And my response is always the same: “It still works.”

I tested Replit in early 2025 and it was clunky. It reminded me to the early days (circa 2021) of copywriting tool Jasper.ai (formerly Jarvis.ai) Where it needed 700 words of instruction to produce 700 words of copy.

However, over the last 6 months there has been a material acceleration in my vibe coding platform of choice. I still use ChatGPT to flesh out the initial instructions, but the output lately is phenomenal.

When I say phenomenal, I know it’s not going to win any awards for most elegant code, but it is going to do exactly what I need it to do. And as a marketer, it’s becoming one of the most valuable skill sets I can add to my portfolio.

Recently I built a simple dashboard for my ad performance. I downloaded the raw data from Meta Ads into Google Sheets, plugged Replit into it, and built a custom dashboard that shows revenue, CAC, ROAS, conversions, and campaign performance.

It took maybe four hours. Now that it’s built, it’ll take ten minutes to clone for the next client or project. Check it out here.

If I really wanted to automate it, I can use Zapier to download the Facebook Lead Ads data into a Google Sheet or AirTable, but it is fine for now.

Would a “real” BI tool do this better? Maybe… Would it cost me hundreds per month and force me to learn a system that fights me the second I want to customize something? Definitely. Would it let me ship something this fast? Nope.

This is the magic of vibe coding: The ability to get things done while simultaneously annoying every engineer I know.

Vibe coding empowers me with just enough technical skill to bend data to my will, without turning my job into a full-time engineering role. It’s the sweet spot between spreadsheets and full-stack development. And if you’re a marketer, you should learn it.

Let’s break down why.

The Case for Vibe Coding

1. You’re no longer waiting for BI teams

Marketers live under the tyranny of “we’ll put that request in the backlog.”
Need a custom field?
Need revenue broken out by device and campaign type?
Need your spend pulled in from platform B and blended with data from platform C?

Cool, see you next quarter.

With vibe coding, download the raw data or connect directly using Zapier, connect Replit, and build what you need. No support tickets, no prioritization meetings.

2. BI tools are powerful, but they’re rigid

BI tools are great, don’t get me wrong. But for marketers, we move fast. Campaigns launch, creatives rotate, tracking breaks, naming conventions get sloppy, UTM parameters mutate in the wild.

We need tools that can flex with us, not tools that we need to work around.

Vibe coding lets you patch, adjust, fix, and rebuild on the fly.

Your line charts don’t care if your code is “ugly,” they only care that the underlying logic works.

3. You’re building things that make you way more valuable

A marketer who can write a Python script, connect an API, or build a small app becomes dangerous in the best way. Suddenly you aren’t just “managing campaigns,” you’re creating infrastructure.

You’re building dashboards, attribution models, creative analysis tools, or even little internal apps that make the team faster.

You become the person who fixes problems instead of the person who waits on someone else to fix them.

4. The future of marketing portfolios is changing

Portfolios used to be for designers and engineers. Not anymore.

Marketers are now expected to show their work, the same way creators share content and designers share Figma files.

Future marketing portfolios won’t be “here are some campaigns I ran.”

They’ll be:

• here’s a dashboard I built to measure my campaign

• here’s a growth model I wrote the earned millions

• here’s a custom landing page experiment I coded and reduced CAC by 20%

• here’s a GTM system I built from end to end and made millions

Marketers who can build are going to leapfrog everyone who can only request.

The Downsides (because there are a few)

1. Your code will be messy

It won’t win any awards. It may even offend someone who uses phrases like “code is art”

2. Vibe systems can break

You’ll forget where something lives. An API key will expire. A Sheet tab name will change. You’ll swear at your laptop.

But here’s the thing. Marketers already deal with breaking systems every day. Keep yourself organized and remember that you have the power to fix it instead of waiting for help.

3. You might accidentally become the “data person”

Once people realize you can build dashboards in an afternoon, every team suddenly has “one quick request.”

Set boundaries. Or don’t. It might help your career.

Why You Should Start Now

I’m not arguing you need to become a full programmer. But you do need to understand how to move data, automate simple workflows, connect tools, and build lightweight apps.

Marketing has been moving in this direction for decades. Agencies now build tracking tools and digital products, not just ads. Vibe coding fills the gap for individual marketers

If you can write a few lines of Python, stitch together a Sheet, and deploy a small script in Replit, you can build anything you need.

A dashboard today.
A reporting engine tomorrow.
Maybe your own internal product next year.

Now, OpenAI has launched the ChatGPT Atlas browser, built on Chromium with ChatGPT integrated directly into the browsing experience. This comes right after Perplexity released its own free version of Comet.

Two AI browsers in a month! Here is what that means for marketing, advertising, and SEO, and what smart marketers can do about it.

Insight 1: Attribution and Analytics Are About to Break

How this changes things

As AI browsers gain adoption, analytics will start to lose accuracy. Users will interact with your brand through AI summaries, previews, and sidebars instead of your site.

Expect rising “Direct” traffic, fewer referral paths, and incomplete attribution. Traditional cookie and third-party tracking models are already collapsing due to regulation, and AI browsers raise the bar further for tracking and fingerprinting. Retargeting signals will weaken, and attribution models will become unreliable.

AI-generated clicks are also beginning to distort ad performance. Some AI browsers, such as ChatGPT Atlas, can autonomously mimic human behaviour by clicking ads and inflating spend without real users behind the activity. That drives up CPCs and wastes budget.

How marketers should adapt

1. Shift from click attribution to visibility attribution.
   This is nothing new, but it is more important than before. Similar to how you should be performing a lift analysis on brand spend, start tracking mentions, citations, and inclusion in AI-generated content. Visibility will matter more than it has in the past. Since many AI answers do not include links, as AI browsers evolve, I expect them to be offering view-through data that marketers can connect to their analytics platforms to understand the indirect impact of their content on conversions.

2. Use micro-engagements to capture intent.
   Embed calculators, tools, or small forms that capture leads before users leave, building an owned audience is more critical than ever.

3. Redefine success metrics.
   Your multi-channel tracking tools must now include AI visibility and brand mentions. Measure inclusion, not just traffic. Track how often your brand appears in AI answers, summaries, and citations across platforms. Replace “How many people visited this page?” with “Where and how is my brand being surfaced by AI?”

Insight 2: Marketing in a World Without Clicks

How this changes things

AI browsers are transforming how people discover and act. The path is no longer “type, scroll, click.” It is “ask, answer, done.”

Instead of sending users to your site, the AI pulls your content into its answer layer. That means fewer pageviews, less ad space, and weaker attribution models.

I no longer need to visit Ratehub for the best mortgage rates or Credit Card Genius for a list of the best credit card products.

Side note: There is no obligation for AI to be compliant with with regulation. In the example above you can see the credit card fee displayed without the APR, contrary to financial regulation.

There is opportunity in this shift. Users arriving through AI overviews are closer to conversion. They already understand the context and are ready to take action.

Research also shows a strong correlation between brand presence on highly linked pages and AI visibility. Google’s AI Overviews heavily favour content from widely cited pages. Other AI systems show weaker but measurable relationships between link authority and visibility.

How marketers should adapt

1. Add conversion points to AI-cited pages.
   If your page is being referenced by AI (you should know which ones are), make conversion frictionless. Add quote buttons, calculators, or short forms near the cited content. AI browsers can help users convert to ensure the user is already primed to act.

2. Invest in PR and brand advertising.
   Build your brand presence on authoritative, high-link pages. PR is no longer just for reputation. It directly improves AI visibility.

3. Write content worth quoting.
   Use concise headings and clear answers. Make it easy for AI systems to extract accurate, credible insights.

4. Capture first-party data early.
   Prioritize email, SMS, and push opt-ins before the AI layer interrupts your funnel. Once users get their answers from AI, you may never see them.

Insight 3: The New Ad Frontier Is AI Context

How this changes things

Ad inventory always follows attention. As AI browsers become the main interface, the new ad space will live inside AI answer panels and sidebar summaries.

This means context-based targeting will replace keyword targeting. Brands will pay to appear as trusted sources or recommended answers inside the AI experience.

How marketers should adapt

1. Be early on AI placement ads.
   When OpenAI and Perplexity open ad APIs, test them early. Almost with exception, early adopters will get cheap visibility before the rest catch up.

2. Write conversational ad copy.
   Ads will need to sound like natural, helpful responses. Write as if your brand is answering a question, not shouting a tagline.

3. Strengthen authority signals.
   AI platforms rank trusted brands higher. Build backlinks, publish verified data, and display credentials to improve authority in both human and AI ecosystems.

The Bottom Line

Atlas and Comet mark a shift from click-driven discovery to answer-driven decision-making.

Search engines indexed your content.
AI browsers interpret it.

You now need to teach the machine to trust you and make it effortless for users to act once they find you.

The marketers who adapt to this will capture high-intent, low-friction leads. The rest will keep refreshing dashboards, wondering where their clicks went.

Most B2B product marketing fails because it focuses on features and capabilities rather than the transformation.

Here’s what great product marketers understand: your buyers aren’t purchasing software, platforms, or services. They’re buying a better version of their professional reality.

The insight: The companies winning in B2B right now are those that map their entire product narrative to specific job outcomes rather than product specifications. Instead of “our platform has real-time analytics dashboards,” it’s “your CFO stops asking why the forecast was wrong.” Instead of “AI-powered workflow automation,” it’s “your team leaves at 5pm instead of 8pm.”

Why This Matters Now

1. Decision fatigue is real. B2B buyers evaluate 5-10 vendors per purchase. The one that helps them visualize their post-purchase life most clearly wins.

2. Internal selling has gotten harder. Your champion needs to sell your product to 6-8 stakeholders on average. Give them a story about transformed outcomes, not a feature list to memorize.

3. ROI is table stakes. Everyone can show ROI. But can you show emotional ROI? The feeling of not being the bottleneck? The credibility gained from accurate reporting?

The Messaging Hierarchy That Works

Structure your narrative like this:

1. First: The frustrating “before” state. Be specific. Generic pain points don’t resonate.

2. Second: The aspirational “after” state. Day-in-the-life details matter.

3. Third: The belief shift required. “This is actually possible for us.”

4. Finally: How your product enables this transition. Features as proof points.

Most B2B marketers invert this completely. They lead with product and hope buyers will imagine the outcome themselves. They won’t.

Why Sales Teams Default to Feature-Selling

Here’s the uncomfortable truth: most B2B salespeople focus on product features because it’s easier to talk about what’s in the deck than it is to actually listen to the sales call and understand how the product solves the real problem.

Features are comfortable. They’re scripted. They’re defensible in internal reviews.

The data backs this up: 59% of business buyers say most sales reps don’t take the time to understand them. Extu That’s not a training problem. That’s a prioritization problem.

Understanding the true problem requires active listening, pattern recognition across conversations, and the intellectual work of translating product capabilities into specific customer outcomes. That’s harder. It requires sales and product marketing to actually collaborate on translating real customer language into messaging that resonates.

The best product marketing teams record sales calls, extract the actual language prospects use to describe their pain, and build their “after state” narratives directly from those conversations. They don’t assume they know the problem. They listen until the real problem reveals itself.

Tools to Showcase the End Solution

If you’re serious about selling the transformation instead of the product, here are the tools that actually help:

Interactive ROI Calculators. Navattic, Reprise, or Arcade. But not the generic “save X hours per week” type. Build calculators that show the specific domino effect of solving the problem. If you’re selling recruitment software, don’t just show “time saved per hire.” Show how that time saved means hitting headcount targets on time, which means launching the product on schedule, which means hitting revenue goals.

Before/After Scenario Videos. Use Loom or Vidyard to create split-screen walkthroughs showing the same task in the prospect’s current state versus with your solution. Make these specific to different personas and use cases. A 90-second video showing a marketing ops manager’s actual daily workflow transformed is worth a thousand feature bullets.

Customer Story Libraries. Clari Copilot, Gong, or Chorus. These conversation intelligence tools help you mine actual sales calls for the language customers use when they describe their transformation. Build a searchable library of “after state” stories indexed by industry, company size, and use case. Your sales team should be able to pull the perfect customer transformation story in 30 seconds.

Interactive Demo Environments. Walnut, Demostack, or Tourial. Create sandbox environments that drop prospects directly into their “after” state. Let them experience what their day looks like with the problem solved. Skip the feature tour. Put them in the driver’s seat of their transformed workflow.

Value Realization Dashboards. Tools like Catalyst or ChurnZero aren’t just for customer success. Use them in pre-sales to show prospects exactly how you’ll measure their transformation. Show them the specific metrics you’ll track together to prove they’ve reached that “after” state.

The Bottom Line

The companies that win in B2B product marketing aren’t those with the best features. They’re the ones that make the transformed future feel inevitable, tangible, and within reach.

Stop selling what your product does. Start selling who your customer becomes.

Remember, like yesterday, when AI video was basically a party trick? You’d generate some weird, Will Smith clip where people had too many fingers and gravity was more of a suggestion than a rule. We’d all laugh, share it on Twitter, and move on.

The video models dropping right now aren’t just better. They’re fundamentally different. We’re not talking about prettier pixels. We’re talking about AI that understands how the world actually works. And that shift is going to scramble everything we know about content, trust, and what it means to “see is believing.”

What’s Actually Changed

Physics stopped being optional

Early AI video treated the laws of motion like loose guidelines. Balls would hover mid-air. Collisions would phase through each other like bad video game glitches. Now? Researchers are baking physical consistency directly into these models. Drop a ball, and it drops. Throw something, and it arcs. Bounce off a wall, and the angle makes sense.

It sounds simple, but it’s the difference between a fever dream and footage you could actually use.

Videos that don’t fall apart after 10 seconds

Most AI video until recently maxed out at maybe 15 seconds before the whole thing started to unravel. Characters would morph, backgrounds would drift, continuity would just... give up. New approaches like SynCoS are pushing that boundary way out by linking frames together more intelligently. At CVPR 2025, long-form video synthesis was everywhere. Models that can hold a scene together across minutes, not seconds.

Real-time generation is becoming real

There’s a new paradigm called “self-forcing” that’s making AI video generation faster and smoother. Instead of waiting around for a render, you’re moving toward on-the-fly synthesis. Think less “export and wait” and more “this is happening live.” That opens the door to interactive video, real-time avatars, content that responds to you in the moment.

Sound is finally part of the equation

Video without audio is a corpse. The latest models get this. Veo 3 generates clips with native sound baked in. Adobe’s rolling out tools that add or clean audio automatically, match B-roll to voiceover, optimize for different platforms. Lip sync, ambient noise, voice matching, it’s all getting bundled into the pipeline.

It’s in the tools you already use

You don’t need a server farm anymore. Midjourney lets you turn static images into video clips. Synthesia, HeyGen, Glance are giving brands and creators point-and-click access to AI video that handles scripting, localization, voice cloning, the works. What used to be a PhD thesis is now a Tuesday afternoon project.

Why Marketers Should Care

“Trust but verify” just became your full-time job

When AI video can fool your eyes, detection becomes a strategic competency. OpenAI’s Sora 2 embeds a visible watermark by default, but watermark removal tools are already circulating. Source verification, provenance tracking, cryptographic signatures... these aren’t IT problems anymore. They’re brand protection.

If your audience can’t tell what’s real, your credibility is the collateral damage.

The content flood is about to get biblical

Anyone can now generate lifelike video in minutes. That doesn’t just raise the bar, it drowns the entire playing field. Attention was already the scarcest resource. Now you’re competing with infinite synthetic content that’s “good enough” to pass.

Volume and velocity will win more battles than quality. That’s the uncomfortable truth.

Creative jobs are shifting, not disappearing

Hollywood’s not dead yet. Human judgment still matters, direction, storytelling, emotional nuance, ethical choices, the ten thousand micro-decisions that separate content from art. But the job description just changed. AI does the grunt work. You do the thinking.

The winners will be the people who figure out how to collaborate with these tools, not compete against them.

Legal chaos is coming

When AI can generate footage of anyone doing anything, copyright and personality rights become a minefield. Can you train a model on someone’s likeness without permission? Can you generate content “in the style of” a creator? What if they don’t explicitly opt out?

Sam Altman said recently that video feels way more “alive” than a static image, which means the stakes for misuse are higher. Lawsuits are warming up in the bullpen.

What to Watch

Detection and provenance tech
Watermarks, content lineage, verification layers, they’ll matter as much as the models themselves.

Regulation (eventually)
Governments are slow, but they’re coming. Expect deepfake disclosure laws, identity protection mandates, and a lot of boring committee hearings.

Hybrid workflows
The smart money is on human + AI pipelines. Let the machine handle the tedium. Keep humans in charge of taste, tension, and trust.

Real-time agents
Interactive video spokespeople. Sales bots that look you in the eye. Choose-your-own-adventure stories that adapt live. It’s closer than you think.

Ethical standards (or attempts at them)
Industry groups will try to draw lines. Who controls their likeness? Who monitors for bias and harm? Who decides what’s allowed? Good luck with that.

The Bottom Line

We’ve crossed the line. AI video isn’t a toy anymore. It’s not a demo you show at a conference. It’s media that behaves like real media: coherent, believable, physical, emotional.

That’s a huge opportunity for anyone who wants to create faster, cheaper, at scale. It’s also a huge risk for anyone who cares about truth, trust, or not getting sued into oblivion.

The question isn’t whether AI will reshape video. It’s how fast, who’s in control, and whether we’ll ever trust our eyes again.

What are you seeing in your own work with AI video? Hit reply, I’d love to hear what you’re experimenting with.

The Federal Trade Commission just proved that. Amazon agreed to pay a major settlement after regulators accused it of making it “knowingly difficult” for users to cancel Prime subscriptions. The case revealed that Amazon’s internal name for its cancel process was the “Iliad Flow.” Yes, the same Iliad that tells a story of a long, painful journey. That was their nickname for the cancellation funnel.

The FTC’s complaint outlined a list of deliberate design choices that made cancellation difficult. The “Cancel” button was buried behind unrelated pages. The wording shifted mid-process. Users had to click through multiple confirmation screens, each one worded to confuse or guilt-trip them into staying.

This wasn’t a glitch. It was intentional. Amazon’s own data showed that the more complicated the flow, the more users gave up before finishing. They treated cancellation as something to optimize against.

That’s the heart of nefarious UX: using good design skills for bad outcomes.

The new frontier of optimization

Conversion rate optimization used to be about improving performance. Now some marketers are using the same tactics, called dark patterns, to slow performance down when it hurts their metrics.

A dark pattern is a user interface design that intentionally tricks, manipulates, or pressures people into taking actions they might not otherwise take.

In other words, it’s when a website or app uses design, wording, or layout choices to benefit the business at the user’s expense.

Common examples include hiding unsubscribe or cancel buttons, pre-checking boxes for paid upgrades, using misleading colours or copy to steer users toward one choice, or creating extra friction when trying to leave a service.

The term was coined in 2010 by UX researcher Harry Brignull, who described dark patterns as “tricks used in websites and apps that make you do things that you didn’t mean to, like buying or signing up for something.”

In short, dark patterns weaponize UX to serve the company’s metrics instead of the user’s intent.

Here are some of the most common:

1. Roach Motel
   It’s easy to get in, but almost impossible to get out. You can sign up with one click, but to cancel you have to go through five steps, log in on a different device, or call customer support.

   

2. Confirm Shaming
   The cancellation button says something like “No thanks, I hate saving money.” The goal is to make you feel bad for leaving.

   

3. Misdirection
   The “Cancel” button is small, gray, or placed beside a much larger “Keep Subscription” button. Your eyes are drawn to the wrong action.

   

4. Forced Continuity
   You sign up for a free trial. The trial ends, and your card gets charged automatically. The cancel button is hidden, or doesn’t appear until you contact support.

   

5. Interface Interference
   The interface itself fights you. Pop-ups appear, buttons shift position, or entire pages are redesigned to make cancellation confusing.

   

The Amazon case showed all of these patterns at once. Regulators said it wasn’t just poor UX. It was deception through design.

Real examples from other companies

Amazon is far from alone.

HelloFresh has been criticized for hiding its cancel button behind multiple submenus. Users have to click through pages that say “pause your plan” or “skip deliveries” before they find the actual cancel option.

Uber One members report similar issues. The cancel button is hidden under “Account” → “Membership” → “Manage.” After you find it, you still have to confirm on several screens that look nothing like the main app.

DoorDash’s DashPass has been called out for burying its cancellation option in support forms. Some users have reported needing to submit tickets just to stop billing.

A 2023 research paper titled Staying at the Roach Motel analyzed hundreds of online subscription sites. More than half required users to go through multiple pages, forced them to type out “cancel,” or made them call a customer service line.

This is a new business model built on friction.

Why this matters

At first, dark UX tricks were small annoyances. A pop-up here. A pre-checked box there. But when cancellation friction becomes part of your growth model, it’s a trust problem.

When users feel trapped, they lose confidence not only in your product but in your brand.

Regulators are catching up. The FTC now treats dark patterns as a form of deceptive practice. In the Amazon case, the Commission said that design itself can be illegal if it manipulates users. Europe’s Digital Services Act includes similar language, and new privacy laws in Canada are expected to follow.

This isn’t just a design issue anymore. It’s a compliance issue.

What marketers and designers should do instead

The irony is that companies do not need nefarious UX to grow. The same skills that create friction can be used to build trust.

If you are designing user experiences, here are a few principles that I follow in my own work:

1. Mirror the sign-up process. If users can subscribe with one click, they should be able to cancel with one click.

2. Label honestly. The button should say “Cancel” if that’s what it does. Not “Continue” or “End Membership Later.”

3. Avoid emotional pressure. Do not shame or guilt users into staying.

4. Make cancellation data visible. If 40 percent of people who start cancellation never finish, that is a problem.

5. Test the offboarding flow. Every company usability-tests onboarding. Almost nobody tests cancellation. You should.

The easiest way to tell if your design is fair is to ask a neutral person to cancel their account. Watch where they get stuck. If it takes more than 30 seconds, it’s probably unethical.

The bottom line

The Amazon case marks a turning point. It proves that regulators are finally paying attention to design choices that harm consumers.

For years, marketers and designers optimized for sign-ups and conversions. Now they need to optimize for honesty and transparency.

Nefarious UX may boost retention in the short term, but it destroys lifetime value in the long run. Users who feel tricked don’t just leave. They tell everyone they know and write blog posts inspired by HeyFlow’s brutal cancellation process…

As someone who has spent nearly two decades in marketing and product, I’ve learned this lesson over and over: the best growth strategy is trust. It compounds.

By confusing your customer to stay, you’re wasting everyone’s time by failing to address the real issue: why are they leaving in the first place.

If you’re nodding your head right now, you’re not alone. The harsh reality is that most cold emails never see the light of day. They’re filtered, flagged, and forgotten before your prospect even knows you exist. But here’s the good news: it doesn’t have to be this way.

After years of testing, twea…

Read more

That’s where everyone looks.

But sometimes the biggest ROAS boost doesn’t come from optimizing your ads. It comes from fixing your tracking.

Read more

When I look back at some of my biggest wins in marketing, they did not come from flashy ad campaigns or headline-grabbing stunts. They came from something much less glamorous: retention.

Most marketers (and founders, if we are being honest) spend the bulk of their energy on acquisition. They want new customers, new subscribers, new eyeballs. I have been guilty of this myself, especially when managing large paid media budgets. There is a dopamine rush in seeing traffic graphs spike after turning on a new campaign. But here is the thing: if you cannot keep the customers you already worked so hard to get, you are pouring water into a leaky bucket.

Research backs this up. Bain & Company found that increasing customer retention by just 5% can boost profits anywhere from 25% to 95%. That is because existing customers not only buy more over time, they also cost less to serve, are more likely to refer others, and are less price sensitive. In other words, retention compounds in a way acquisition rarely does.

My Own Lessons in Retention

At Wise Publishing, where I led marketing and product growth, we had over a million subscribers on our email and push notification lists. Early on, we spent heavily on acquisition by buying ads, testing new channels, and building out SEO. That got us growth, but the real profit came when we figured out how to keep people engaged.

One of my more controversial experiments was with push notifications. We went from sending one notification a day to six. On paper, this sounds like a recipe for mass unsubscribes, and yes, we annoyed a chunk of users. But here is the reality: revenue tripled. Unsubscribes spiked, but new subscribers outpaced the losses, and the lift in revenue was worth the trade off. That experiment taught me an uncomfortable truth about retention. It is not about keeping everyone. It is about keeping the right people engaged and profitable while having the stomach to lose the ones who were never going to stick anyway.

At StorageVault, a completely different type of business, retention looked different. Self storage is often a “set it and forget it” product, but student customers were notoriously flighty. To solve this, we created a seasonal playbook that made the entire storage journey frictionless, from contest-based offers to trained front line staff who could answer questions on the spot. By reducing churn and improving renewal rates, we drove more than $2.39 million in seasonal revenue in a single campaign. That is retention in action: fewer drop offs, more renewals, bigger lifetime value.

How the Best Companies Retain Users

Retention is not one size fits all, but the best companies treat it as a core growth strategy. Here are a few playbooks I have seen work, both in my own experience and across industries:

1. Onboarding that actually works.

The first 7 to 14 days are critical for retention. Apps like Duolingo know this. They gamify onboarding with streaks, levels, and push reminders that keep users coming back daily until the habit sticks. At Wise, we leaned on a strong welcome sequence that got readers hooked on the value of our content before asking them to upgrade or engage deeper.

2. Lifecycle email and push sequences.

Retention lives and dies on whether you stay relevant. Spotify does this brilliantly with their personalized playlists, year end “Wrapped” campaigns, and re engagement emails. At Wise, personalization was not as advanced, but we used segmentation to make sure subscribers got content aligned to their interests, which boosted click through rates and kept list churn low.

3. Loyalty programs and rewards.

Starbucks is the poster child here. Their loyalty app has over 30 million members in the US alone, and it drives nearly half of their revenue. The genius is that rewards are not just about freebies, they are about creating stickiness. Customers reload gift cards, earn stars, and choose Starbucks over competitors because they are already invested.

4. Surprise and delight.

This does not have to be expensive. Zappos famously empowered customer service reps to send flowers or hand written notes when they spotted opportunities. Those small touches created lifetime loyalty. At StorageVault, we did not send flowers, but we did make sure contest winners felt celebrated and supported. That human touch built long term goodwill.

5. Community and belonging.

Some of the most powerful retention strategies tap into identity. CrossFit is not just a workout, it is a tribe. Tesla is not just a car, it is a movement. When customers feel they belong, retention becomes natural. I have seen smaller finance brands achieve this by creating exclusive communities around investing or money tips, where customers stick around because the brand is part of their daily conversations.

Why Marketers Should Rethink Their Spend

The math is simple. Acquiring a new customer can cost five to seven times more than keeping an existing one. Yet most marketing budgets still skew toward acquisition. Imagine if even 20% of that budget shifted toward retention initiatives: better onboarding, more personalized communication, smarter loyalty programs.

In my experience, the best growth does not come from doing more acquisition. It comes from balancing acquisition with retention and making sure the leaky bucket is sealed. If you can do both well, growth compounds in a way that paid campaigns alone cannot deliver.

Closing Thought

If you are a marketer reading this, here is my challenge. Audit your current strategy. How much of your time and budget is spent on chasing new customers versus keeping the ones you already have? I would bet acquisition wins by a landslide. Flip that ratio, even slightly, and you will unlock growth that feels almost unfair.

Retention is not sexy. It does not always win awards. But it wins revenue, and that is what we are really here for.

The 4 E’s are Energy, Energize, Edge, Execute. The P is Passion. When combined, they form a complete picture of what it takes to thrive as a leader. Some people naturally lean toward one or two of these traits, but the best leaders round out all five.

Personally, my favorites are Execute and Energize. Execute because great ideas are not rare. Every company is full of smart people with clever strategies. The hard part is taking an idea and turning it into measurable results. Energize because work is a team sport. I often spend more hours in the week with coworkers than I do with my own family. Keeping each other motivated, positive, and inspired makes a huge difference.

Before breaking down each of the 4 E’s and the P, let’s step back and talk about why they matter. Gallup has found that highly engaged teams deliver 21 percent greater profitability than disengaged ones. McKinsey reports that companies with decisive leaders who prioritize execution achieve 60 percent higher total shareholder returns than their peers. A Harvard Business Review study showed that CEOs who combine vision with strong execution skills are 4.5 times more likely to deliver top-quartile performance. In other words, the 4 E’s and a P are not just theory. They drive real business outcomes.

Energy

Energy is the personal drive a leader brings to the table. It is about showing up with enthusiasm, stamina, and resilience even when the workload is heavy or the market is tough. High energy leaders set the pace.

For marketers, examples of Energy include:

• Staying curious and proactive about new channels and tactics instead of waiting for directions.

• Bringing urgency to a product launch and being the one who rallies people to meet a tight deadline.

• Maintaining focus during long campaign cycles where testing and optimization might take months before showing results.

Leaders with Energy inspire others through their own actions. Deloitte research shows that 70 percent of employees who rate their managers as “high energy” also report higher productivity in their own work.

Energize

Energize is about transferring your own motivation to others. It is not enough to be personally driven. A leader must be able to create momentum across a team.

For marketers, examples of Energize include:

• Running a creative brainstorming session where everyone leaves feeling excited and clear about next steps.

• Celebrating small wins on a campaign so the team does not lose steam before the big results come in.

• Creating an environment where collaboration across sales, product, and design feels fun instead of forced.

This one resonates with me because of how much time we spend with coworkers. A positive culture is built on leaders who energize their teams. Gallup has shown that when managers actively engage employees, turnover drops by 43 percent and productivity rises. That is why Energize is just as important as Execute in my book.

Edge

Edge is the courage to make tough calls. Leadership is not just about consensus or harmony. Sometimes it means choosing between two imperfect options and standing by that choice.

For marketers, examples of Edge include:

• Deciding to cut a campaign that is underperforming even if it had strong advocates in the room.

• Picking one core brand message to double down on instead of trying to be everything to everyone.

• Saying no to shiny new channels that do not align with strategy, even if competitors are testing them.

Welch often said that Edge separates good managers from great leaders. Research by Korn Ferry found that companies led by executives with strong decision-making skills were two times more likely to outperform their peers financially.

Execute

Execute is the ability to deliver results. This is where many leaders stumble because execution requires discipline, persistence, and coordination across multiple moving parts.

For marketers, examples of Execute include:

• Delivering a lead generation program that not only launches on time but also hits the cost per acquisition target.

• Managing the rollout of a new website or app with all content, SEO, and user experience components working seamlessly.

• Tracking attribution data correctly so the team knows which channels are profitable and can adjust budgets.

Great ideas are abundant. The rare skill is execution. In fact, a study published in the Journal of Business Strategy found that 61 percent of strategic failures are due to poor execution rather than poor strategy. This is why I rank Execute as the most important of the five.

Passion

Passion is the glue that holds the rest together. Leaders who are passionate about their work inspire belief and commitment. Passion is contagious, and when employees believe their leader cares deeply, they give more of themselves.

For marketers, examples of Passion include:

• Showing genuine excitement when presenting campaign results to leadership.

• Advocating for the customer experience even when it is harder or slower to implement.

• Being vocal about the brand’s mission and values in both internal and external communications.

Deloitte found that purpose-driven organizations, which often start with passionate leaders, achieve 40 percent higher employee retention. Passion is what convinces a team to keep pushing when challenges arise.

Putting It All Together

The beauty of the 4 E’s and a P is that no one trait stands alone. Energy fuels your own stamina. Energize spreads that drive to others. Edge ensures decisions get made. Execute delivers the outcomes. Passion ties it all together and makes people want to follow you.

For marketers, these five traits are not abstract. They show up in campaign management, in cross-functional collaboration, and in the daily grind of testing, optimizing, and scaling programs. A well rounded leader practices all five, not just the one or two that come naturally.

The research is clear. Leaders who embody the 4 E’s and a P drive better outcomes for their teams and their companies. Teams are more engaged, businesses grow faster, and the work becomes more rewarding. That is why I study this model, use it weekly, and recommend it to anyone who wants to lead in their career.

Two years ago, on August 25, 2023, Google released the Helpful Content Update. It promised to reward “people-first” content. It punished low quality, SEO-first writing.

It’s the biggest update since the Panda update in 2011 and, to me, it is now obvious why. Google intends to use the SERP as a product/form focus and move publishers to the Discover and News platforms. I can only assume this is an attempt to clean up the SERP from affiliate-publisher websites like Capterra that target ‘best’ keywords but bring users to paid affiliate top-lists and fail to provide a true answer about what is ‘best’.

As a result, when I look at the SERPs today, the winners and losers are obvious.

⬆️ Products.
⬆️ Forums.
⬇️ Publishers.

Each has been impacted differently.

Products Are Winning

Product websites are doing better since the Helpful Content Update. Most SEOs know this. The numbers back it up.

In the analysis below, I analyzed the organic traffic (Ahrefs) from the most notable Canadian Personal Finance websites. I looked at September 2023 vs September 2024 and 2025. The chart is sorted by largest increase in traffic to smallest from 2023 to 2025 (2-YOY)

Banks, credit card sites, and mortgage providers are holding steady or even gaining. If you are a business selling a product or service, you are in a stronger spot than you were three years ago.

Forums Are a Mixed Bag

Forums are trickier to judge. Some exploded in traffic. Reddit is the clearest example. In 2023 it was already massive. By 2025, it’s close to a billion visits a month. Whitecollar Investor’s forum also grew year over year.

But not every forum is winning. Canadian Money Forum is down. Red Flag Deals has lost more than a third of its traffic. Forums live and die based on engagement and community. If the conversations are active, Google rewards them. If the activity dies, so does the traffic.

Publishers Are Losing

This is where the real damage is.

Publishers have not recovered from the Helpful Content Update. It has been almost three years. Traffic is still far below pre-HCU levels.

Financial Post, for example, had a great 2024. They doubled their search traffic that year. But in 2025, they lost almost half of it again. Loans Canada is flat. Moneysense is down more than 50% year over year. Forbes Advisor Canada is down over 90%. Savvy New Canadians lost almost all of their organic traffic.

According to Similarweb data, 46 of the 50 largest publishers in the world are seeing less organic traffic from Google compared to last year. The decline is so steep that publishers are starting to fight back. Penske Media Corporation, which owns Rolling Stone and Variety, has even filed a lawsuit against Google, arguing that its dominance and algorithm changes are starving publishers of the search traffic they once relied on.

Yes, a few bright spots exist. Maple Money managed a small rebound in 2025. But the overall trend is ugly.

Publishers are being pushed off the SERP.

The Shift to Discover

Where are they going? Google Discover.

I spoke with Mike Vannucci, President of Finance Buzz. He’s seeing more traffic from Discover. Many publishers are. Especially those who syndicate content through Yahoo, MSN, and other large portals.

That is the new playing field. Discover feeds are becoming the place where publishers compete for attention. It is a different game than search. It’s more like social media. Headlines matter more. Timing matters more. And the audience is far less predictable.

This is not good news for publishers. Search used to be predictable. Discover is not.

The Ranking Factor Nobody Talks About

The numbers tell the story clearly. Publisher traffic is collapsing, down 9% year over year in 2024 and then plunging another 47% in 2025. That puts publishers at a 51% loss over two years. Forums (excluding Reddit) are also sliding, down 2% in 2024 and then dropping 33% in 2025 for a total two-year decline of 35%. The only group holding strong is product websites. They grew 6% in 2024, another 10% in 2025, and are up 16% over two years. Products are thriving while publishers and forums are being erased from search.

The data also hints at something else.

To me it appears that your website category itself may be a ranking factor.

If your site is categorized as a publisher, you are swimming upstream. You can produce the best content in the world. But you won’t rank for bottom of funnel keywords if Google thinks you are a publisher. Those spots are reserved for product sites.

This is a scary thought for anyone building a media business. It suggests that your category limits your ceiling.

The Future

So where does this leave publishers?

Stuck between Google and AI.

Google is moving them off the SERP. AI is cutting them out of top of funnel. And Discover is not a stable replacement.

The only path forward for publishers might be deeper partnerships. Syndication with large platforms like Yahoo. Strategic deals with MSN. Stronger integration into communities like Reddit. Or focusing only on highly trusted, timely data that AI needs to scrape.

But the truth is, relief is not coming. Not from Google. Not from AI. Publishers are being forced to reinvent themselves.

Two years after the Helpful Content Update, the picture is clear. Products win. Forums split. Publishers lose.

And the gap is growing.

In this article I will share the results, what worked, what didn’t, and what I think local professionals should focus on instead. This applies to anyone who depends on local clients. Mortgage brokers. Insurance brokers. Accountants. Financial advisors. Lawyers. If your business lives on trust and referrals, this is for you.

Where customers come from

After years of marketing all types of services from storage units to mortgages, I’ve noticed the same pattern. People go to friends and family first. They ask around for a referral. Only after that do they turn to Google Maps.

If you are not showing up when they search, you are invisible. That is the real starting point for local marketing. Not ads. Not billboards. Google Maps.

What better way to test this than by doing the opposite? I set up a Facebook campaign, built a landing page, and pinned my hopes on breaking the mold. I chose reverse mortgages because they offered the highest commission payout, giving me more room to spend more on ads before dipping into losses.

My $2,999.89 experiment

Here are the results on my Facebook campaign.

I spent $2,999.89.
I had 609 clicks.
I paid $4.92 per click on average.
I had 55 leads.
That worked out to $54.54 per lead.

And here’s the kicker. Out of 55 leads, I had to call every single one. Most of them didn’t answer. Some hung up. Some were not serious. In the end, only one converted.

That one client did close (barely). The commission was $8,000 (It was a reverse mortgage). So on paper, the campaign was profitable. My return on ad spend (ROAS) was 2.6. That is not terrible, but very far from great. For reference, a growing business should have a ROAS of about 3-4. An established business should be around 10

The worst part was the work. Fifty-five calls for one client. That is the reality of Facebook leads. They are often people who couldn’t qualify with their friend’s broker. Then they couldn’t qualify with the first broker they found on Google. Finally, they arrive at me. The bottom of the barrel. That is why conversion rates are always low.

I’m not alone here, when I talk to other medium to large brokers, 2 to 3 percent conversion rate on paid ads is normal. oof!

What this taught me

The money was not wasted. It reinforced what not to do. Paid media can work, but it is a grind. For most local professionals, it is better to build trust where customers actually look. That means investing in organic local presence.

The two most important channels are:

1. Google Maps.

2. Local referrals.

How to win on Google Maps

Your Google Business Profile is the front door to your business. Here is how to make it work:

1. Claim and verify your profile. If you don’t, someone else will. Verification also gives you control over reviews and updates.

2. Pick the right categories. Choose “Mortgage Broker” if you are one. Add secondary categories if you also offer related services.

3. Add your location. Google wants to show local businesses near the searcher. If you can rent a desk in a real estate office or brokerage, do it. Even if it costs $100 to $1,000 a month, it is worth it. You get an address to pin on Google Maps, plus referrals from the brokerage.

   1. DO NOT say that you service an area rather than having a local address. Google Maps users want to see a pin where you located. No pin, no call.

4. Collect reviews. Ask every happy client to leave one. Reviews build trust and push you higher in search results.

5. Post updates. Share news, offers, or tips right in your profile. Google rewards activity.

6. Add photos. A profile with real photos looks alive. Add your office, your team, and even pictures of your neighborhood.

7. Track performance. Google gives you data on how many people call or click for directions. Watch it.

8. Make it look attractive. consumers will usually narrow their search down to a couple of options in their local area. The one they choose is usually the one with the most attractive photos, best reviews, and friendliest people. I know this from years of marketing Self Storage. :)

If you do this right, you can become the top listing in your area. That is better than spending thousands on ads.

To learn how I created $13 million in additional revenue on Google Maps, click here.

Build referrals locally

Ads reach strangers. Referrals reach warm leads. One of the best investments I’ve seen was renting a desk in a large brokerage. When real estate agents know you, they will refer clients. It is almost automatic.

Networking also matters. Join your local Chamber of Commerce. Sponsor a kids’ sports team. Host a free financial literacy night at the library. These things don’t just build awareness. They show you are part of the community.

Don’t forget your website

Google Maps is key, but people will also click through to your site. Make sure it is ready.

1. Use local keywords. “Mortgage broker Toronto” works better than just “mortgage broker.” Add your neighbourhood too.

2. Create service pages. Have one for “first time home buyers,” another for “refinancing,” and so on.

3. Add testimonials. Social proof helps people trust you.

4. Make it fast. People bounce if your site takes more than a couple seconds to load.

5. Add clear calls to action. Make it easy to call, email, or book an appointment.

SEO takes time (years), but it builds an engine that works for you 24/7.

Other local strategies that work

• Direct mail. Flyers still work if targeted by neighborhood.

• Local media. A story in your community newspaper builds trust.

• Events. Sponsor or host a seminar. People remember faces.

• Partnerships. Team up with accountants, lawyers, and real estate agents. Trade referrals.

Each of these has one thing in common. They build your presence where your customers already are.

The bottom line

I don’t regret spending $2,999.89. It taught me that ads can work, but they are expensive and time-consuming. For local businesses like mortgage brokers, insurance brokers, and financial advisors, the better path is clear.

Show up where your clients search. Build relationships where they live.

Google Maps and referrals should be your foundation. Ads can come later.

If you are starting a local business, remember this: people want to work with someone trusted. Be visible. Be local. Be the first name that comes up when they ask their friends or type “near me” into Google.

That is how you market a local business the right way.

Not terrible. Not great.

We made huge progress over my tenure there. One of the big initiatives, was to cut average callback times from 45 minutes to under 20. It had a massive impact on revenue and ultimately even the price we could charge for our services.

However, we still had a massive problem where the majority of callback leads weren’t picking up the phone when we called. Why? People thought we were spam.

Our outbound numbers looked anonymous. Caller ID couldn’t display our brand. Most leads assumed we were robocallers. Straight to voicemail.

This is how we fixed it.

Read more

But today the ground has shifted. The tools have changed. And the biggest players in the game are starting to feel it.

Agencies Built on Scarcity

Advertising agencies once had an advantage because the means of production were scarce. Cameras, editing bays, distribution channels, creative teams, these were expensive and difficult to access. Agencies were a way to rent the team without having to buy all the expensive production resources. If you wanted to play in the big leagues, you had to go through them.

That model worked when television was dominant, media buying was complex, and creative assets required armies of specialists. Agencies charged big retainers and took a cut of media spend. Everyone from Fortune 500 brands to scrappy startups bought into the system because there was not a viable alternative.

For reference, The most expensive commercial ever made is the 2004 Chanel No. 5 advertisement, "No. 5 The Film," which cost approximately $33 million to produce

But what happens when scarcity disappears?

The AI Earthquake

AI has blown a hole in the traditional model. We are only in the early stages and already AI can produce video that rivals what global agencies put out. Entire ads can be scripted, filmed, and edited without ever booking a studio or hiring a crew.

The Kalshi ad during the NBA Finals is a perfect example. Built entirely with AI tools for under $2,000, it hit more than 3 million views online. It looked polished. It had a concept. And it worked. Compare that with the six-figure budgets agencies still demand for similar output.

Another example came from The Original Tamale Company in Los Angeles. A mom-and-pop shop leaned on AI tools to create a quirky ad. It went viral with more than 22 million views. No agency in sight. Just AI, creativity, and hustle.

When a teenager in their mom’s basement can use off-the-shelf AI tools to make something that spreads farther than a Madison Avenue production, you know the game has changed.

Wall Street is Taking Notice

Investors have started to punish the traditional players. WPP, the world’s largest advertising holding company, reported a 71 percent drop in pre-tax profits in 2025 and slashed its dividend by half. Its stock has fallen sharply as clients pull back spending and question the agency model.

This is not just a blip. It is a sign that big agencies no longer control the levers they once did. Brands are seeing that they can get high-quality work without paying bloated fees. Investors are seeing it too.

The decline in agency stocks reflects a broader truth: the old way of working is unsustainable.

Production Value is Now a Commodity

What agencies used to sell was access to high production value. But thanks to AI, that is no longer scarce. Anyone with a laptop can spin up visuals that look like they came out of a million-dollar studio.

The difference between “professional” and “amateur” production has blurred. A TikTok ad shot in someone’s living room can drive more engagement than a polished TV spot. In fact, consumers often prefer raw, authentic content over something that feels manufactured.

Agencies that continue to pitch production as their value proposition are in trouble. Production is now a commodity. What matters is the idea, the story, and the distribution.

The New Role of Agencies

So where does this leave agencies? Dead? Not quite. But they need to evolve.

The production side of agencies is finished. AI has made high-quality content creation a commodity. What survives is everything else that makes campaigns work: strategy, distribution, and above all performance.

With the rise of affiliate marketing and performance-driven partnerships, measurement will become the lifeblood of the modern agency. Clients will not pay for hours or headcount. They will pay for outcomes. Agencies that can prove lift, track conversions, and optimize spend will thrive.

The future agency will be part strategy, part performance engine, and part distribution. Its advantage will come not from competing with AI on production, but from using AI as fuel for smarter targeting, sharper insights, and, most of all, faster iteration.

Small is the New Big

Ironically, the best agencies may start to look like they did decades ago: small, nimble, and scrappy. Think three or four people with laptops who can compete with 300-person agencies because they know how to wield AI effectively.

Portfolios are no longer just for designers, they are for all digital marketers to demonstrate how you can weld the tools to create million dollar brands on thousand dollar budgets.

We have already seen independent creators on YouTube and TikTok rival the reach of multinational campaigns. Brands will increasingly turn to these small teams and creators rather than paying for traditional agency retainers.

The industry may fragment, but it will not vanish. It will just shrink.

Lessons from the Music Industry

The situation mirrors what happened in music. For decades, recording studios controlled production. Then software like GarageBand and distribution platforms like Spotify removed barriers. Suddenly anyone could record an album in their bedroom and reach millions.

The labels that survived were not the ones who clung to the old model. They were the ones who adapted, signing new kinds of artists, investing in streaming, and finding new revenue models.

Advertising agencies face the same challenge. Adapt, or risk irrelevance.

The Human Advantage

Despite all this disruption, humans still matter. AI can generate infinite variations, but it cannot understand human nuance the way great strategists and storytellers can. Agencies that combine human insight with AI capability will thrive.

The winners will be those who know culture, who know what makes people laugh, cry, or click. AI can help produce the work, but humans will still provide the spark.

A World Without Gatekeepers

The future of advertising will be more democratic. A small restaurant, a teenager in a basement, or a niche e-commerce brand will have access to the same creative firepower once reserved for Coca-Cola and Nike.

That is both exciting and terrifying. Exciting because creativity is no longer constrained by budget. Terrifying because the volume of content will only increase. Cutting through will be harder than ever.

Which brings us back to the agencies. They are not needed as gatekeepers anymore. But they could be needed as guides.

Conclusion: Evolution, Not Extinction

The traditional agency model is dying. The days of charging $200,000 for a glossy ad are over. AI has made high production value accessible to anyone, and Wall Street is already punishing the giants for failing to adapt.

But this does not mean agencies are finished. It means they must evolve. From selling production to selling strategy. From bloated headcounts to lean AI-enabled teams. From controlling scarcity to thriving in abundance.

The new agency will be smaller, faster, and sharper. And if it can harness AI instead of fearing it, it might just survive the revolution.

The answer isn’t always more budget. It’s smarter ideas.

At Toronto’s first stock market themed bar, CKTL & Co., the prices move like a live ticker.

As guests order, select drink prices rise or fall on screens around the room, all synced to real demand. A QR…

Read more

I’ve been job hunting for the last few months. In the past, my lead rate sat around 10% and my conversion rate at 1.5%. This time it’s 5% and 0% on 300+ applications.

Oof.

I create a cover letter for every job (yes, with ChatGPT’s help). I message the hiring manager or recruiter when I can. And yet… nothing.

I spoke with a couple of recruiters last month and the job market vibes are, frankly, meh. It’s moving, but more like a polite shuffle than a sprint. Budgets are tight, hiring managers are picky, and “good enough” gets ghosted harder than a Hinge date who claims to “love walks.” Companies are not just hiring, they are hunting unicorns with laser scopes.

The unemployment rate in Ontario is now higher than during the 2008 financial crisis and is creeping toward pandemic levels.

The two things I’ve learned about this job market:

It’s a buyers market

The average unemployment rate in Toronto is normally 6.8 to 7%. With it now at 10%, employers have about 50% more candidates to choose from. Compared to pre-pandemic levels of 5 to 6%, it is closer to 100% more.

That is shocking, and it means I have about half the chance of landing a role as I did before. My numbers prove it. Historically: 10% lead rate, 1.5% conversion. This time: 374 applications, 5% lead rate, 0% conversion.

Lack of an MBA is excluding me from recruitment filters.

Employers want top performers. The best predictors of success are conscientiousness and IQ, but since companies cannot ask every candidate to take personality and IQ tests, they use education as a proxy. It is not perfect, but it is safe.

That is my problem. I do not have an MBA. My track record suggests I can outperform most, but because I lack those three letters on my résumé, I often do not even make it to the recruiter stage.

Exploring AI

Always looking to improve, I tried automating my applications with AI. After manually slogging through 30 apps in a day, I dreamed of an AI “shotgun strategy,” hundreds or even thousands of applications with minimal effort.

I narrowed it down to two platforms: jobcopilot.com and loopcv.pro. I went with Loopcv, signed up for the premium plan, uploaded my résumé, set up SMTP with my Gmail, and let it run.

The results? Brutal.

I didn’t do much of a purchase analysis. I kept it simple. I plugged both sites into Ahrefs, checked which had steady traffic over the longest period, and used that as a proxy for being more established and less buggy.

I bought the premium plan on Loopcv, setup my resume, my SMTP with the personal Gmail account, and boom. I was off.

The results left much to be desired… I hold no ill will toward any company that tries to innovate. I’m always excited to try a new tool that will make my life easier, but AI job application software has a long way to go.

The Problems

1. Applying incorrectly

Instead of using the job board or LinkedIn, it defaulted to emailing recruiters directly, often sending applications to the “accommodations” email address at the bottom of postings. Awkward.

2. Applying for the wrong jobs

After letting the AI run throughout the night, I woke up to see that I had applied for jobs I wasn’t qualified for.

• “Administrative Assistan - Security Services”

• “Vice President, Medicine, Quality and Safety”

3. It didn’t complete the application for most jobs

It seemed to find a fair amount of jobs in the 12 hours I let it run, but applied to almost none of them. It emailed a few, but as far as applying through the proper channels, it was pretty bad at it.

4. Finally, the negative keyword list isn’t realistic.

The platform lets you add negative keywords, which exclude job postings containing certain terms. The problem is I should not have to add “administrative assistant” to the list. The AI should know a VP of Marketing is not applying for that role. They are completely different jobs with different skill sets. If the expectation is that I build an exhaustive list of exclusions, the software is not really doing its job.

Was there any upside?

A little. As a job search aggregator, it is not bad. It does pull listings into one place. But as an application tool, it is nowhere near ready.

Conclusion

I was shocked at how far off the results were from what AI should be capable of. It felt like there was barely any AI at all. I should not need to hack together a giant keyword exclusion list, the software should figure that out.

Maybe I will try Jobcopilot at some point. For now, I am back to manual applications.

The four functions of management, planning, organizing, leading, and controlling, might sound like something pulled from a stuffy business textbook, but in reality, this is the framework behind almost every successful project, campaign, or business decision I’ve made. Once you understand them, you start to see them everywhere. And if you work in marketing, you quickly realize that the people who excel in their careers are the ones who execute all four functions consistently, no matter their title.

Strong managers who master these skills bring measurable results. Gallup found that talented managers drive 48% higher profit compared to average ones. Performance-management leaders see 67% better employee performance. And managers account for 70% of the variance in team engagement, an engaged team delivers more, faster, and better.

A well-rounded professional doesn’t just dabble in these functions. They plan, organize, lead, and control like second nature. And it shows… managers who can competently perform these functions, stand out from their peers.

Here’s how I see each function, why it matters, and the marketing activities that live inside each.

Planning

This is about setting the direction before you start moving. You define where you’re going, why, and how you’ll measure success. Each year for the past 10 years I’ve created an annual marketing plan which has been key in establishing direction, role clarity, and how each person can contribute to the company goals.

Marketing activities that fall under planning:

• Defining campaign goals (e.g. “Increase email subscriber base by 25% in Q2”)

• Building channel strategies (deciding how much budget goes to search, social, display, or email)

• Conducting competitive analysis to identify gaps and opportunities

• Creating audience personas and mapping the customer journey

• Forecasting campaign ROI based on historical performance

When I plan well, I avoid chasing vanity metrics and keep my campaigns focused on the numbers that matter to the business.

Organizing

Once you have the plan, you need to set up the structure to make it happen. This is where you align people, tools, and timelines so execution runs smoothly.

Marketing activities that fall under organizing:

• Assigning roles for creative, copywriting, ad buying, analytics, and reporting

• Creating campaign timelines and Gantt charts

• Setting up project management tools like Asana, Trello, or Jira for visibility

• Coordinating with external vendors, agencies, or freelancers

• Preparing the marketing tech stack (CRM integrations, ad account setups, tracking pixels)

If organizing is skipped, even the best ideas can collapse under missed deadlines, unclear ownership, or tech issues that could have been avoided.

Leading

This is the human side of management. Leading means motivating, guiding, and removing roadblocks for the people involved in executing the plan.

Marketing activities that fall under leading:

• Presenting campaign strategy to the team in a way that builds excitement and clarity

• Coaching team members on best practices for their channels

• Facilitating creative brainstorms to spark new ideas

• Resolving conflicts or misalignments between departments (e.g. creative and product)

• Keeping morale high during crunch periods with recognition, encouragement, and clear priorities

Leading well means your team not only delivers but wants to deliver. I’ve seen campaigns hit impossible deadlines because the team was united and motivated.

Bonus: I also reference Jack Welch’s 4E’s and a P. I think it helps make the leading function a little more tangible.

Controlling

Controlling is about ensuring your results match your plan. It’s measurement, analysis, and adjustment.

Marketing activities that fall under controlling:

• Monitoring KPIs like CPA, ROAS, CTR, and conversion rate in real time

• Running A/B tests on creative, landing pages, and messaging

• Auditing ad spend to ensure budget is used efficiently

• Producing post-campaign reports with actionable insights

• Implementing changes mid-campaign if metrics fall short of targets

In marketing, controlling often separates average campaigns from great ones. A 5% improvement in CTR or a drop in CPA by $50 can be the difference between scaling a campaign or shutting it down.

My Criticism of The 4 Functions of Management

One valid criticism of the four functions of management framework is that it does not explicitly call out communication, which I believe is easily half of what business administration actually is. Planning, organizing, leading, and controlling all rely on the ability to communicate clearly, consistently, and persuasively. Without strong communication, even the best plans can fall apart, teams can lose direction, and performance tracking becomes meaningless. In practice, communication is not a separate “fifth function”, it is the connective tissue that allows the other four to work.

Why I Keep Coming Back to This Model

I have led marketing teams, managed multi-million-dollar budgets, and rebuilt entire brand platforms, and I still run my work through these four functions. They’re a checklist, a framework, and a gut check all in one.

If you’re early in your career, learn these now. If you’re mid-career, sharpen them. If you’re leading a team, teach them. Because once you can plan effectively, organize resources, lead people, and control results, you’re not just “doing marketing.” You’re running it like a business. And that’s where the real impact happens.

Key Metrics

• CPA before rebuild: $400–$500

• CPA after rebuild: $100

• Conversion lift: +300%

• Landing page relevance: Matched to user location + ad intent

• Organic performance: Higher rankings for local storage terms

The Problem

When I was asked to manage the marketing department for a major self storage portfolio, the websites for all our brands were stuck in the past.

• Cluttered CTAs: On the homepage above the fold, there were six competing calls-to-action. Users didn’t know what to click.

• Inefficient paid campaigns: Most paid traffic went to the homepage or to a single store’s page, even for ads targeting other regions.

• High CPA: $400–$500 for a conversion.

• Brand perception: Sites looked dated and lacked the trust factor modern consumers expect.

• All the websites were built on different platforms. We needed to consolidate the all brands onto one single backend.

Modernize the Brand Without Losing Recognition

There was skepticism from senior leadership about changing logos, they feared customers wouldn’t recognize us.

Solution: Subtle modern updates. We refreshed the look without breaking brand recognition.

Fix the User Interface and Funnel

• Reduced above-the-fold CTAs from six to one clear, high-intent CTA.

• Designed a guided flow to funnel users from the homepage to the most relevant pages.

• Added geolocation technology to detect where a user was visiting from and automatically show facilities closest to them.

Rebuild in Django for Power and Flexibility

I hired a software engineer and rebuilt the site in Django.

Why Django?

1. Python’s growing ecosystem meant long-term scalability.

2. The ability to pull live facility data (availability, pricing, features) directly from our database.

This made the site more dynamic and conversion-focused.

Create Hyper-Local Landing Pages

We built location-specific landing pages that:

• Showed multiple facilities in a given area.

• Displayed live pricing.

• Matched the ad group’s targeting and copy.

• Optimized for both conversion rate and local SEO.

This drastically improved relevance for paid campaigns and helped us climb organic rankings.

The Results

• Online conversions up 300%.

• Paid search CPA dropped to $100 on a 1100 LTV.

• Organic traffic increased for local storage searches.

• Users spent less time searching, more time booking.

Takeaway for Growth Marketers

A website redesign isn’t about making it look pretty.

It’s about removing friction, matching intent, and structuring the funnel so that every click has a purpose.

In our case, that meant:

• Cutting cluttered CTAs.

• Using tech (geolocation + live data) to improve relevance.

• Aligning paid and organic strategies around high-intent local pages.

I developed new design mockups that highlighted our portfolio, including four self-storage brands and two portable storage brands. Incorporating updated photography and drone footage of flagship locations, the new site brought a strong visual presence to our brand family.

The content strategy emphasized our corporate history, leadership, governance, and future growth plans—striking a balance between investor transparency and brand storytelling.

The result was a modern, scalable website that continues to support StorageVault’s identity and investor relations needs.

https://www.reddit.com/r/shopify/comments/1mm4wju/i_blew_hundreds_on_sms_marketing_and_it_almost/

Owned channels like SMS and push notifications can be game changers for revenue. But they come with a cost. They will irritate people. They will cause unsubscribes. And if you do it wrong, they will damage your brand.

I’ve lived this. When I went from one push notification a day to six, revenue tripled overnight. That sounds like a win, but the unsubscribes spiked too. The lesson? Growth in owned media is never free.

The trick is knowing your limits, tracking the right metrics, and putting structure around your program so you can scale without burning your list. That’s what this guide is for.

Step 1: Define the rules

Every SMS program needs clear rules from day one.

• Pick one goal. Do you want more new orders? Repeat orders? Higher average order value? A growing list? Reviews and user content? Pick one and align every decision to it. Without a north star, your program will drift.

• Set guardrails. Decide how often you will send, when you will be silent, and how you’ll measure annoyance. For example: a maximum of two sends per week, quiet hours at night, and unsubscribes under 1% per send. These rules protect your list and your brand.

• Define your KPI. Choose the single metric that represents success. Revenue per message, conversion rate, unsub rate, or net list growth. Whatever you pick, track it relentlessly and judge all experiments against it.

Step 2: Get compliant

Compliance is not fun, but it is the foundation.

• Consent is non-negotiable. Capture explicit opt-ins through checkout boxes, popups, or keywords. Store the timestamp and the source so you can prove it.

• Identify yourself. Every message should include your brand and a clear STOP command. This builds trust and keeps you safe.

• Respect quiet hours. Nobody wants a text at midnight.

• Follow the law. In the US, register with 10DLC or toll-free verification. In Canada, follow CASL: express consent, clear identification, and simple opt-out.

Step 3: Connect your stack

Data is the fuel.

• Inputs. Feed your SMS tool with customer data, order history, and behavioral events like viewed, added to cart, and checkout started. This is what powers personalization.

• Outputs. Track every click with UTM tags and send it into your analytics. Build a dashboard to monitor sends, clicks, orders, and unsubscribes. Even a Google Sheet will work. Without this loop, you are flying blind.

Step 4: Acquire high-intent subscribers

Not all opt-ins are equal. Go for quality.

• Checkout opt-ins. This is the best place to capture people who are already engaged. Set the expectation: “Get order updates and early access.”

• On-site popups. Incentives like 10% off or first access to drops work well. Always collect email and SMS together so you can build both channels.

• Post-purchase invites. Ask people to join after they buy. Position it as membership: “Restocks and members-only drops.”

Step 5: Launch a Welcome Flow

The welcome flow is where you train your new subscribers.

• Message 1 (immediate). Thank them, deliver a code or link, and remind them how to opt out.

• Message 2 (day 2–3). Share something useful like a tip or mini guide. This shows you aren’t only here to sell.

• Message 3 (day 5–7). Introduce scarcity with a time-sensitive offer or early access. Now you’ve earned the right to sell.

Step 6: Turn on the 5 money flows

Automation beats blasts. Always.

1. Abandoned Cart. Send a reminder 30–90 minutes after abandonment, then again 20–24 hours later with social proof or low stock urgency.

2. Browse Abandon. Catch window shoppers with a note like “Still thinking about {Product}? Here’s how it fits…”

3. Post-Purchase. Confirm the order, share care tips, and then cross-sell after 7 or 30 days.

4. Win-Back. At 45–90 days of inactivity, send “We miss you” with something new or a discount code.

5. Review/UCG. After 7–14 days, ask for a rating or photo review. Sweeten it with a perk.

Step 7: Broadcast cadence

Blasts should be the exception, not the rule.

• Start slow. One per week, max two.

• Ramp carefully. Increase only if revenue per message climbs and unsubscribes stay low.

• Balance the content. Half should be value (tips, restocks, early access). The other half can be offers.

Step 8: Minimum viable segmentation

Segmentation makes messages feel relevant. Start with just a few.

• VIPs. Two or more orders or above-median lifetime value. Reward them with exclusives.

• Coupon lovers. Target them with promotions and bundles.

• Full-price buyers. Focus on quality and experience, not discounts.

• Category fans. If someone buys in one category, show them new items there.

• Dormant users. No clicks in 60 days? Snooze them or run a re-permission campaign.

Step 9: Write texts humans like

Copy makes or breaks SMS.

• One job per text. Click or reply, not both.

• Keep it tight. 130–160 characters, plain language, first name if you have it.

• Use urgency and proof. “Back today: {Bestseller} in {Color}. Last run sold out in 6h.”

• Avoid gimmicks. Shouting SALE with 10 emojis kills trust.

Step 10: Test weekly

Small tests compound into big wins.

• Try offer vs no offer.

• Morning vs evening.

• Short copy vs long.

• Emoji vs none.

• Product A vs Product B.

Call winners by revenue per message, then conversion rate, then unsub rate. Lock in, move on.

Step 11: Attribute properly

Don’t let SMS steal credit or get ignored.

• Tag every link with UTMs.

• Count “assists” where SMS drove the click but the purchase came later.

• Watch sends, clicks, orders, CVR, RPM, unsub %, complaints, and net list growth.

• Run the math:

  • RPM = Revenue ÷ Messages Sent

  • ROI% = (Revenue – Discounts – SMS Cost – Ops Costs) ÷ SMS Cost

Step 12: Frequency and friction controls

Protect your list.

• Add a kill switch: pause if unsub >2% or complaints spike.

• Snooze anyone inactive for 60 days.

• Send “Still want texts?” check-ins.

• Always honor STOP instantly.

Step 13: Scale without torching the list

Growth comes from discipline.

• Offer text-only perks like early links and surprise drops.

• Coordinate with email and push. Don’t triple-ping people in one hour.

• Clean your list often. Remove hard bounces and keep carrier trust high.

Step 14: 30-day launch plan

• Week 1. Get compliant. Turn on opt-ins, Welcome, and Abandoned Cart.

• Week 2. Add Post-Purchase, Reviews, and your first VIP broadcast.

• Week 3. Add Browse Abandon, Win-Back, and a second broadcast.

• Week 4. Tune the winners, adjust frequency, and publish your first SMS report.

Benchmarks: What “Good” Looks Like

• SMS conversion rate. 21–30% overall. 11–20% in ecommerce. Higher in tech.

• Push CTR. 2–5% on average. Android usually higher than iOS.

• Email CTR. 1.7–2.0% with opens in the 30–40% range. Automations like cart, welcome, and browse flows deliver most of the orders.

Takeaway: SMS usually wins on conversion, push wins on speed, and email wins long-term. Ignore vendor averages. Focus on your own RPM, unsubscribes, and list growth.

Free Templates (steal these)

Abandoned Cart
“{First}, you left {Product}. Want us to hold it for an hour? Finish here: {link} –{Brand} (Reply STOP to opt out)”

Back in Stock
“It’s back: {Product} in {Color}. Last run sold out in 6h. Early link: {link} –{Brand}”

Win-Back
“We saved you 10% on your next {Category}—valid 48h. Code: BACKAGAIN → {link} –{Brand}”

Value Tip
“Pro tip for {Product}: {1-line care/use hack}. Full guide: {link} –{Brand}”

Conclusion

SMS is powerful because it cuts straight through the noise. But that power is easy to abuse. If you treat it like a firehose, you’ll burn your list and your brand. If you treat it with respect, you’ll build a reliable revenue stream that complements your email and push programs.

Start small, follow the rules, track the right numbers, and scale carefully. The brands that win with SMS are the ones that think long-term.Free templates (steal these)

Sources:

https://www.reddit.com/r/shopify/comments/1mm4wju/i_blew_hundreds_on_sms_marketing_and_it_almost/

https://simpletexting.com/sms-marketing/benchmarks/conversion-rates-cvr/?utm_source=chatgpt.com

https://clevertap.com/blog/push-notification-metrics-ctr-open-rate/

https://www.businessofapps.com/marketplace/push-notifications/research/push-notifications-statistics/

https://www.airship.com/blog/a-marketers-guide-to-push-notification-benchmarks/

https://mailchimp.com/resources/email-marketing-benchmarks/

https://www.mailerlite.com/blog/compare-your-email-performance-metrics-industry-benchmarks

https://www.omnisend.com/2025-ecommerce-marketing-report/

This blog is for the curious marketers, the ones who thrive in uncharted territory, testing bold ideas before anyone else dares to try them. No recycled case studies from 2010. No hollow “growth hacks” that sound good on LinkedIn but die in execution. Just battle-tested strategies you can put into play immediately — from campaigns that made headlines to experiments that quietly drove millions.

If you’re ready to think differently, break things (strategically), and create the kind of marketing that gets people talking, you’re in the right place.

I’m Kris Bruynson, a marketer who’s managed somewhere between 60-80 million in ad spend, launched campaigns that have driven millions in revenue, and tested just about every tactic you can imagine (plus a few I’d rather forget). Along the way, I’ve learned what works, what doesn’t, and what looks great in a pitch deck but crashes in the real world. This blog is where I share the hard-earned lessons, playbooks, and unconventional ideas that actually move the needle, so you can skip the costly experiments and get straight to results.