AMI Ads

I’ve spent over $100k/day on affiliate traffic this year. Here’s what actually works (and what nobody teaching “Meta hacks” will tell you). I’ve been in affiliate for years now — sweeps, leadgen, casino, ecom funnels — and our campaigns consistently push 7–8 figures a month across multiple verticals. And I’m tired of watching new affiliates get wrecked by outdated strategies from 2019. So here’s what’s actually printing right now in 2025: Cost caps are undefeated. If you’re still raising budgets manually like it’s the golden era of ABO, you’re fighting the machine. At scale, cost caps handle volatility better than any human. My days of staring at dashboards ended the moment I realized Meta respects consistency more than “clever” optimizations. I used to stress when spending $500/day. Now I’ll push $100k+ and barely check the account some days. Why? Cost caps + clean signals = the algorithm does the heavy lifting. Stop testing with separate campaigns. This is where every beginner nukes their accounts. They build giant “testing structures” with 12–20 adsets and wonder why nothing spends. The truth? Meta knows if an ad will win within 100 impressions. Add new creatives directly into your proven adsets. If Meta doesn’t give them budget, it already decided they won’t beat your currents. Trust the machine. Creative volume is the entire game now. We push 80–150 new creatives per week across verticals. Not because we’re hunting unicorns — but because Andromeda literally rewards pattern diversity. New hooks. New scroll-stops. New angles. New funnels. The affiliates stuck at $3–10k/day are usually running the same 4–5 ads from last month. AI is a cheat code for speed. I’ve launched full affiliate funnels (prelanders, advertorials, hero images, UGC-style videos) for under $300 total. No studio shoots. No “creators.” No $2k UGC packages. Just AI. I’ve validated entire offers in 48 hours with nothing but AI creatives + agency accounts. Is it perfect? No. But it gets you to the truth faster. Here’s the part nobody wants to admit: You don’t need $20–50k to scale. You need: • a god-tier offer • relentless creative volume • a funnel that sends clean signals • and a setup that works with Andromeda, not against it If you’re stuck right now, your bottleneck is almost always one of three things: 1. your offer isn’t strong enough 2. your creative volume is way too low 3. your signals are confusing Meta instead of guiding it The game didn’t get harder. It got more predictable. And the affiliates who adapt fastest are eating right now. If you want the exact systems we use to scale $100k/day campaigns — including our creative volume engine, cost-cap structure, signal setup, and AI workflow — comment BLUEPRINT and I’ll send you the full affiliate scaling manual. (must like + follow)

@sasasokas CASINO

$214k in casino deposits from one creative. And it didn’t even look like a casino ad. no jackpots no casino lobby no fake winners Just a slot machine screaming while players keep spinning underneath. The whole ad explains the dopamine loop in 2 seconds. ChatGPT → Object POV MakeUGC → generate Scenes → animate Creative cost: $1 Scaled 10× overnight. We break down these creatives inside our private casino traffic group. Comment "CASINO" and I’ll send the invite.

@yllevan2 Still going hardddd bro

cpi and freecash are genuinely so cooked right now been seeing lower epcs than normal and cant scale that much anymore time to switch verticals…

Meta just rolled out “pay to remove ads”. From an auto insurance buyer POV, this actually matters more than people think. If the users most likely to pay for ad-free are: – higher income – more privacy aware – more intentional with purchases then what’s left in the auction? Lower-intent traffic. More impulsive users. More noise. Which raises real questions for insurance advertisers: – Does CPL creep up because the “clean” audience exits? – Does lead quality get more volatile? – Does Meta lean harder on thinner, less reliable signals? – Do compliance-heavy verticals like auto insurance feel this first? This also makes one thing unavoidable: Pixel manipulation becomes more important, not less. When high-intent users disappear from the auction, you can’t afford to let your pixel learn from every lead anymore. If you keep feeding Meta: – low-quality submits – unqualified drivers – price-shopping junk leads the algorithm will optimize straight into the ground. Filtering what conversions you send back becomes a survival skill: – only firing profitable leads – only training on real downstream value – shaping the audience Meta thinks “works” This accelerates a split that was already happening: People who pay with money vs people who pay with attention For auto insurance, that likely means: – more reliance on advertorials + education – heavier pre-qualification before the form – first-party data and postbacks actually mattering – pixel feedback control becoming part of the media buying stack This isn’t an overnight collapse. But long term, anyone running lazy leadgen and blindly firing every conversion is going to feel this first.

The Avatar Advantage (what auto insurance brands refuse to believe) Most “audience research” in auto insurance is fake work. I’ve seen the same carrier offer go from: $62 CPL → $14 CPL without changing traffic, funnel, or payout. Same offer. Same states. Same budget. 4x more profitable. What actually changed: Demographics = maybe 5% Avatar depth = everything Not “drivers 25–54” Not “low credit” Not “high intent” But: – how they describe their last rate increase – what pissed them off about their insurer – what they’re scared will happen if they switch – the exact words they use when they complain Those words become: – your hooks – your advertorial headlines – your form copy – your follow-up logic One avatar. Run for 12 months. Month 1 vs month 12 isn’t optimization. It’s compounding. By month 12: – CPL drops – conversion rate stabilizes – creatives last longer – buyers trust the traffic more Most auto insurance brands die because they rotate avatars every 2 weeks instead of going deep on one. Depth beats breadth. Every time.

everyone in affiliate land has the same fantasy plan (not really) build an offer scale it to $3m–$10m/yr sell it for 4–6x disappear cool story except most auto insurance businesses are worth exactly $0 on exit and here’s why nobody buys revenue they buy systems that print without you read that again if you stop working and the leads stop coming you don’t own a business you own a stressful job with ad accounts here’s what most auto insurance affiliates actually do – find one carrier or aggregator – scale it on meta or native – hit $250k–$400k/month in spend – screenshot dashboards – start talking about “exits” on paper it looks good example math: $3.6m annual gross 20% net after ops = $720k profit at 4x = $2.9m exit sounds decent then due diligence starts and everything falls apart red flag 1: founder dependency you’re still the hub – you talk to the buyers – you manage traffic – you approve creatives – you handle compliance issues – you fix postback breaks if you disappear for 30 days, revenue drops 40–60% no buyer is paying millions for that risk red flag 2: single-channel exposure 80–90% of leads come from one source meta, taboola, or google buyers don’t care that “it’s stable right now” they care that: – cpl doubled once already – compliance rules change quarterly – accounts get flagged randomly one channel = one failure point that kills multiples fast red flag 3: no moat your “business” is: – the same forms – the same carriers – the same angles – the same landers nothing stops someone from cloning you in 2 weeks buyers don’t acquire things that can be recreated for $50k red flag 4: messy numbers – revshare mixed with cpl – carrier clawbacks not normalized – no clean cohort reporting – no real LTV by source – personal expenses in the business to a buyer this screams: “we don’t trust your EBITDA” discount or walk red flag 5: rented customers 90% of revenue is new leads almost zero backend no email monetization no retargeting depth no cross-offer logic you’re renting traffic not owning relationships buyers want predictability not roulette what actually sells in auto insurance the groups that exit at 5x–8x aren’t louder they’re boring and structured they have: 1. documented systems lead flow, QA, compliance, scaling rules traffic runs without the founder touching buttons 2. diversified traffic meta + native + search + email monetization one source drops, revenue doesn’t collapse 3. buyer leverage multiple carriers per state negotiated floors volume-based revshare uplifts buyers pay for bargaining power 4. repeat monetization policy renewals cross-sells (auto → home → renters) email and sms doing 20–30% of monthly rev this is where multiples come from 5. clean books carrier-level P&Ls true net after clawbacks clear CAC → LTV math no guesswork 6. real team media buyers ops compliance buyer management the founder is optional that’s the asset 7. runway new states new buyer types new traffic sources buyers want upside not a capped machine here’s the part nobody likes you cannot bolt this on later you can’t wake up at $300k/month and say “ok now let’s make it sellable” buyers see through that immediately the affiliates who exit well built like they were selling from day one they sacrificed: – fast screenshots – sloppy scaling – fragile wins in exchange for: – durability – leverage – real enterprise value ironically those are also the guys who end up making more money long-term anyway if you want a real exit in auto insurance stop thinking like a media buyer start thinking like an operator who might sell this thing one day

Unpopular opinion in auto insurance media buying (from a 50M a year media buyer): you don’t have to choose between Broad or Cost Caps. Here’s the 70/30 split we use to scale auto insurance without nuking the account. There’s a quiet war going on in auto insurance buying. Camp 1: Broad-only buyers “Trust the algo. No targeting. Lowest cost. Let Meta figure it out.” Camp 2: Cost-cap diehards “Never trust Zuck. Cap everything. Force the CPL.” Both are half right. Both are incomplete. Broad is for volume. Cost caps are for control. Auto insurance needs both. This is the exact structure we run on auto insurance accounts spending $2k–$5k/day to smooth volatility while still scaling. The 70/30 split 70% of budget → Broad (Lowest Cost) 30% of budget → Cost Caps Why this works in auto insurance. The engine: Broad (70%) Goal: scale + data. Auto insurance needs volume to train the pixel across driver types, states, credit bands, risk profiles. Broad finds new pockets: – different driver behaviors – unexpected state-level demand – cheaper segments Meta won’t find if you choke delivery If you only run cost caps, you eventually stall. Learning never stabilizes and spend gets throttled. Broad is how you expand. The insurance policy: Cost Caps (30%) Goal: margin + stability. We take the top-performing creatives from Broad (actual approvals, not just CTR) and move them into a separate campaign with cost caps. Setup example: Target CPL: $40 Cost cap: $45–$50 What happens: On bad days (high CPMs, junk traffic, bad auctions): – Broad may bleed – Cost cap barely spends On good days: – Cost cap wakes up – Snipes the cheapest, highest-intent drivers – Pulls blended CPL down This matters in auto insurance where volatility kills confidence fast. The result: blended performance Broad gives you scale. Cost caps protect the floor. You don’t need to be right every day. You just need to not die on bad ones. Quick setup Campaign A – Scaling CBO Broad Lowest Cost 3–5 proven auto insurance creatives Open budget Campaign B – Sniper ABO Top 1–2 creatives only Cost cap at break-even High budget (it won’t overspend anyway) Stop being religious about bidding strategies. Broad finds drivers. Cost caps protect profit. If you want the exact creatives, hooks, and offers we use to make this structure work in auto insurance: Reply “AUTO” and I’ll send one of these (must like + follow)

here’s how we’re scaling auto insurance info + lead systems to $50k+/mo without hitting the creative wall most people stall for the same reason they find one angle that works milk it for 1–2 weeks then performance drops and they’re back to “what hook do we test next?” we hit that ceiling early when we were doing ~$12–15k/mo the problem wasn’t traffic it was creative throughput we started testing AI creative systems specifically for auto insurance and leadgen most tools were useless generic hooks no understanding of compliance, intent, or payout ceilings then we rebuilt the process around one rule: AI doesn’t create strategy, it scales proven strategy here’s what changed everything for us instead of asking AI to “make ads”, we feed it structure: – the offer (carrier type, payout range, states) – the avatar (high-risk, seniors, parents w/ teens, low credit, etc.) – the angles that already convert (rate shock, denial fear, overpayment math) – the constraints (compliance, claims language, disclosures) from there the system does one job only: multiply what already works our internal flow now looks like this: 1. take 1 winning ad (image + copy) 2. extract the core elements: – persona – emotional trigger – promise – CTA 3. generate 20–50 controlled variations: – different driver situations – different state-level framing – different math anchors ($87/mo vs $1,042/yr) 4. push all outputs into Airtable with: – angle tag – persona tag – state tag – CPL / approval tracking we’re pushing 80–120 new creatives/week per offer without guessing without brainstorming without turning media buyers into copywriters hard numbers: – creative output: +6x – time per concept: down from ~2 hours to ~15 minutes – CPL variance tightened by ~22% – fewer account-level shocks because we’re rotating intent, not just visuals the biggest unlock wasn’t “better AI” it was removing ourselves as the bottleneck auto insurance is not about clever copy it’s about volume + precision if you can consistently produce: – compliant – persona-specific – intent-aligned creatives you don’t cap at $10k/mo you scale until buyer demand taps out most teams never get there because they can’t ship fast enough

Just hit your first $10k day in auto insurance with ripped ads and think you made it? You’re about to fall back to $0 next month if you don’t fix this immediately. Here’s the exact hiring + systems order we use once an auto insurance account starts scaling. Customer support is your first hire Not when it gets overwhelming. Tomorrow. $6–8/hour from the Philippines gets you someone who handles: – inbound lead questions – duplicate submissions – obvious junk leads – basic CRM notes This alone improves approval rates by ~10–20% and reduces buyer scrubs. Next move: put them in a group chat with buyers Slack / Telegram with: – lead buyers – account managers – tech / tracking Now quality issues get solved same day instead of showing up as clawbacks 30 days later. This prevents: – buyer pauses – payout drops – “quality suddenly declined” emails Time saved: 10–15 hours/week Money saved: $5k–$20k/month in avoided scrubs. Next problem: creative risk If you’re scaling auto insurance on ripped creatives only, you’re on borrowed time. What happens every time: – winner gets flagged – domain burns – account resets – spend drops overnight Solution: Hire a static editor + short-form video editor. Rates: – static editor: $5–8/hour – video editor: $100–150/week Their job: – turn ripped concepts into original formats – localize by state / persona – produce 3–5 new ads per week minimum Run ripped + custom in parallel. Never wait for something to die before testing backups. Final step: own the offer layer Pure arbitrage caps fast in auto insurance. To break past it you need: – direct buyer relationships – custom payout tiers – exclusive flows or routing That’s how you move from: $10k days → $30k weeks → $1M+ months in spend. The unsexy stuff: – support – QA – creative systems – buyer communication adds more long-term value than any new “angle” ever will. This is how auto insurance operators stop resetting every 30 days and start compounding.

After 9 years of managing auto insurance Facebook ads, I’ve seen more funnels collapse overnight than I can count. This is the exact process I use to bring performance back without blowing up the account. Most auto insurance buyers panic the moment CPLs spike or approvals drop. I see it constantly on audits and calls. One bad day and people start pausing campaigns, duplicating ad sets, changing budgets, launching 10 new creatives at once. After managing millions in auto insurance spend, here’s the truth: Sudden drops are normal. Bad reactions are what kill accounts. The problem is almost never the drop itself. It’s the emotional response to it. Why auto insurance campaigns collapse suddenly Most performance drops have nothing to do with “broken ads” or Meta suddenly hating you. The most common causes I see in auto insurance: – Day-of-week behavior shifts – Pay cycles and bill timing – Seasonal intent changes (renewals, tickets, rate hikes) – Creative fatigue inside narrow driver personas – Delivery randomness during optimization – Backend issues (approval delays, call center lag, routing problems) None of these mean your funnel is dead. They mean behavior changed temporarily. The mindset before touching anything One bad day in auto insurance means nothing. I’ve seen accounts go from: $38 CPL → $72 CPL → $41 CPL in the span of four days with zero changes. People rebuild entire accounts after a single bad afternoon. That’s how you erase months of learning. Before touching Ads Manager, the goal is not to “fix.” The goal is to understand. The troubleshooting process I actually use A. Zoom out to 3–5 days The first thing I do is look at the last 3–5 days, not the last 6 hours. I want to see: – Is the decline gradual or sudden? – Is it consistent or volatile? – Is it isolated to one campaign or account-wide? One bad day = noise. A trend = signal. I’ve had auto insurance accounts hit zero approved leads one day and 7x ROAS two days later without touching a thing. B. Check external factors before Ads Manager Next, I check everything outside Meta: – Is it a weekend? – Is it the first of the month? – Are renewals peaking or dropping? – Did call center hours change? – Any approval or underwriting delays? Auto insurance is extremely sensitive to timing and ops. Ads can be fine while approvals tank. C. Check frequency and creative fatigue If CPLs rise slowly while frequency climbs, it’s usually fatigue. In auto insurance this happens fast, especially inside: – High-risk drivers – Seniors – Low-credit personas The fix is rarely a full rebuild. Usually it’s: – Rotating in a variation – Changing the opening frame – Resetting delivery without changing the offer Small moves beat big resets. D. Check where Meta is delivering If creative is fine, I look at audience delivery. For interest targeting: – Audience may be too tight – Budget outpaced audience size For Advantage+: – Meta may have shifted into a lower-intent pocket This is common after short-term volatility. You want to confirm you’re still reaching fresh drivers, not hammering the same exhausted cohort. E. Soft resets, not nuclear ones If performance still doesn’t recover, I use soft resets: – Duplicate the campaign – Keep the same ads – Re-introduce proven structures – Restart delivery without changing the core logic This resets Meta’s path without wiping learning. Full rebuilds are almost never necessary in auto insurance unless the structure itself is broken. What not to do when auto insurance results drop These are the fastest ways to make it worse: – Pausing everything instantly – Launching too many new ads at once – Making large budget swings – Rebuilding before identifying the cause This creates algorithm chaos and prolongs the drop. Sometimes the correct move really is: Do nothing. Let it stabilize. What this approach actually does This process: – Protects high-intent driver learning – Prevents wasted spend during panic testing – Keeps approvals stable – Restores performance without nuking the account Most auto insurance accounts don’t need “new ideas.” They need discipline. Wrap-up Every auto insurance account experiences drops. The difference between accounts that recover and ones that spiral is how the operator reacts. Calm, structured troubleshooting beats emotional action every time. If you want deeper breakdowns on auto insurance structures comment "AUTO" and ill share our frameworks (must like + follow)

unless you get insanely lucky (bitcoin at $1 type lucky), you’re not investing your way out of poverty with retail trades. especially in auto insurance / leadgen. here’s the math nobody wants to hear: 10% yearly return on $1,000 = $100 that doesn’t even cover one bad month, let alone build a life. now compare that to cashflow plays. one average auto insurance buyer: – $8–$15 CPL – $40–$120 payout per qualified lead – 20–40% margin if you know what you’re doing run 50 leads/day at $10 margin = $500/day = $15,000/month cashflow now the psychology flips. you’re no longer “hoping” investments save you you’re deploying surplus cash from a machine that prints daily this is why almost every serious affiliate does the same path: 1. build cashflow first (leadgen, insurance, finance) 2. stabilize income at $10k–$30k/month 3. then invest from excess, not desperation trading small money feels productive but it’s just slow gambling cashflow businesses compress time. auto insurance isn’t sexy but it turns effort into capital fast and capital is what actually gives you optionality later.

Everyone said “test more creative.” I did the opposite and took an auto insurance account from 0.75 ROAS to 2.5 ROAS in 5 days. The problem wasn’t the ads. It was how Meta allocates budget in auto insurance accounts. Background (Oct – early Nov) Vertical: Auto insurance leadgen Structure: CBO October performance: – Spend: $13k – ROAS: 1.14 – CPA: $53 Structure: – 8–10 new concepts per week – Each concept = 1 ad set, 3 variations Profitable, but chaotic. Early November (Nov 1–17) Reacted to Andromeda by consolidating: – Fewer ad sets – Larger batches – Kept one proven ad (“Ad A”) that had been converting for 3 months Performance improved: – Spend: $9k – ROAS: 1.35 – CPA: $48 Everything looked fine. Then November 17 happened. The collapse (Nov 17 – Dec 16) November 17: – Best day of the month – 24–26 approved leads at ~$35 CPA That same day, I launched 6 new auto insurance ads (“Batch 4”). One ad (“Ad B”) immediately absorbed most of the budget. Ad B metrics: – High hook rate – Strong video engagement – Tons of clicks Business metrics: – $68 CPA – 0.85 ROAS – $2.5k spend Meanwhile, Ad A (the proven one): Before Nov 17: – 1.38 ROAS – $43 CPA – $16k spend After Nov 17: – 0.66 ROAS – $80 CPA – $4k spend Same ad. Same script. Same landing page. Suddenly “dead.” Panic mode (Nov 18 – Dec 16) I did everything people recommend: – Paused Ad B – Relaunched Ad B – Added ABO alongside CBO – Increased budgets to “push through learning” – Decreased budgets to “stop bleeding” – Changed targeting Results: – Nov 18–30: 0.92 ROAS, $72 CPA – Dec 1–16: 0.75 ROAS, $87 CPA Total damage: – ~$18k spent – ~$8k lost The reset (Dec 16–17) I wiped the slate clean. – Paused everything for 24 hours – Launched ONE campaign – ONE ad set – 9 ads total Included: – “Ad A2” — a minor iteration of the original proven auto insurance ad Budget: – $300/day (down from $900 chaos) No new concepts. No testing spree. The recovery (Dec 17–22) – Dec 17: $39 CPA, 1.87 ROAS – Dec 18: $65 CPA, 1.03 ROAS – Dec 20: $28 CPA, 1.72 ROAS – Dec 21: $25 CPA, 2.13 ROAS – Dec 22: ~$23 CPA, 2.5 ROAS 5 days. 55+ approved leads. Zero new creative. What actually happened Meta optimized for engagement, not insurance approvals. The new ad (“Ad B”) had: – Higher hook rate – Higher hold rate So Meta pushed it into: – The same high-intent driver segments that Ad A had already “trained” for Result: – Ad B consumed the best audiences but converted poorly – Ad A got leftover traffic and “died” The account collapsed because: – High-engagement ads cannibalized high-intent audiences – Conversion efficiency was ignored in the early allocation phase Why this is especially dangerous in auto insurance – Approval ≠ click – Engagement ≠ intent – Meta reallocates budget in the first 6–12 hours – Insurance approvals show up 24–72 hours later By then, damage is done. The lesson More creatives ≠ more performance in auto insurance. If you introduce high-engagement ads too early: – They steal your best driver cohorts – They poison the learning phase – They kill proven ads Sometimes the highest-ROAS move is removing optionality from the algorithm. One strong ad > ten competing ones. If you want: – The exact auto insurance hooks that convert without cannibalizing – The offers that survive scaling – The advertisers actually running clean structures profitably Reply “AUTO” (must like + follow)

@ auto insurance affiliates you can get the SAME insurance offer to show up 2–3 times in Google Search and capture 2–3x the inbound lead volume from the exact same keyword set here’s how to do it (without breaking Google policy) take your top 5–10 converting auto insurance offers (same buyer, same payout, same backend) create 3–4 separate search builds each build = a different intent angle example: – one build targets price pain “cheapest car insurance after rate increase” – one targets driver type “auto insurance for high-risk drivers” – one targets life event “car insurance after moving states” you’re basically making Google treat each as a unique intent match change the following for each build: – headlines (problem vs persona vs situation) – descriptions (rate drop vs comparison vs urgency) – sitelinks (quote, compare, check eligibility, switch now) use separate campaign IDs separate budgets separate bid strategies track each build independently now the important part: your landing page MUST match the angle price angle → rate comparison framing persona angle → driver-specific language event angle → “this is why your rate changed” framing Google doesn’t care that it’s the same insurance buyer they care that the user intent is different we’ve run this exact structure for years in auto zero policy issues massive volume lift most affiliates try to jam every keyword into one campaign then wonder why CPCs spike and volume caps out intent stacking is how you scale insurance search without increasing risk same offer same buyer multiple demand surfaces

Here’s how and why creative volume actually matters for scaling auto insurance. This is the formula we use, and how 1 concept turns into 100+ insurance ads: Concept × Trigger × Driver Persona × Format Start with one core concept. Concept example: If your premium keeps going up, it’s not random. It’s how insurers price risk, and this is how you lower it. Now multiply it by real triggers (problem angles): – If your rate jumped after a minor accident – If your rate went up after a speeding ticket – If your policy renewed higher “for no reason” – If you just moved ZIP codes – If your insurer quietly repriced you this year From 1 concept, you already have 5 angles. Now layer personas. – If you’re a young driver with a clean record and your rate jumped – If you’re a parent insuring multiple cars and your rate jumped – If you’re a rideshare driver and your rate jumped – If you’re a homeowner bundling policies and your rate jumped – If you’re over 40 with no claims and your rate jumped Those same 5 angles × 5 personas = 25 ads. Now multiply by formats. At minimum in auto insurance you can run: – Hi-fi static (rate comparison screenshots) – UGC selfie video (driver ranting about a price increase) – Native screenshot (renewal email, policy notice) – Split screen problem/solution native – Short GIF or motion statics 25 ads × 5 formats = 125 ads from a single concept. That’s what creative volume actually means. It’s not taking: “If your rate went up after a ticket” and making 20 slightly different headlines. It’s taking one core insurance truth and mapping it across: – multiple real-world triggers – multiple driver profiles – multiple native formats That’s why big auto insurance advertisers can run hundreds or thousands of ads without “fatigue.” They’re not guessing. They’re covering the entire risk surface. If you have 5–10 core insurance concepts (rate increases, cancellations, SR-22, non-standard drivers, lapses, bundling, state changes), you can realistically support 1,000+ ads without repeating yourself. Meta isn’t pushing volume for fun. It’s pushing relevance. And in auto insurance, relevance is everything.

We A/B tested 1,200 ad hooks specifically for auto insurance funnels under $500K/month. The top 5% all had this one thing. Last year, I mass-produced what I thought were “winning hooks.” Aggressive. Benefit-driven. “Save $X.” Fast cuts. Stock footage. Price-first angles. You know the type. You’ve seen them everywhere in auto insurance. And for about 2–3 weeks, they worked. Then one morning: CPMs doubled. CTR got cut in half. CPL went through the roof. So I did what most people don’t actually do when performance collapses. I stopped guessing and started logging everything. Over 6 months, I tested 1,200 hooks across 47 auto insurance funnels. All under $50K/month. All competing against advertisers spending 10x more. Here’s what actually worked — and it’s not “pattern interrupts” or “lead with savings.” The real problem with hooks in auto insurance Most small auto insurance advertisers copy hooks from big spenders. That’s the mistake. Big brands aren’t playing the same game. When a national carrier opens with: “Drivers are switching and saving $742/year” They’re leveraging brand trust built over decades and tens of millions in spend. When you run that same hook at $20K–$30K/month, you’re paying Meta to show a generic claim from a brand nobody recognizes. There’s no trust foundation. So your hook has to do something different. What we found in the top 5% After 1,200 tests, the pattern was clear. The top-performing auto insurance hooks didn’t try to grab attention. They delivered instant expertise. In the first 2 seconds, the viewer learned something real about how insurance pricing actually works. Not teased. Not hinted. Delivered immediately. Example: Generic hook: “Most drivers overpay for car insurance.” Instant expertise hook: “Your ZIP code affects your car insurance more than your driving record — here’s why.” The second one teaches something before the viewer even knows it’s an ad. That’s what builds trust when you don’t have a brand. Why this works for sub-$50K auto insurance funnels At this spend level, Meta doesn’t have enough conversion data to optimize cleanly. You’re not feeding it 50+ conversions per week. So Meta guesses. And when hooks are vague, Meta optimizes for cheap clicks — not qualified drivers. Instant expertise hooks solve this. They self-select the audience. Only drivers who recognize the problem keep watching. Only drivers who feel personally affected click. You’re qualifying traffic before the landing page. One funnel example: Old hook: “Lower your car insurance in minutes.” New hook: “Drivers with clean records still overpay because insurers re-rate every 6 months.” Same offer. Same funnel. Same payout. CTR went from 1.1% → 2.6% CPL dropped 38% Approval rate went up, not down. The 4 instant-expertise hook types that worked in auto insurance Type 1: The correction “You’re not being penalized for accidents — you’re being penalized for coverage gaps.” The key is explaining why in the hook itself. Type 2: The hidden mechanism “Insurance companies raise rates after 6 months even if nothing changed. It’s called passive re-rating.” Viewers feel like they finally understand the system. Type 3: The specific number “Drivers with lapses over 30 days pay 23–31% more on average.” Specific numbers signal real data, not marketing. Type 4: The reframe “High insurance rates aren’t about bad driving. They’re about how insurers classify risk.” This creates an instant mental shift. Hooks that looked good but died fast “Wait for it” hooks Cheap clicks, low intent. Shock / outrage hooks High comments, bad buyers. Fake scarcity Meta learns low-quality signals fast. High-production brand ads Signal ‘corporate’, not ‘helpful’. Auto insurance buyers want clarity, not polish. How we write these hooks now Step 1 List insider knowledge about insurance pricing, underwriting, and re-rating. Step 2 Pick insights drivers would actually say “wait, what?” to. Step 3 Deliver the full insight in the first sentence. Step 4 Film it like you’re explaining it to a friend. Step 5 Test expertise angles against each other — not against generic hooks. The database Every hook is logged. By persona, state, driver type, credit band, and funnel type. We track: – CTR – CPL – Approval rate – Longevity Patterns are very clear once you stop guessing. If you want access to the exact hooks that are working right now in auto insurance: Reply “AUTO” (must like + follow) This is the same framework we use before scaling spend.

That’s not the right way to think about it. ROAS doesn’t tell you if your auto insurance campaigns are profitable. Your unit economics do. The real question is: “What’s my target CPL based on my buyer payouts, scrub rate, and cash flow?” Here’s what you should actually track daily in auto insurance: – Average payout per sold lead – Approval rate (what % actually gets paid) – Scrub / clawback rate – Net revenue per lead – What’s left to spend on traffic In insurance, this is your equivalent of CM2. Net revenue per lead = (payout × approval rate) − expected clawbacks That number tells you exactly how much room you have. Your target CPL should always sit below that. Especially early on. Example: – Advertiser payout: $45 per approved lead – Approval rate: 70% – Effective revenue per raw lead: $31.50 – Average clawback risk: $1.50 Net revenue per lead ≈ $30 If you want a $10 margin per lead: Target CPL = $20 Everything else is noise. Once you know your target CPL, you can work backwards into ROAS, CPA, or whatever platform metric you want to use. But those are outputs — not decision metrics. Early stage rule: Set your CPL so you’re profitable on first pass. No LTV fantasies. No “it’ll optimize later.” Once you have stable buyers, predictable approval, and cash flow buffer, you can push CPL higher to scale. ROAS is just a translation layer. Unit economics decide whether you live or die.

never sold an auto insurance asset yet. but that’s the only reason we’re building them. cashflow is great, sure — but leadgen businesses trap cash everywhere. why you should build auto insurance ops to exit, not just to run: CASH running insurance at scale locks money in dumb places. networks: 60–90 day payment terms advertisers: clawbacks + quality holds traffic sources: prepaid balances call centers: payroll float compliance buffers: cash you can’t touch you can be net-profitable and still feel cash-poor. that money isn’t yours. it’s an operations enabler. MATH insurance exits aren’t sexy multiples, but they’re real. 2–3x annual profit is normal for: – clean data – diversified buyers – stable CPL – documented ops $4M yearly profit × 2.5 = $10M paid at once. no waiting. no payment terms. no chargebacks. BUILD buyers don’t want “good buyers” or “strong traffic.” they want: – buyer diversification – predictable lead flow – clean compliance history – ops that run without founders – documentation for everything your job isn’t to squeeze today’s margin. it’s to build something boring and transferable. TIMING you’re ready to sell when: – scaling effort > marginal upside – growth means more risk, not more reward – the system works without heroics that’s the signal. auto insurance isn’t a forever game. it’s a build → stabilize → exit game. build the asset. get paid once. move on clean.

@cashwaychris Dm

Buying FB ad accounts Drop your connects - Testing new providers

Stop trying to make your auto insurance ads look “professional.” Polished creatives die fast. This ugly split-screen format is a native ads cheat code. 1/ The pattern interrupt Native feeds are full of stock photos, smiling families, and generic “Save on insurance” headlines. A raw split-screen stops the scroll immediately: Left: dashboard warning lights, cracked windshield, accident photo Right: renewal notice, premium increase, cancellation email It looks real. It looks stressful. It doesn’t look like an ad. 2/ Double validation (problem + consequence) You’re not just showing a car problem. You’re showing: – the trigger (ticket, accident, warning light) – the punishment (rate hike, cancellation, insane premium) That combo bridges emotion and money. This isn’t “get a quote.” This is “this could cost you $2,400 this year.” 3/ Curiosity gap The brain immediately asks: – “Is this why my premium jumped?” – “Am I about to get dropped?” – “Is there a way around this?” You’re not explaining anything yet. You’re letting their fear fill in the gaps. 4/ CTR velocity On Taboola and Outbrain, CTR is everything. Higher CTR means lower CPC and more volume. These ugly but specific split screens consistently: – beat clean brand creatives – unlock cheaper clicks – let you outbid competitors with half the budget Pretty ads reassure brands. Ugly ads make people click. Test the split screen.

auto insurance lesson that surprises most people: we removed a small “service fee” from one geo thinking it would boost CVR it didn’t results after removing it: – form completion down – lead quality down – close rate down – worst performing market in 14 days we added the fee back within 7 days: +12% form completion +18% policy close rate +22% revenue per lead why this happens: “free” attracts curiosity traffic not buyers a small fee (or friction checkpoint): – filters unserious users – increases intent – makes the quote feel valuable same traffic same creatives better economics cheap leads are often the most expensive ones.