Nobody says this part out loud. You can’t out-execute a competitor on their own axis. They’ve been climbing that hill longer than you have. Compete where they’re strong, and you lose slow.

Real differentiation is a refusal. You look at the axis everyone grades themselves on, and you walk off it. You build a new one where your weirdness is the whole point.

In my framework that’s the eighth domino: New Opportunity. Stop competing in the category. Replace it. Three brands did exactly that. The move, the result, and the part you can steal.

Liquid Death: they sold death in a water aisle

The bottled-water category had one axis. Purity. Mountains, glaciers, a woman doing yoga, “untouched by humans since the Pleistocene.” Everyone competed on who looked the cleanest.

Mike Cessario walked off that hill in late 2018. He put still water in a 16.9-ounce aluminum tallboy that looks exactly like a tallboy of beer, slapped a skull on it, and called it Liquid Death. The tagline isn’t “pure.” It’s “Murder Your Thirst.”

That is a healthy product marketed like an unhealthy one. On purpose. He wasn’t selling hydration. He was selling something to hold at a party when you don’t want to drink and don’t want to explain why. The category sold water. He sold an identity.

The result isn’t a vibe. It’s a number. The first launch video cost about $1,500 and pulled over three million views before the product was even for sale. By March 2024 the company had raised funding at a $1.4 billion valuation. Reported revenue ran around $333 million for 2024, up from roughly $3 million in 2019.

Water. The most undifferentiated product on Earth. A billion-dollar brand because one founder refused to compete on purity.

The lesson: your category has a default axis everyone grades on. Name it. Then ask what you’d build if that axis didn’t exist.

Duolingo: they made the mascot a menace

Language apps competed on outcomes. Fluency in three months. Speak like a local. Land the job. Every one of them, the same earnest promise on the same anxious axis.

Duolingo stopped promising fluency to strangers and started entertaining them instead. They took Duo, the green owl that nags you to practice, and turned him loose on social. Twerking on the office desk. Thirsting over celebrities. Threatening you, lovingly, if you skipped your Spanish. They even staged the owl’s death as a campaign and watched the internet lose its mind.

It reads like a stunt. It’s a moat. A mascot nobody can copy. Disney can’t license Duo. Babbel can’t reskin him. The competition can match your features in a sprint and your price in an afternoon. They cannot become an unhinged bird the world already has a relationship with. That’s the fifth domino: a mechanism so specific to you it can’t be replicated.

The chaos sat on a real engine. By the third quarter of 2024, Duolingo reported 37.2 million daily active users, up 54% on the year. Full-year revenue landed near $748 million, up roughly 41%. The owl with 16 million-plus followers wasn’t the joke. It was the top of the funnel.

The lesson: features get copied. Price gets matched. A character the market is attached to gets neither.

{{newsletter}}

Avis: they admitted they were losing

Both of those are recent. So is the instinct to say “sure, but that only works for punk water and meme owls.” It doesn’t. The cleanest version of this move is sixty years old, and it ran in a category as dull as it gets. Renting cars.

In 1962, Avis had trailed Hertz forever. Hertz held roughly 61% of the market to Avis’s 29%, and Avis had lost money for thirteen straight years. The obvious play is to claim you’re just as good as the leader. To stand on his hill and insist you belong there.

Doyle Dane Bernbach did the opposite. The campaign opened with a confession: “Avis is only No. 2 in rent a cars. So why go with us? We try harder.” They didn’t hide the weakness. They led with it, and turned it into the reason to choose them. The little guy sweats the details the giant got lazy about.

A company voluntarily admitting it’s losing is so strange, and so obviously true, that everything it says next feels honest too. That’s the seventh domino: embracing limitations until the weakness becomes the proof. Within a year Avis went from losing $3.2 million to earning $1.2 million, its first profit in over a decade. By 1966 the gap had closed to 49 against 36.

The lesson: the flaw you’re hiding is often the most believable thing you have. Say it first, on your terms, and watch it buy back every other claim.

The pattern under all three

Punk water. A deranged owl. A rental car that admitted it was second. Three different decades, three categories nobody calls exciting, one identical move.

None of them got better at the thing the category already rewarded. Liquid Death didn’t make purer water. Duolingo didn’t promise faster fluency. Avis didn’t out-Hertz Hertz. Each one found the axis everyone was crowded onto and stepped off it. Health product, sold like a vice. A brand built on a character instead of a feature. A weakness used as the headline.

That’s the whole discipline. Differentiation is not a louder version of your competitor’s pitch. It’s the decision to be graded on something they can’t win.

So before you polish another value prop, run the audit. What axis is your whole category competing on? Where is everyone standing? Then find the hill they left empty, and go plant your flag in it.

There’s no spell here. There never is. Just the nerve to stop competing and start replacing.

That era is over.

Reviews are no longer decoration on your profile. They’re a measured signal feeding the local algorithm, and increasingly the raw material an AI chews up and reads aloud to a customer who will never see your website. Get this wrong and you don’t just lose stars. You lose the room.

What reviews do now

Start with the weather. In BrightLocal’s 2026 Local Consumer Review Survey, 97% of consumers read reviews for local businesses. Eighty-five percent say a positive set makes them more likely to use you. This part you knew.

Then the shift. Reviews now carry real algorithmic weight. Whitespark’s 2026 Local Search Ranking Factors report puts review signals at roughly 16% of what decides the local pack, and that share keeps climbing year over year. Not a tiebreaker. A factor.

So reviews do two jobs at once. They persuade the human. And they rank the business. Most owners optimize for the first and forget the second.

Velocity beats volume

The instinct is to chase a big number. Two hundred reviews. A wall of five stars.

Wrong target.

Sterling Sky’s 2025 study found the ranking lift shows up early, around ten reviews, then flattens fast. What keeps you visible after that isn’t the total. It’s the flow. A steady drip of recent reviews tells Google you’re alive and busy right now.

Recency is the part nobody schedules. In that same BrightLocal survey, 74% of consumers only trust reviews from the last three months. A glowing testimonial from 2023 reads like a closed restaurant.

So stop sprinting to a number. Build a system that produces three or four honest reviews a month, forever. Boring. Compounding. The only thing that works.

The trap that gets you penalized

Good intentions die right here. You want more good reviews, so you screen for them. Happy customers get the Google link. Unhappy ones get a private feedback form.

That’s review gating. Google prohibits it outright.

And it’s not just Google policing this anymore. The FTC’s rule on fake reviews, in effect since late 2024, bans suppressing negative reviews and dangling incentives for positive ones, with penalties that scale per violation. No discounts for stars. No filtering by mood. No staff posting from the back office.

The compliant version is almost insultingly simple. Ask everyone. Ask fast. Hand them a direct link or a QR code while the experience is still warm. Then get out of the way.

{{newsletter}}

Reply like a human, on the clock

Responses are their own signal now. In BrightLocal’s data, 80% of consumers are more likely to use a business that responds to every review. Whitespark tracks response rate as a ranking input. Two birds.

But templated replies backfire. Copy-paste “Thanks for your feedback!” forty times and people read it as exactly what it is. Read each review. Reply like a person who was there.

Negative reviews aren’t fires to put out. They’re auditions. A calm, specific, human reply to a one-star rant does more for the stranger reading it than for the angry customer who left it. You’re not writing to the critic. You’re writing to the next hundred people deciding whether you can be trusted when something goes wrong.

The AI is reading them for you

This is the shift almost nobody has internalized.

Google now generates AI summaries of your reviews, stitched together by Gemini and stamped “Summarized with Gemini”. It pulls out sentiment and attributes and hands a customer the verdict before they read a single full review. In BrightLocal’s survey, 82% of consumers already read those AI summaries. Twenty-three percent will decide on the summary alone.

The space is shrinking too. Sterling Sky reports AI-powered local results often surface one or two businesses where the old pack showed three. Fewer slots. A machine picking who fills them, partly off the language in your reviews.

So the words matter, not just the stars. When a customer writes “best plumber in Mesa for emergency repairs,” they’re feeding the exact phrasing the AI will use to recommend you. You can’t write that yourself. That’s gaming the system. But you can ask the right question at the right moment: what did we fix, and where. Specific service, specific place, in their words. Then let the machine find it.

The whole spell

No shortcut here.

Reviews reward the business that does the unglamorous thing on repeat. Earn them honestly. Earn them often. Reply like you mean it. Let the language stay human, because the algorithm and the AI both reward what real people actually say.

Build that engine now. When the next customer asks an AI who to call, you want to be the name it already trusts.

Go fix your review system this week. Not the star count. The system.

Someone searches “plumber near me.” Google answers with three businesses in a box, a row of stars, and a map pin. That box gets the click. Your beautiful homepage sits two scrolls down, waiting for traffic that already decided.

That box is fed by your Google Business Profile. Not your website. Your profile.

First, the name changed. Then the app died.

If you learned this as “Google My Business,” you learned the old name. On November 4, 2021, Google renamed it Google Business Profile. Same listing, new label.

Then they killed the app. The standalone Google My Business app is gone. You manage your profile straight from Google Search and Google Maps now: search your own business name while signed in as the owner, and the edit controls appear right there.

Google also pruned hard. As of July 31, 2024, the chat, messaging, and call-history features were removed. The free GBP-built websites shut down. Q&A is being phased out. If your playbook still leans on any of those, it’s running on a dead engine.

What Google actually ranks

Google sorts local results on three things. Relevance, distance, prominence. How well you match the search, how close you are, how trusted you are.

You can’t move your building closer to every searcher. So you win on the other two. And the lever is bigger than most owners think. Across the local pack, Google Business Profile signals carry about 32% of the ranking weight. Reviews another 20%. Your website, 15%.

Read that again. The profile and the reviews outweigh the site roughly three to one for that map box.

So work the profile like the storefront it is:

• Nail the primary category. It’s the single strongest local-pack factor there is. “Emergency plumber,” not “home services.” Be specific or be invisible.
• Get your name, address, and phone identical everywhere. Same format on your site, your profile, every directory. Sloppy data confuses the algorithm and the customer.
• Fill every field. Hours, services, products, attributes, photos. An empty profile is a closed sign.
• Post like the place is open. Updates, offers, photos. A profile last touched in 2023 reads as a business that might be gone.

Reviews are the engine. Recency is the fuel.

Reviews are 20% of the local pack, and most people work them wrong. They beg for a pile of stars once, then stop.

Google cares less about your total and more about your pace. A shop earning two or three honest reviews a month, answered thoughtfully, often outranks the place with 400 reviews and nothing since spring. Volume is a trophy. Velocity is a signal.

So build a habit, not a campaign. Ask every happy customer the day the job’s done. Reply to all of them, the good and the ugly. A calm, specific response to a bad review sells harder than a wall of five stars, because the next reader is watching how you handle it.

{{newsletter}}

The map box isn’t the only door anymore

The ground shifted under everyone’s feet. Google now answers a huge share of local questions with an AI Overview before the map even loads.

The split depends on what’s asked. For a ready-to-buy search like “tacos San Francisco,” the local pack still shows over 90% of the time and AI Overviews only about 15%. But for a research question like “how long does an eye exam take” or “average cost of dental implants in Phoenix,” AI Overviews appear more than 90% of the time. That data is Whitespark’s, Q2 2025.

The pattern is simple. Bottom of the funnel, the map wins. Top of the funnel, the AI answers, and your profile alone doesn’t get you quoted.

AI Overviews pull from your website, from Yelp and Reddit and forums, from review platforms, from anywhere your name shows up with substance. So the business that answers the real questions on its own pages, and gets mentioned across the web, is the one the AI repeats. The business with a tidy profile and a thin website gets skipped.

A perfect Google Business Profile is table stakes now. It is not the whole game.

The wizard’s confession

No trick to it.

Local SEO is just being the most obvious, most trusted answer in your town, in two places at once. The map box, where you win on category, accuracy, and a steady drip of fresh reviews. And the AI answer, where you win by being genuinely useful on your own site and talked about off it.

Claim the profile. Fix the category. Earn a review this week. Then write the page that answers the question your customer types before they ever search your name.

Be the answer. Not an option.

His marketing was never the problem. His name, address, and phone number were.

That’s NAP. Name, Address, Phone. The same three fields on your site, your Google Business Profile, and every directory that ever scraped you. When they match everywhere, Google believes you’re one real business. When they don’t, it can’t tell which version is true. So it trusts you less. So you rank lower.

Not a hack. The cost of admission.

What changed, and what didn’t

For years the advice was “build citations.” Get listed on 200 directories, watch your rankings climb. Then around 2022 the SEO crowd started asking whether citations even mattered anymore. The needle had moved. Google had gotten better at understanding businesses without a wall of directory links.

They were half right. As a ranking lever, citations got weaker. In BrightLocal’s Local Search Ranking Factors, your Google Business Profile now carries about 32% of local pack weight. Citations? Around 6%. If you’re hunting for the thing that moves you up the pack, it isn’t a directory blast.

Then the twist. The weak ranking lever became a hard trust gate. Same source: “with the explosion of AI over the last few years citations only seem to be becoming more important in business visibility.” Inconsistent NAP doesn’t just fail to help. It bleeds trust. A 30-location brand with messy data has a dozen storefronts arguing with each other across the web.

So the old tactic died and the signal underneath got stronger. Both things are true.

Google Business Profile is your source of truth

Stop thinking of your listings as equal. They aren’t.

Your Google Business Profile is the master record. It feeds the map pack, powers “near me,” and increasingly tells the AI engines who to quote when someone asks for the best plumber in town. Every other listing exists to agree with your profile. When a directory says something different, it isn’t a second opinion. It’s a contradiction Google has to resolve, and contradictions cost you.

So the order of operations is simple. Lock the profile first. Name exactly as it reads on your signage. One address, formatted one way. One phone number, the local one, not a tracking line that changes. Then make everything else match that. Not the other way around.

Your website comes next. Same NAP in the footer, on the contact page, in the schema markup. If your own site disagrees with your own profile, you’ve lost the argument before a directory ever weighs in.

{{newsletter}}

AI search raised the stakes

This is the part the 2020 playbooks miss.

Google’s local results are mutating in front of us. Sterling Sky’s 2026 data, pulled from 179 profiles across 34 law firms, shows AI-powered local packs surfacing only about 32% as many businesses as the old three-pack. One or two names instead of three. Often no call button at all. Clicks straight from the profile are sliding, worst on mobile.

Fewer slots, higher bar. The AI choosing who fills those slots reads your identity off the open web. Clean, consistent NAP and you look like one confident business it can cite. Scattered NAP and you look like noise. Noise doesn’t get quoted.

The customer-side cost is just as blunt. BrightLocal found 80% of people lose trust in a business with inconsistent contact details. A wrong phone number isn’t a ranking problem then. It’s a person standing outside a door that moved two years ago, calling a line that rings nowhere.

What to fix this week

You don’t need a six-figure tool stack. You need an afternoon and some discipline.

Audit first. Search your business name, your old names, your phone number. Write down every listing and how each one spells your NAP. The deviations will surprise you.

Fix the profile, then the website, then the big directories: Google, Bing, Apple, Yelp, Facebook, the two or three that own your industry. Skip the long tail of junk directories. They were never the point.

Then put it on the calendar. NAP drifts. You move, you change numbers, a data aggregator resurrects an address you killed. The locksmith didn’t break his listings on purpose. Time did. A quarterly look keeps the rot out.

No spell here. NAP consistency is just refusing to lie to the machine about who you are and where to find you. Get the boring three fields right, and every clever thing you do next finally counts.

Go pull your listings. You won’t like what’s there.

That isn’t strategy. That’s a tax you pay every quarter.

The fix isn’t picking a side. It’s wiring the two together so one feeds the other. And in 2026, with AI eating the top of the search page, integration stopped being a nice-to-have. It’s how you keep your clicks at all.

The page you’re fighting over is shrinking

Nobody put this on the roadmap. Google’s AI Overviews now answer the question before anyone scrolls. A pre-registered field study from the Indian School of Business and Carnegie Mellon found that when an AI Overview shows up, organic clicks on that query drop 38%. Zero-click searches jumped from 54% to 72%. Same satisfaction scores. Just fewer clicks reaching your site.

Then there’s AI Mode. One billion monthly users inside a year, with queries doubling every quarter. People type longer, weirder, more specific questions now. The ten blue links you optimized for a decade are becoming a footnote under a paragraph Google wrote.

The easy organic click is dying. Not dead. Dying. Pretending otherwise is how you lose a year.

Why two weapons beat one

So you lean harder on paid? Only if you think paid is free traffic you’d have gotten anyway. It isn’t.

Google ran the experiment. Over 400 studies on accounts that paused their search ads, then watched what organic picked up. The answer: 89% of the clicks from search ads were not recovered by organic. Pause the ads, and that traffic mostly vanishes. It does not politely migrate to your blue link below.

Read it the other way. Your organic ranking does not make your ad redundant. Your ad does not make your ranking redundant. Two listings on one page take more real estate, catch more intent, and survive the moment AI swallows one of the two. Presence is the strategy. Redundancy is the point.

Let each channel pay the other’s bills

This is where integration stops being a slogan and becomes a workflow.

Paid funds the SEO roadmap. Your ad search-terms report is the fastest keyword research on earth. It tells you which queries actually convert, with real money attached, in days instead of the six months an organic test takes. Mine it. Then build organic pages for the winners and stop guessing.

SEO trims the paid bill. Once a page ranks on its own, you stop renting that click forever. Shift the spend to the terms you can’t win organically, or to the new product with no ranking history yet. That’s the launch move: paid buys the front page on day one while SEO compounds underneath for month twelve.

Both sharpen the message. A landing page that converts paid traffic converts organic traffic too. Test the headline with ad dollars. Roll the winner everywhere.

{{newsletter}}

The new prize: getting cited, not just ranked

The 2026 wrinkle changes the math.

When an AI Overview appears and you’re cited inside it, Seer Interactive’s analysis of 53 brands and 5.47 million queries found you earn 120% more organic clicks than the pages that AI ignored. Being cited still trails not having an AI Overview at all, by 38%. But the gap between cited and uncited is the whole ballgame now. Get named in the summary, or get buried under it.

Paid plays defense here. On the queries where AI is strangling organic clicks, your ad is sometimes the only listing of yours a buyer sees above the fold. Seer’s data shows paid CTR held far steadier through the AI Overview rollout than organic did. So you cover the organic losses with paid on AI-heavy terms, and chase pure-organic clicks on the queries where no AI Overview shows up yet. One budget. Two fronts. Allocated on purpose.

Run it as one machine, not two reports

Google built the controls for this. Performance Max finally added campaign-level negative keywords and channel reporting in 2025, plus account-level brand exclusions. That matters for integration: it stops your automated campaigns from quietly buying the branded clicks you already win for free, so paid spends on demand you’d otherwise miss.

Then measure both channels in one view. Not an SEO dashboard and a PPC dashboard that never speak. One report, by query, showing organic position, ad spend, AI Overview presence, and what each combination actually converted. When you can see that a term has no organic shot but cheap, profitable paid clicks, you stop arguing about budgets and start moving them.

There is no spell here. There never is.

SEO and PPC together is just the discipline of treating one search result as one battlefield instead of two line items. Mine the paid data. Earn the citation. Cover the gaps. Stop paying the tax of two teams who refuse to talk.

Wire them together. Then go own the page before AI finishes rewriting it.

There’s one metric that does predict it. Almost no one tracks it. I call it Share of Brand Voice: how often your brand gets mentioned, searched, and cited in your category, measured against everyone else fighting for the same wallet.

Not your reach. Your share of the conversation.

Why the conversation beats the click

The math is uncomfortable. In any given quarter, only about 5% of your potential buyers are in the market. The other 95% aren’t ready, aren’t looking, and won’t sign today no matter how clever your funnel is. That’s the Ehrenberg-Bass Institute’s “95-5 rule,” and it has held up across category after category.

Performance marketing fights over the 5%. It’s a knife fight in a phone booth. Everyone bidding on the same in-market keywords, watching their margins bleed out.

Share of Brand Voice is how you win the other 95% long before they ever enter the room.

Anyone can buy likes, hearts, shares, and views. Share of voice is whether they remember your name when the money is finally on the table.

The pattern nobody wants to hear

Decades of advertising data point the same direction. Brands whose share of voice runs ahead of their share of market tend to grow. Brands that let it slip behind tend to shrink. Binet and Field even put numbers on the split: roughly 60% of budget into building memory, 40% into capturing demand that already exists.

Most founders run it backwards. All activation. No memory. They wonder why every lead is expensive and every quarter starts from zero.

Memory compounds. Activation doesn’t.

{{newsletter}}

How to actually measure it

You don’t need an enterprise license or a six-figure brand study. You need three honest inputs, tracked over time:

• Branded search volume. How many people type your name into Google, not just your category. Pull it from Google Search Console and a keyword tool. Rising branded search is the cleanest demand signal there is.
• Mentions and citations. Where your name shows up without you paying for it: podcasts, posts, roundups, other people’s content. Tools like SparkToro make this visible.
• Visibility versus named rivals. Pick three to five competitors. Track the same numbers for them. Share is a comparison or it’s nothing.

Pick the three you can pull this week. A rough number you watch beats a perfect number you never check.

The wizard’s confession

There’s no spell here. There never is.

Share of Brand Voice is just the discipline of being remembered on purpose: showing up, telling the truth, and saying your own name often enough that the market learns it before it needs you. Build that memory now, and when the buying window opens, you aren’t a search result fighting nine others. You’re the name they already trusted.

Start counting the metric your competitors ignore. Then go be the loudest honest voice in your category.

Start with the numbers, because the numbers are the whole lesson.

I had 400 followers after my first 30 days on TikTok. A full month of showing up twice a day, talking marketing into a phone, for a few hundred people.

Around day 34, one video broke out. It pushed me past 10,000. By May 2024 I was at 70,000. Today it’s 110,000+.

Everyone wants the day-34 video. Almost no one wants the 33 days before it. That gap is the whole game.

Why twice a day beats one perfect post

Volume bought the breakout. You can’t predict which video pops, so the move is more shots and a sharper craft on every one.

I posted twice a day. Hundreds of videos. Most did nothing. A handful did everything. The ones that worked taught me the hook, the pacing, and the on-screen text the next one needed. You don’t find your voice by planning it. You find it on rep two hundred.

One perfect video is a lottery ticket. A daily cadence is a printing press.

I treated TikTok like a search engine

The algorithm wasn’t the only engine. Search was.

About 1 in 4 TikTok sessions begins with a search, so I stopped writing captions for vibes and started writing them for the words people actually type. Hook, caption, on-screen text, all aimed at a real query: “how to market a local business,” “TikTok SEO,” “fast Google indexing.” A video that ranks for a search keeps pulling months after the feed forgets it.

Most creators optimize for the For You page. I optimized for the search bar too. That’s where the compounding lives.

Followers are vanity. Leads are the point.

The line I’ll die on: anyone can get likes, hearts, shares, and views. None of that is the scoreboard.

The 110,000 only mattered because they turned into over 10 leads a week for my agency, and took the business international. A follower you can’t convert is applause. A follower who books a call is revenue. I built the audience to move the second number, not the first.

{{newsletter}}

The playbook, stripped down

If you want to run this yourself, it’s four moves and zero magic:

1. Pick one platform and post twice a day for at least 30 days before you judge it.
2. Write every caption for a search, not just the feed. Hook and on-screen text included.
3. Make hundreds, not dozens. Treat each video as a rep, not a referendum.
4. Measure leads, not likes. If the audience never moves a business number, the audience is a hobby.

The uncomfortable part

There was no growth hack. There never is.

It was 34 boring days, one break, and the discipline to keep posting after it. The people who “blow up on TikTok” almost always have a quiet month you never saw. Most quit inside that month, at 400 followers, one good week before the curve bends.

Pick your platform. Post twice a day for a month before you decide it doesn’t work. Then come find me on TikTok and tell me what broke out.