AMI Ads @ami_ads

after managing $27,483,192 in auto insurance facebook spend, I’ve seen more funnels collapse overnight than I can count this is the exact process I use to recover performance without blowing up the account most affiliates panic the moment CPL spikes or approvals drop one bad day and they: pause campaigns duplicate ad sets change budgets launch 10 new creatives after years in this vertical, here’s the truth drops are normal bad reactions kill accounts the problem is rarely the drop it’s how you respond to it why auto insurance campaigns drop most performance swings have nothing to do with “broken ads” common causes: – day-of-week behavior – pay cycles – seasonal intent shifts – creative fatigue in tight personas – delivery randomness – backend issues (approvals, call centers, routing) none of this means your funnel is dead it means behavior changed temporarily mindset before touching anything one bad day means nothing I’ve seen: $14 CPL → $29 CPL → $16 CPL in 3–4 days with zero changes people rebuild entire accounts after one bad afternoon that’s how you destroy months of learning before touching anything don’t fix understand the process A. zoom out to 3–5 days – is it a trend or a spike – account-wide or isolated one bad day = noise consistent drop = signal I’ve seen accounts go from 0 approvals to 5x ROAS two days later no changes B. check outside meta – weekend vs weekday – start of month – renewal cycles – call center issues – approval delays ads can be fine while backend is broken C. check fatigue if CPL rises slowly and frequency climbs it’s fatigue common in: – high-risk drivers – seniors – low-credit fix is not a rebuild rotate refresh slight variation D. check delivery – audience too tight – meta shifting to lower intent you need fresh drivers not the same exhausted cohort E. soft resets – duplicate campaign – keep ads – restart delivery don’t nuke the structure what not to do – pause everything – launch 10 new ads – swing budgets aggressively – rebuild without knowing why sometimes the best move is nothing let it stabilize what this does – protects high-intent traffic – avoids wasted spend – keeps approvals stable – restores performance most accounts don’t need new ideas they need discipline if you want the full Meta Ads Playbook we use to stabilize and scale comment “META” (must like + follow)

there’s a way you can use google trends to make better creative decisions in insurance leadgen you already know ads work best when they tap into existing beliefs google trends shows you how familiar a market already is with a concept because the more something is searched the less your ad has to explain and the more people it naturally resonates with EXAMPLE TIME let’s say you’re comparing angles for insurance “cheap car insurance” “lower monthly premium” “switch providers” “save on car insurance” you layer them in google trends now you’re comparing angles inside the same category right side you see which angle has the highest overall search volume that’s your strongest baseline message left side you see how demand changes over time which angles are stable which ones are growing TLDR google trends shows you what people already understand so you can build creatives that don’t need heavy explanation and instead tap into existing intent use it as a filter for your angles and your creatives will hit harder

anyone doing $83,417/month+ in insurance leadgen knows best practices aren’t optional yes, data and testing matter when you’re scaling but for most accounts it’s about putting the right elements in the right place at the right time take this example just by adjusting one step in the funnel we generated +$83,417/month extra no new traffic no new creatives no complex testing just fixing fundamentals here’s what we changed: 1️⃣ improved form flow we adjusted the order of questions ZIP first simple inputs early contact details later users stayed longer intent increased 2️⃣ surfaced trust at the right moment added simple reassurance before submit “drivers in your area are switching” clear expectations friction dropped the result? +$83,417/month same traffic sometimes it’s not about complex tests it’s about understanding how users behave and making the obvious path easier to take

everyone said “test more creative” I did the opposite and took an auto insurance account from 0.74 ROAS to 2.51 ROAS in 5 days on ~$18,472 spend the problem wasn’t the ads it was how meta allocates budget background vertical: auto insurance structure: CBO january: – spend: $13,284 – ROAS: 1.12 – CPA: $54 structure: – 8–10 concepts per week – each concept = 1 ad set, 3 variations profitable but messy february: – spend: $9,183 – ROAS: 1.34 – CPA: $49 everything stable then it broke march 17: – best day – ~25 approved leads at ~$36 CPA same day I launched 6 new ads one ad took over metrics looked good: – high hook rate – strong engagement but: – $69 CPA – 0.84 ROAS – $2,487 spend meanwhile the proven ad: before: – 1.39 ROAS – $43 CPA – $16,000+ spend after: – 0.67 ROAS – $81 CPA – $4,000 spend same ad looked “dead” panic phase paused relaunched added ABO changed budgets results: – ROAS dropped to 0.75 – CPA hit $87 ~$18,000 spend ~$8,000 loss reset paused everything launched: – 1 campaign – 1 ad set – 9 ads included a slight variation of the original winner budget: $300/day no new concepts recovery (late march → early april) – day 1: 1.86 ROAS – day 3: 1.72 ROAS – day 5: 2.51 ROAS 55+ approved leads zero new creatives what happened meta optimized for engagement not approvals new ads had better engagement so they got pushed into high-intent audiences but converted worse they drained the best traffic and killed the winners why this matters in auto insurance – approval ≠ click – engagement ≠ intent – allocation happens fast – approvals come later by the time you see it damage is done the lesson more creatives ≠ better performance sometimes fewer options = higher ROAS one strong ad > ten competing ones if you want the exact hook files that avoid this reply “HOOKS” (must like + follow)

this system for insurance leadgen is for you if you’ve been running ads for 6+ months and haven’t hit your first $1,000 day the testing phase ends here you know exactly who you are you follow 100+ accounts talking about meta ads you’ve watched hours of content on funnels, creatives, scaling you’ve saved every “$10k/day case study” you see and you’re still stuck at $0–$200/day not because you don’t know HOW but because you won’t EXECUTE properly you’re addicted to tweaking changing campaigns every day switching angles too fast killing things before data comes in there is no missing piece you already have what you need you just don’t stick to a system you keep testing instead of scaling here’s what actually happens you launch you don’t see results in 24–48 hours you change everything repeat that’s why you stay stuck meanwhile people doing $12,487/day know LESS than you but execute better they pick one angle one funnel one structure and run it properly they let data come in they optimize based on backend they scale what works while you’re still resetting campaigns this system is built for you the over-tester the over-optimizer the one stuck in cycles because it forces execution day 1: pick one angle day 2: build the funnel day 3: launch campaigns day 4–10: test creatives without touching structure day 11–30: scale based on RPL you either follow it or you stay stuck there’s no middle ground and here’s what happens by day 7 you have real data by day 14 you hit breakeven by day 30 you’re pushing toward $1,000/day or you prove you don’t execute properly most people will keep tweaking keep overthinking keep sitting at $0–$500/day while others pass $20,384/month same knowledge different execution this isn’t for people who want to “learn more” this is for people ready to run traffic properly if that’s you this is your moment you either execute or you stay stuck your choice

when starting out, I was copying everyone copying their funnels their creatives their angles and it worked got me to $20,384/month in revenue but there’s a ceiling to copying other people’s systems at some point I wanted to try a different angle something nobody was really pushing in insurance leadgen and the consensus was “that won’t work” there was no playbook to follow so for the first time I had to figure it out myself I committed anyway at first nothing happened campaigns didn’t move no traction then it started working one angle hit then another when it clicked, everything changed I realized I could figure things out myself that sounds obvious but early on you just follow what works because you don’t know better once you get experience you start seeing patterns you start making your own bets that angle became a serious advantage scaling past $52,817/day while others were stuck copying the same stuff the lesson wasn’t about the angle it was about breaking the belief that you have to follow what everyone else is doing once you get past that you start finding your own edges

we A/B tested 1,284 ad hooks across $487,293/month in auto insurance spend. the top 5% all had this one thing last year, I mass-produced what I thought were “winning hooks” aggressive benefit-driven “save $X” angles you know the type and for about 2–3 weeks, they worked then performance collapsed CPMs doubled CTR dropped CPL spiked so instead of guessing I logged everything over 6 months 1,284 hooks tested 47 funnels all under $50k/month competing against accounts spending 10x more here’s what actually worked the problem with most hooks small advertisers copy big brands that’s the mistake big brands can open with: “drivers are saving $742/year” they have trust you don’t so your hook needs to do something else what worked in the top 5% they didn’t try to grab attention they delivered instant expertise in the first 2 seconds the viewer learns something real not teased delivered immediately example generic hook: “most drivers overpay for insurance” expert hook: “your ZIP code affects your rate more than your driving record — here’s why” that’s what builds trust why this works at sub-$50k/month meta doesn’t have enough data so it guesses and vague hooks bring low-quality traffic expert hooks filter only relevant users engage you qualify before the click example old hook: “lower your insurance in minutes” new hook: “drivers with clean records still overpay because insurers re-rate every 6 months” same funnel same payout CTR: 1.1% → 2.6% CPL: -38% approval rate increased the 4 hook types that worked type 1: correction “you’re not penalized for accidents — you’re penalized for coverage gaps” type 2: hidden mechanism “insurers raise rates every 6 months even if nothing changed” type 3: specific numbers “drivers with lapses over 30 days pay 23–31% more” type 4: reframe “high rates aren’t about bad driving — it’s about risk classification” what didn’t work “wait for it” hooks shock angles fake scarcity overproduced ads auto insurance buyers want clarity not polish how we write hooks now list insider knowledge pick insights that trigger curiosity deliver the full idea instantly test expertise vs expertise we log everything CTR CPL approval rate longevity patterns become obvious if you want the exact hook files behind this reply “HOOKS” (must like + follow)

the secret to improving your media buying is spending money i know so many who won’t even test $50/day start panicking the moment real budget is on the line even though they talk about scaling and running ads all day yet won’t risk a small % of their bankroll on themselves (avg affiliate will sit around $5k–$10k/month with $20k–$50k saved, sample size of people i’ve seen) so here’s the cure launch campaigns spend money accept you’ll burn $500–$1,000 learning that will teach you more than any course because once money is on the line you actually start paying attention that’s where the real learning happens

building a new insurance funnel LIVE after hitting $3,217 in a single day, exposing everything step by step no one does this publicly because they can’t built this whole funnel off almost no sleep after burning through my first 4 angles nothing worked CPL too high RPL too low had to go deeper went back to the data looked at every drop-off every weak signal tested new hooks launched again same thing miss after miss kept pushing needed a win finally hit one angle everything clicked CPL dropped RPL jumped profitable instantly scaled it same day first $3,217 day now pushing higher same funnel same system just execution

what $0 → $512,847/month in spend ACTUALLY looks like (from a $48,392,173 auto insurance affiliate) most people think scaling auto insurance is slow, capped, and fragile that’s only true if you run it like ecom direct-response auto insurance behaves differently direct-response vs “normal” funnels ecom funnels: – product-first – brand-safe – soft persuasion – gradual trust direct-response auto insurance: – conversion-first – aggressive qualification – zero brand sentiment – cold → action immediately same platforms completely different economics the only 3 levers that matter (in order) 1. offer – state-specific – persona-specific – payout ceiling upfront this drives 60–70% of CPA 2. creative – script > visuals – one intent per ad – one problem per persona 3. landing page – pre-qualifies before form – filters low-value drivers – forces intent nail these and scale becomes mechanical the case study numbers – vertical: auto insurance – payout: $183–$257 per converted policy – target CPA: <$92 – break-even CPA: ~$121 – initial CPA: $67 – initial return: ~3.1x why this scaled 1. massive pain – high-risk drivers – lapsed policies – DUIs, tickets, low credit 2. infrastructure already existed – proven scripts reused – same psychology, different states – adapted, not reinvented 3. heavy pre-sell – long-form advertorial – VSL-style education – price anchored after trust people didn’t shop they committed scaling path – day 1–3: $1,200–$3,400/day – day 4–7: $6,300–$10,800/day – week 2: $15,200–$21,700/day – end of month: ~$512,847 spend no hacks just margins + intent control the real takeaway auto insurance scales when: – you can afford higher CPA – you control what hits the pixel – you stop treating leads equally most affiliates fail because: – payout ceiling too low – funnel leaks junk traffic – pixel learns garbage this isn’t beginner level but it proves one thing when fundamentals are right auto insurance doesn’t scale slowly it scales aggressively if you want the exact hook files behind these campaigns reply “HOOKS” and i’ll send them (must like + follow)

“you’re doing $2,847,193/month in revenue... but are you actually profitable?” I get this question all the time and honestly, it’s the dumbest question in this space do you think we’re running leadgen just to push volume for fun? there’s no VC money no vanity metrics no “growth at all costs” this is built to make money the entire goal is backend profit we aim for $6–$10+ margin per lead that’s the sweet spot that lets us scale while still printing real cash so yes, it’s profitable and I share these numbers for people who think volume alone matters because it doesn’t profit does

a media buyer doing $1,184,327/month told me his audience didn’t respond to “overpaying” angles anymore, which is why his account plateaued but we changed one thing instead of leading with the problem, we opened the first 10 seconds with pure savings and outcomes actual numbers real scenarios what the user cares about that shift came directly from the research we did for his creatives we tested multiple angles on his account and once we found the ones that truly resonated, we paired them with the right creative style that’s when it really took off you can see it in the naming version 2 version 3 version 4 all scaled each one spent over $127,483 that’s how you know it wasn’t a lucky ad the entire concept worked final numbers: $982,417 spent $11.84 CPL 83,000+ leads generated every angle you test without research is just an expensive guess

go deeper into one angle at a time remember, if the angle is validated and scaling for others, it’s almost always a skill issue BUT obviously, some angles perform better than others even if they’re proven so I run a stop loss of $500–$700 per angle if it shows zero momentum, I move on if I’m close to breakeven or slightly negative I stay on it longer because at that point it’s not the angle it’s execution

back by popular demand, we’re reopening the auto insurance funnel vault after analyzing $7,482,193 in live spend this is our most requested resource: the top-performing auto insurance funnel breakdowns running right now not just ads the entire stack: paid traffic → advertorials → landing pages → forms/calls → follow-ups → monetization everything this is the most copied asset we’ve shared in insurance so we packaged it properly one folder zero fluff here’s what’s inside: – leadgen advertorial funnels doing $3–$7 CPL – call-optimized landing pages (LPV → call) – state-specific routing logic – scrubbing + postback setups used at scale – follow-up flows recovering 20–35% lost leads – monetization setups for auto + bundled insurance all pulled from campaigns actively spending use this to: cut weeks of testing steal proven structures plug into what’s already working if you want the folder, you know the drill retweet drop “FUNNEL” below we’ll send it over

my student just hit $72,843 in a single day running insurance leadgen if you would’ve asked me a few years ago if this still scales like this I would’ve said you’re crazy his story is insane as well started with almost nothing reinvested everything back into traffic had multiple setbacks along the way joined our circle about a year ago went through ups and downs doubled down on learning and execution the results speak for themselves he’s already cleared over $104,672 in profit this month same model just executed properly

my new testing campaign dropped CPL by ~40-50%. here’s how it works I switched from an ABO to a CBO for testing. every ad set gets a min spend of $50–100. total campaign budget is $1,500–$2,000 that’s it the old ABO had 10–12 ad sets running at once. only ~30–40% were actually working. the rest were dragging blended CPL up to $14–$16 because every ad set had its own budget regardless of performance with a CBO, meta focuses spend on what’s actually working → winners scale to $300–$600/day → losers stay around $50–$80 → budget isn’t forced into underperformers → blended CPL drops to $8–$10 RPL stayed $17–$20+ margin increased hard in a way it becomes a scaling campaign. the min spend makes sure every ad set gets data but meta decides where the real budget goes goes to show there isn’t one way to test. ABO works, but this setup is performing better right now

when i learned this account structure scaling from $10k to $100k/month became too easy my account structure and creative volume based on monthly spend in insurance leadgen under $30k/mo: → 1 CBO, 1 ad set for winners, 1 ad set for testing → test 20–30 new creatives per month minimum → typical CPL: $8–$12 → target RPL: $14–$18 $30–100k/mo: → scaling CBO with cost control → ABO for testing, one angle per ad set → add a zombie campaign for fatigued ads → test 40–60 new creatives per month → typical CPL: $9–$14 → target RPL: $16–$22 $100k+/mo: → full structure: scaling CBO, testing ABO, whitelisting campaigns, zombie campaigns, copy testing campaigns → aggressive push on whitelisting + new angles → test 80–120+ creatives per month → typical CPL: $10–$18 → target RPL: $18–$30+ to be completely honest its not as much about creative volume as it is about angle volume hitting different concepts different hooks different intent levels that’s what actually unlocks scale that said as spend increases creative output needs to increase with it you can’t spend $100k+/mo testing 20 creatives you’ll stall some accounts scale with fewer creatives but only if the angles are strong enough to carry the spend

watched an affiliate turn $1,237 into $6,482,193/year in auto insurance sounds fake it’s not here’s how: he didn’t “scale ads” with $1,237 he used it to access buyers with capital started with $300/day on facebook kept ~$900 buffer for variance + scrubs sent traffic to live auto insurance buyers numbers at the start: $1.31 avg CPC $3.08 EPC $300/day spend ≈ 229 clicks ≈ $705 revenue ≈ $405 margin/day within 30 days: $11,482 ad spend $26,317 revenue $14,835 margin instead of withdrawing he raised caps buyer cap went from $300/day → $2,000/day added 3 more buyers running the same offer now: 4 buyers × $2,000/day = $8,000/day spend same math: $8,000/day spend $3.02 EPC ≈ $18,124/day revenue ≈ $10,124 margin monthly: $240,000 spend ≈ $543,720 revenue ≈ $303,720 margin scaled across geos + segments ended up running: 8 buyers $15k–$20k/day spend annualized: $5,482,193 ad spend ≈ $12,038,417 revenue ≈ $6,556,224 margin he didn’t invent anything didn’t build a brand didn’t write copy the shift: stop asking “how do I make money with small budgets?” start asking “how do I access budgets that make mine irrelevant?” auto insurance is capital arbitrage you’re routing intent to companies with balance sheets they fund the scale you take the spread that’s how this actually scales comment “OFFERS” and i’ll send you what he’s running (must like + follow)

here’s a real world reminder to always challenge your assumptions in insurance leadgen i was recently looking at an account doing ~$70k/day in revenue after digging into the data, there were three conclusions that didn’t make sense at first 1. higher CPL traffic was more profitable the campaigns running at $12-$14 CPL were outperforming the $7-$9 CPL campaigns better RPL better approval rate more backend revenue this goes against what most affiliates believe but it was clear in the data 2. cold paid traffic outperformed everything else the broad facebook traffic was converting better on the backend than “warmer” segments higher contact rate higher close rate again, doesn’t make sense if you’re thinking in basic terms but backend told the truth 3. “native looking” content alone wasn’t enough we tested a lot of soft, organic-style creatives they looked good they got engagement but when it came to actual lead quality the ads built specifically for conversion outperformed everything the ones written with intent and structure not just content repurposed into ads anyways the reason I’m sharing this is because all three go against what most people think they know about leadgen the lesson is simple there are no fixed rules everything is contextual if you can’t challenge your assumptions and follow the data you won’t scale drop your ego trust the numbers and you’ll print

there’s a crazy playbook happening right now in insurance leadgen anytime an angle starts scaling to serious numbers, it gets copied almost instantly across the market you launch something, it starts doing $30k/day push it harder, hit $50k/day and within a week you’ll see the same hook, same angle, same messaging everywhere other affiliates pick it up networks spread it buyers start recognizing the pattern and your edge disappears you paid for the testing you burned the budget you found the winner and everyone else gets to run it for free they might do worse numbers they might have worse lead quality but they’re still competing for the same traffic and that hits your performance CPMs spike fatigue hits faster buyers start capping this used to take months now it happens in days I’ve seen angles go from $0 to $200k/day across the market and then die just as fast because everyone piles in which means the real game isn’t just finding winners it’s extracting as much revenue as possible before saturation hits and moving before everyone else catches up most people complain about copycats the ones doing $1M+/month expect it and build around it because in this market everything that works gets copied the only real edge is speed