigns require these four teams to work together to produce data-driven potent ads that are successful in driving users to convert in the product conversion funnel.

As such, a Digital Marketer in 2023 needs to be able to work well with cross-functional teams, and/or be able to manage such teams.

2. Media Campaigns

Needless to say as digital marketers, we need to have the right technical skills to run media campaigns. However, I will go a step further here to highlight two points – the breadth of media campaigns and hands-on experience.

As we move towards a world powered by artificial intelligence (AI), Digital Marketers need to be more than single ad platform specialists. Being able to understand and manage multiple campaigns across different ad platforms will be vital.

In addition, with a looming recession, companies may be forced to tighten their resources and thus more Digital Marketers will be expected to have hands-on experience in managing the campaigns.

3. Drive Customer Growth

A huge part of what Digital Marketers do is to drive user acquisition. However, it is important to note that growing the customer base does not stop at the top of the funnel. It goes way deeper than that.

Digital Marketers need to be able to bring in quality users who will end up as valuable customers. This means focusing on bringing in relevant users, and working closely with Data and Product teams to retain customers.

4. Business Strategy

With increased help from AI, Digital Marketers need to move beyond technical expertise and towards other skill sets such as strategic thinking, creativity and problem-solving. 

At the end of the day, digital marketing needs to help the company achieve its business goals. The more strategic Digital Marketers are, the better position they will be in to formulate strategies and tactics to bring success to their company.

At this point, an astute reader might be saying “Hey Mervyn, there are only 4 skills here. What happened to the other 3?”. To that, my answer is that the remaining three are from a book I have read recently – Think Again by Adam Grant

5. Think Like a Scientist

To level up as Digital Marketers, we need to develop the habit of thinking again. Instead of simply forming opinions based on experience, come up with hypotheses, test them with data and arrive at your conclusions. 

When building up our skill set to be more strategic, we Digital Marketers should approach business strategies as experiments.

6. Abandon Best Practices

More often than not, we Digital Marketers resort or fall back to what we think are ‘best practices’. However, in doing so, we are preventing ourselves from further improving our current routines. 

What we need to do is to focus not just on results but also on the process. A bad process with a good outcome is luck, and a good process with a bad outcome might be a smart experiment. We need to adopt process accountability and continually pursue better practices.

7. Make Time to Think Again

Last but not least, we need to make time to think again. The world of Digital Marketing and Technology is ever-changing, and thus it is imperative that we set aside time for us to rethink and learn. 

Take the time to assess how much you are learning, or how much closer are you moving towards your goals. All this will help you decide on your next steps, your next experiment or your next campaign. 

Ultimately, we are rapidly moving towards a world filled with artificial intelligence and automation. Digital Marketers of today need to evolve as well to ensure that we stay relevant and continue to contribute substantially to our companies.

Are you up for the change?

AI-Powered Search Engines: 3 Ways It Will Change Digital Marketing

I got to be honest. Writing this article, I had help from ChatGPT (an AI language model). While not to the extent of it (he, she or they?) writing the article for me, but good enough proof that the future of work is indeed changing.

First, a quick premier to set the context. 

What is ChatGPT and why is it causing an AI browser war?

ChatGPT is an AI language model developed by OpenAI. Since its web interface version was released to the public in November 2022, it has taken the world by storm due to its capability to generate human-like text and its ability to perform a wide range of tasks such as answering questions, summarising text, and producing creative writing.

Most recently, Google fired the first salvo when they announced their own experimental AI chat service, Bard, which will be accessible in the upcoming weeks. A couple of days later, Microsoft (which has already invested $1 billion into OpenAI) announced that it is launching a new Bing with OpenAI’s GPT-4 model (ChatGPT was built using GPT-3 which only has data until 2021). This will provide a ChatGPT-like experience within the search engine. In addition, Microsoft is planning to include AI features in their Edge browser as well.

This kicks off the AI wars between Microsoft and Google. As we pray that this will not degenerate into an ending like The Matrix, let’s check out the three biggest impacts AI-powered search engines will have on us Digital Marketers.

Search Marketing To Take Larger Proportion of Ad Marketing Spend

With access to the user’s search history, location and behaviour, AI-powered search engines will deliver highly personalised search results and advertisements that can be used to create more effective targeted marketing campaigns. This would lead to higher conversion rates and better ROAS (return on ad spend) for digital marketers. As a result, it would be expected that our investment in Search Marketing will only grow in importance.

Furthermore, with Microsoft’s Bing returning with a vengeance like a Jedi, we will no longer be able to get away with just Google SEM (search engine marketing). Time to broaden our Search Marketing repertoire and pick up Bing advertising.

Increased Efficiency Leading to a New Breed of Digital Marketers

Imagining having JARVIS (Ironman) or Griot (Black Panther) helping you with your manual tasks. It is highly possible that AI-powered search engines may automate many of our manual tasks such as keyword research and bid management. This will free up precious time and resources and thus allow us to focus on more strategic initiatives.

This means that we as Digital Marketers will need to develop and focus on other skillsets besides technical expertise such as strategic thinking, creativity and problem-solving. 

Unique Content is Required to Stand Out

As AI-powered search engines focus more on conversational results instead of the traditional search query results, search engines will likely drive fewer clicks to your content.

In a world where AI may synthesise large amounts of data to produce a summarised answer, sites with similar content will be buried away. Therefore to stand out, an even greater importance will be placed on creating unique content. A tip here would be to add your own unique view on top of what AI services like ChatGPT provides.

In conclusion, AI-powered search engines will transform digital marketing by improving the ROI through enhanced targeting, creating a new breed of digital marketers, and forcing content creation to be more unique. 

What other ways do you think AI-powered search engines will affect us as Digital Marketers?

The Duopoly of Facebook and Google

In Economics, we learn that in an Oligopoly, the consumer suffers because of potential collusion leading to higher prices and lack of consumer choice.

Google and Facebook Rigged the Ad Market

With a combined share of ~53%, the digital ad space is currently being dominated by Google and Facebook. In such a duopoly (a special form of Oligopoly), collusion is always a possibility, and it turned out to be true when Google colluded with Facebook to favour its own exchange.

Oligopoly Trend is Not Waning

This problem of Oligopoly is definitely not going away anytime soon. It is exacerbated by further consolidation in the ad supply space (Applovin acquiring Mopub & ironSource acquiring Tapjoy) and Facebook/Google withdrawing further behind their walled gardens.

What can be Done?

Brands / Advertisers definitely have their work cut out for them. To stand out and succeed in time to come, I believe advertisers would need to focus on the following:

  1. Grow their product’s core value: Costs of digital advertising will be on the rise, and thus it will be important to provide real value to customers/users.
  2. Increase their users’ lifetime value: With the rising costs, revenues need to go up by either extending your customers’ lifetime or improving the ways you are monetising.
  3. Diversify to market disrupters like TikTok: We need to start diversifying to other digital marketing channels so that we do not continue to feed to their duopoly. It is probably easier said than done, but this is why we need more disruptors like TikTok.

The digital ad space will certainly start to feel tighter but do not mistake that for it being smaller.

We just need to grow bigger and better with it!

Attribution: The World Is Not Fair

We want a fair and just world. A world where all our marketing partners are attributed equally. And, we would like to think that is the case. Sorry to burst your fairy tale bubble, but we are certainly far from the truth.

World is unfair

In 2020, Facebook and Google will continue to rule over Digital Ad Spend land. Estimated by eMarketer, the duo will seize 61% of the US Digital Ad Spend. I guess this should not come as a surprise to many. With copious user data coupled with the smartest AI algorithms, is there any doubt as to why Facebook and Google are leading the race?

Yes, there is no doubt. But it is not only because of their superior technology and user base. Facebook and Google do not play fair.

Facebook and Google US Digital Ad Spend Share in 2020
Source: eMarketer

What is Attribution in Digital Marketing?

Alright, let me set the context straight-up first. What the heck is Attribution? To put it simply, Attribution refers to credit allocation to marketing interactions. In relation, there are two key concepts on Attribution that will be relevant here.

First up is the concept of which marketing interaction gets the credit. On the fundamental level, there are five basic methods (as illustrated below). Relevant to what we are discussing later, we can just refer to the “Last Interaction” model where the last marketing interaction gets all the credit.

Marketing Attribution Models

The second concept to note is the type of marketing interaction. Broadly speaking, there are only two – Click-through attribution and View-Through attribution. Don’t worry all this mumbo-jumbo is simpler than it sounds. It is the players like Facebook and Google which complicate it.

Click-through simply means the credit is given when the user actually clicks on an ad, whilst View-through means the credit is given when the user views the ad.

Sounds simple enough?

When a View Becomes a Click

What is a view? What is a click?

It may sound simple but when you really think about it, it is going to be borderline philosophical. Take some time and think through the following scenarios:

  • The ad image has loaded only the top 25% but the user has already scrolled past it.
  • The web page is loaded and there is a potential banner ad to be shown below-the-fold.
  • A video ad auto-plays but the user immediately pauses it.

Would any of you consider the above as a view? Here’s the kicker. The answer is yes and no. Yes according to Facebook but not according to Google. According to Appsflyer, a major Mobile Measurement Partner, Facebook considers an ad unit as a view as long as the ad unit is rendered. Even if it is not necessarily in view. And for videos, all it takes is for 1 pixel of which to appear on the screen. In contrast, Google requires at least 50% of the ad unit to be visible. The majority of the rest have pretty stringent rules too. They require 100% of interstitials and banners to be loaded before a view is counted.

Source: Appsflyer

At this moment you might think it is mighty of Google to be implementing such strict rules on itself. Don’t be rejoicing too soon. When we move on to a click, which I thought should have way less ambiguity, Google performed magic. For video ads that have been watched for 10 seconds or more, Google will transform that view to a click!

Source: Appsflyer

Impact on my Attribution Game

So what have all these funky definitions got to do with not playing fair?

Because Facebook and Google have risen to such importance to advertisers, all 3rd party partners such as Appsflyer who wish to continue partnering with them have to play by their rules, or risk being left out in the cold. In an ideal world, attribution rules should be the same for all players, and in my opinion, should be decided by an independent 3rd party.

When Facebook is allowed to consider unseen ads to be counted as a view, and Google is allowed to ‘magically’ convert a view to a click, we advertisers will constantly be playing in a world where we can never truly understand what channel works best with our customers.