Back to Blog

    11 Years of Growth Hacking: Why AI Changes Everything

    tom-stromJanuary 15, 2025

    In 2014, I got my first "real" growth hacking job. My title was "Growth Lead" (which meant I was the only person working on growth), and my toolkit consisted of:

    • Excel
    • Google Analytics
    • A $500/month ad budget
    • Whatever scrappy tactics I could think of

    I spent most of my time manually analyzing data, setting up campaigns, running A/B tests, and trying to squeeze every possible conversion out of tiny budgets.

    It was exhausting, exhilarating, and honestly, kind of fun.

    Eleven years later, I'm building an AI company that automates most of what I used to do manually. And I keep asking myself: what does "growth hacking" even mean anymore?

    The Early Days: Hustle Over Everything

    When I started in growth, the community was obsessed with "hacks"—clever tricks to get disproportionate results with minimal resources.

    Some classics from my early playbook:

    • Scraping competitor review sites to find potential customers
    • Manually messaging hundreds of people daily on LinkedIn
    • A/B testing 50 different landing page variants
    • Reverse-engineering competitor ad strategies
    • Building fake waitlist demand to create FOMO

    Did these work? Sometimes. Were they scalable? Not really. Did they teach me invaluable lessons about user psychology and experimentation? Absolutely.

    The skill wasn't in having sophisticated tools. It was in being resourceful, analytical, and willing to try weird things.

    The mindset: Growth hacking was about doing things that didn't scale to figure out what worked, then finding ways to scale it.

    That mindset hasn't changed. But the tools absolutely have.

    The Data Awakening (2016-2018)

    Around 2016-2017, the industry matured. Companies started hiring "growth teams" instead of lone growth hackers. Tools got better. Data became more accessible.

    I joined Campanja and suddenly had access to:

    • Real ad budgets (six figures, not hundreds)
    • Proper analytics infrastructure
    • A team to execute ideas
    • Sophisticated attribution modeling

    The game shifted from "clever hacks" to "systematic experimentation." We'd run dozens of tests simultaneously, analyze results rigorously, and compound learnings over time.

    The companies winning at growth were the ones with:

    • Better data infrastructure
    • Faster experiment velocity
    • More sophisticated analytical capabilities
    • Smarter prioritization frameworks

    It was less about individual genius and more about systematic excellence.

    The challenge: This approach required resources. You needed engineers to build tracking, analysts to interpret data, specialists to run campaigns. Growth became expensive.

    Small companies couldn't compete with large companies' resources. The "scrappy startup beats giant corporation" stories got rarer.

    The Automation Wave (2018-2020)

    Then automation tools started getting really good.

    Platforms like Facebook and Google introduced automated bidding. Email tools added AI-powered send time optimization. Analytics platforms built automatic anomaly detection.

    Suddenly, a lot of what we did manually was getting commoditized:

    • Bid optimization: automated
    • Audience targeting: automated
    • Budget allocation: automated
    • Timing optimization: automated

    My first reaction: This is great! We can focus on strategy instead of execution.

    My second reaction: Wait, if the tools do all the optimization, what's my actual value?

    This was an uncomfortable question. I'd built my career on being good at tactical optimization. If machines could do it better and faster, what was left?

    The answer, I realized, was the things machines couldn't do:

    • Understanding deep customer psychology
    • Crafting compelling positioning
    • Identifying new channels and opportunities
    • Designing experiments to test strategic hypotheses
    • Making judgment calls with incomplete information

    But here's what bothered me: those skills were harder to demonstrate. It's easy to show "I improved ROAS by 30%." It's harder to show "I helped the team think more strategically about positioning."

    The nature of growth work was changing, and not everyone was ready for it.

    The Plateau Period (2020-2022)

    For a couple years, things felt stagnant. The tools were good but not revolutionary. The strategies were well-known. Growth felt more like optimization than innovation.

    The playbook for B2B SaaS growth was basically:

    • Content marketing for SEO
    • Paid ads on Google and LinkedIn
    • Product-led growth if possible
    • Sales outreach for enterprise
    • Optimize, optimize, optimize

    Everyone was doing the same things. Winning meant executing slightly better than competitors, not doing fundamentally different things.

    The insight from this period: Growth was becoming more about compounding advantages—better data, faster iteration, smarter team—than breakthrough tactics.

    The "hack" mentality was dead. Professional, systematic growth was the new standard.

    Then AI Actually Arrived (2023-Present)

    I started experimenting with GPT-3 in 2022, but it felt like a toy. Interesting, sure, but not transformative for actual growth work.

    GPT-4 in early 2023 changed my mind completely.

    For the first time, I could have a genuine analytical conversation with an AI:

    • "Why did conversion rate drop last week?"
    • "What audience segments are underperforming?"
    • "How should we reallocate budget across channels?"

    And get thoughtful, contextual answers in seconds.

    The moment I knew everything had changed:

    I was analyzing campaign performance for a client. Normally, this would take me 2-3 hours:

    • Pull data from multiple sources
    • Build pivot tables
    • Look for patterns
    • Form hypotheses
    • Test hypotheses
    • Write up findings

    Instead, I fed the data to GPT-4 and had a conversation:

    Me: "Analyze this campaign data and tell me what's working and what isn't."

    GPT-4: "Performance is strong overall, but I notice a significant drop in conversion rate for mobile users in the 25-34 age range. This started approximately two weeks ago. The landing page load time for mobile increased by 40% in that timeframe. I'd recommend checking for recent changes to mobile experience."

    That would have taken me at least an hour to discover. The AI found it in seconds.

    And it wasn't just faster. It caught a pattern I probably would have missed because I wasn't specifically looking at mobile load times.

    That's when I realized: The nature of growth work is fundamentally changing again.

    What AI Actually Changes

    After spending the last two years building AI tools for marketing, here's what I think AI fundamentally changes about growth:

    1. Analysis Speed

    What used to take hours now takes seconds. This isn't just convenience—it changes what's possible.

    You can now:

    • Test dozens of hypotheses in an afternoon
    • Analyze performance daily instead of weekly
    • Catch issues in real-time instead of post-mortem
    • Explore more experimental approaches without huge time investment

    The implication: Iteration velocity becomes the key competitive advantage. The team that learns fastest wins.

    2. Democratization of Sophistication

    Advanced analytical techniques used to require specialized skills:

    • Cohort analysis
    • Attribution modeling
    • Statistical significance testing
    • Segmentation analysis

    Now? You describe what you want in plain language, and AI handles the technical implementation.

    The implication: Small teams can operate with capabilities that previously required large, specialized teams.

    3. Idea Generation at Scale

    I used to spend hours brainstorming growth experiments. Now I can generate hundreds of ideas in minutes, then use AI to help evaluate and prioritize them.

    Is every AI-generated idea brilliant? No. But volume matters. Ten mediocre ideas that you actually test beat one brilliant idea you never get around to implementing.

    The implication: Ideation is no longer the bottleneck. Execution is.

    4. Personalization Without Code

    Creating personalized experiences used to require engineering resources. Now you can prototype personalization with AI, test if it works, and only then invest in proper implementation.

    The implication: The barrier to testing sophisticated growth tactics has collapsed.

    5. Learning Curve Compression

    I spent years developing intuition about what works in growth. Now someone starting today can shortcut a lot of that learning by working with AI that has seen thousands of case studies and can suggest approaches based on similar situations.

    The implication: Experience matters less; learning agility matters more.

    What Hasn't Changed (And Won't)

    Despite all the AI capabilities, some things remain fundamentally human:

    1. Strategic Judgment

    AI can analyze data and suggest optimizations. It can't tell you if you're building something people actually want or if you're in the right market.

    The strategic questions—who are we serving, why do they care, how do we win—remain human decisions.

    2. Creative Differentiation

    AI can generate copy, design variants, and test approaches. But genuinely creative positioning that makes you stand out? That's still human territory.

    Everyone will have access to similar AI tools. Differentiation will come from strategic creativity, not tactical execution.

    3. Cross-Functional Influence

    Growth requires convincing product to build features, engineering to prioritize infrastructure, sales to try new approaches.

    That's relationship building and organizational navigation. AI doesn't help with office politics.

    4. Ethical Judgment

    AI will suggest tactics that work but might be questionable:

    • Aggressive retargeting that feels stalker-ish
    • Dark patterns that boost conversion but harm trust
    • Data practices that are legal but creepy

    Deciding where the line is remains a human responsibility.

    5. Long-Term Vision

    AI optimizes for measurable outcomes. It can't tell you that you should invest in brand building even though it hurts short-term metrics.

    Strategic patience and conviction remain human strengths.

    What This Means for Growth People

    If you're in growth, here's my honest assessment of how AI changes your work:

    Skills That Are Getting Commoditized

    • Manual data analysis
    • Routine optimization
    • Campaign execution
    • Performance reporting
    • Basic A/B testing

    If your primary value is doing these tasks, you're in trouble. AI will do them better, faster, and cheaper.

    Skills That Are Getting More Valuable

    • Strategic thinking (what should we optimize for?)
    • Creative positioning (why should people care?)
    • Experimental design (what should we test?)
    • Cross-functional collaboration (how do we align the org?)
    • Ethical judgment (should we do this just because we can?)
    • AI fluency (how do we leverage these tools effectively?)

    The New Growth Hacker Profile

    In 2014, a good growth hacker was:

    • Analytically strong
    • Tactically creative
    • Willing to hustle
    • Comfortable with basic technical tools

    In 2025, a good growth person is:

    • Strategically creative
    • AI-augmented
    • Focused on high-leverage decisions
    • Comfortable working alongside AI agents

    The shift: From hands-on execution to orchestration. From doing the work to directing AI that does the work.

    The Uncomfortable Truth

    I need to be honest about something: not everyone will make this transition successfully.

    Some people in growth love the tactical execution. They like running campaigns, analyzing spreadsheets, and optimizing conversions. That work is deeply satisfying.

    AI is taking that work away. Not entirely, but increasingly.

    If you don't find joy in the strategic, creative, and collaborative aspects of growth, this transition will be painful.

    I've watched talented growth people struggle with this. They're excellent at the tactics that are being automated, but they haven't developed the strategic muscles that are becoming essential.

    There's no judgment in this. People have different strengths. But the industry is changing whether we like it or not.

    How I'm Adapting

    Building Cogny is partly me future-proofing my own career. I'm betting that the future of growth is:

    Less about:

    • Manual analysis
    • Routine optimization
    • Tactical execution

    More about:

    • Strategic direction
    • Creative differentiation
    • AI orchestration

    Instead of analyzing data myself, I'm building AI that analyzes data better than I could.

    Instead of running campaigns manually, I'm creating agents that can execute based on strategic direction.

    Instead of optimizing tactics, I'm focusing on what to optimize for and why.

    The meta-skill: Learning to work effectively with AI agents. Prompting them well, interpreting their outputs critically, combining their speed with human judgment.

    Advice for Growth People

    If you're in growth and wondering how to navigate this shift, here's what I'd suggest:

    1. Start Using AI Tools Daily

    Don't wait until AI is "ready" or perfect. Start using Claude, GPT-4, or Cogny today. Build fluency with AI assistance.

    The people who will thrive are the ones who get comfortable working with AI now, not the ones who wait until it's mandatory.

    2. Shift Focus to Strategy

    Spend less time on tactical execution, more time on strategic questions:

    • Are we in the right market?
    • Is our positioning compelling?
    • What experiments would change our trajectory?
    • How do we build defensible advantages?

    Let AI handle "how do we optimize this campaign?" Focus on "what should our growth strategy be?"

    3. Develop Cross-Functional Skills

    Growth is increasingly about coordination, not just execution. Work on:

    • Influencing product roadmap
    • Aligning sales and marketing
    • Building relationships across the org
    • Communicating strategy to leadership

    These skills don't automate.

    4. Get Comfortable with Less Certainty

    AI enables faster iteration but also more experimentation in uncertain areas. You'll be making more bets with less complete information.

    The mindset shift: from "I need to be right" to "I need to learn quickly."

    5. Build in Public

    Share what you're learning. Document experiments. Teach others.

    As tactical execution commoditizes, your ability to think clearly and communicate effectively becomes more valuable. Build that muscle.

    6. Don't Lose the Hustle

    AI makes sophisticated tactics accessible, but it doesn't replace scrappy creativity.

    The best growth people will combine AI capabilities with hustle—using AI to move faster and try more things, not as an excuse to do less.

    What Excites Me Most

    Despite the uncertainty and disruption, I'm more excited about growth now than I've been in years.

    Why?

    Because the barriers are dropping. You don't need a huge team, massive budget, or years of experience to try sophisticated growth tactics.

    A solo founder can now:

    • Analyze campaign performance like an expert
    • Generate and test positioning variants
    • Personalize user experiences
    • Optimize across complex funnels
    • Learn from thousands of case studies

    The playing field is leveling. Scrappy startups can compete with giant corporations again—not through manual hustle alone, but through AI-augmented creativity and speed.

    This is the most interesting time to be in growth since I started.

    The people who embrace AI, develop strategic skills, and maintain creative hustle will have incredible leverage.

    The people who resist change or rely only on tactical execution will struggle.

    The Future I'm Building

    At Cogny, we're betting on a future where:

    • Small teams operate with enterprise-level capabilities
    • Analysis happens in seconds, not hours
    • Experiments run constantly, not quarterly
    • Learning compounds faster than ever
    • Strategic creativity becomes the key differentiator

    We're not building AI to replace growth people. We're building AI to make growth people superhuman.

    The vision: A growth team of three operating with the capabilities of a team of thirty. Not by working harder, but by leveraging AI to handle everything that doesn't require uniquely human judgment.

    One Last Thought

    Eleven years ago, I started in growth because I loved the combination of creativity, analytics, and impact. I loved finding clever ways to drive results with limited resources.

    AI hasn't changed what I love about this work. It's changed how I do it.

    I still get to be creative, analytical, and impactful. But now I'm augmented by tools that handle the tedious parts and let me focus on the interesting challenges.

    If you'd told 2014 Tom that in 2025 he'd be working alongside AI agents that could analyze data, generate insights, and suggest experiments in seconds, he would have been ecstatic.

    This is the future we wanted. We just didn't realize how much it would change along the way.

    The fundamentals remain: understand your users, test ruthlessly, learn quickly, and always be looking for the edge that makes you different.

    The tools change. The mindset doesn't.

    And that's exactly why I'm so excited about what comes next.

    ---

    About Tom Ström

    Tom is CEO and co-founder of Cogny, where he's building AI-powered marketing automation. He's spent over 11 years in growth hacking and marketing technology, from scrappy startup tactics to enterprise AI platforms. Previously, he co-founded Campanja and built optimization systems for Netflix, Zalando, and other major brands. He's passionate about making sophisticated growth tactics accessible to teams of any size.

    Ready to level up your growth game?

    Cogny brings AI-powered analysis and optimization directly into your workflow. Stop spending hours on manual analysis and start focusing on strategic growth. Book a demo to see how AI can augment your growth team.