ROAS Mutilator

17yo high school dropout | silent. faceless. efficient.
digital & physical ecom products to +$1.3M

[ROAS Mutilator @ROASMutilator]

@nguyentomyy webby

[ROAS Mutilator @ROASMutilator]

steal winning product ideas before they trend (takes 90 seconds):
search "[your niche] reddit"

-sort by: most commented this month
-find the pain points mentioned 50+ times
-create the solution congrats, you just did market research that actually matters while everyone else is guessing on aliexpress

[ROAS Mutilator @ROASMutilator]

half agree, "optimize everything" is just expensive guessing without diagnosis first.

find your single biggest funnel drop off (ad to landing? atc to purchase?) and fix that before testing 3 avatars or 47 headlines.

most brands stuck at $300k have one structural leak (wrong audience temp, price shock, messaging mismatch) that no cro testing fixes because they're optimizing the wrong variable

real expertise is knowing what not to test after diagnosing where revenue actually dies

[ROAS Mutilator @ROASMutilator]

@margx_ecom solid but €50/day profitable ≠ scalable

cpc drops might just be easy pockets that exhaust at €150/day.

wait 5 days stable before adding budget or you'll hit real auction costs and flip unprofitable fast

[ROAS Mutilator @ROASMutilator]

@Bogzabs96 this format kills for supplements/complex products where "why this works" matters more than "does this look good on me."

claiming 40% hook rates means nothing if conversion rate tanks because you educated people who didn't need convincing, they needed dopamine and urgency

[ROAS Mutilator @ROASMutilator]

@mannybarbas_ interesting setup, but how are you handling creative fatigue with only 2 ad sets per week launching?

[ROAS Mutilator @ROASMutilator]

lmao this is a "how to burn money in 30 days" checklist

ripping tiktok creatives without testing your own gets you banned and autods fulfillment times kill repeat rate before you even have one

stoploss at -$200 per product assumes you're testing 10+ products to find one winner, which means you need $2k+ just to validate anything.

most people following this run out of cash before finding product-market fit because they're optimizing for speed not survival

[ROAS Mutilator @ROASMutilator]

yep, but asking 6 questions kills completion to 15%, they just bought, they're not doing homework.

ask ONE on thank you page: "what almost stopped you from buying?"

that's your real blocker list.

save age/frequency questions for email 3 days later when product arrives and they actually care again

[Ecom Matteo @Ecom_Matteo]

WINNING PRODUCT OPTIMISING IDEA:

If you have a winning product, make sure to integrate a post purchase survey after your customer has bought.

IF YOU ARE USING SHOPIFY:
- Use the app "Zigpoll"

IF YOU ARE USING FUNNELISH:
- Integrate a link to "Typeform" in your Thank You Page screen.

What to put in your survey?

You want to get more information from your customer.
- Why did they buy?
- Which age group?
- Why were they hesitant to buy?
- How often does the problem you are solving occur?
- What have they tried?

Make sure to ask questions whereby you can improve your product, adjust your sales page, get new marketing angles, etc etc etc

[ROAS Mutilator @ROASMutilator]

@andrejtumo yep, but value stacking only works if the "€300 value" feels real, inflating perceived value with garbage bonuses (ebooks, guides) kills trust faster than no offer at all

[ROAS Mutilator @ROASMutilator]

@benradack cool theory, but high to low cost cap only works if you have volume, brands under $5k/day just starve ads when they drop the cap because meta can't find enough cheap conversions to hit the floor, so everything stops instead of self-optimizing like you think

[ROAS Mutilator @ROASMutilator]

@williamkast_ great post william

[ROAS Mutilator @ROASMutilator]

@GlennNieuwenh GOAT

[ROAS Mutilator @ROASMutilator]

You shouldn’t study ecom brands just because they’re spending big.

A lot of “killer ad accounts” are propped up by aggressive subscription mechanics, not marketing skill.

When you can force rebills, upsells, and price jumps…

You can afford bad creatives, high CPMs, and sloppy funnels.

Real marketing skill shows up when the frontend has to work or the brand dies.

[ROAS Mutilator @ROASMutilator]

@yegormethod @eptwts at 14***

[ROAS Mutilator @ROASMutilator]

@Camicees yep, but here's what kills this approach: you're mining complaints without checking recency, a reddit thread from 2022 about shipping times doesn't matter if competitors fixed it in 2024. stale research makes you address solved problems while missing current pain points.

[ROAS Mutilator @ROASMutilator]

product bundles aren't about convenience. they're about margin expansion.
analyzed bundle performance across 40+ stores for 14 months.
found something nobody talks about:
bundles don't increase conversion rate. they increase profit per customer by 40-60%.
here's the actual math that makes bundles work (and why most people structure them wrong):
the bundle mistake everyone makes:
beginner approach:

product a: $29.99
product b: $24.99
bundle both: $49.99 (save $5)

they think: "small discount encourages bundle purchase"
result: 8% of customers take the bundle.
revenue impact: minimal.
here's what actually works:
the bundle economics need to hit three psychological thresholds:
threshold 1: discount feels meaningful (20%+ off) threshold 2: total price stays under impulse-buy ceiling threshold 3: perceived value exceeds price paid
most bundles fail because they optimize for margin over psychology.
the correct bundle math:
product a standalone: $35 product b standalone: $30 true bundle price: $48 (26% discount vs $65)
customers see:
saving $17 (feels substantial)
paying $48 (still under $50 impulse threshold)
getting $65 worth of products
result: 28% take rate on bundles.

but here's where it gets interesting:
you're not actually discounting 26%.
because customers were never going to buy both products separately.

they were going to buy product a for $35.
now they're buying the bundle for $48.
you just increased aov by $13 (37% increase) while the customer thinks they got a deal.
the backend math that matters:
standalone sale:
revenue: $35
cogs: $12
gross profit: $23
profit margin: 66%

bundle sale:
revenue: $48
cogs: $20 (both products)
gross profit: $28

profit margin: 58%

margin decreased 8 points.
but absolute profit increased by $5 per transaction.
on 1,000 orders, that's $5,000 more profit.
the advanced bundle strategy:

most stores offer one bundle option. that's leaving money on the table.
the three-tier bundle framework:
tier 1: product only
product a: $35
baseline offer
60% of customers choose this
tier 2: essential bundle

product a + product b: $48
positioned as "most popular"
32% of customers choose this

tier 3: complete bundle

product a + product b + product c: $67
positioned as "best value"
8% of customers choose this

blended aov: (0.60 × $35) + (0.32 × $48) + (0.08 × $67) = $42.02

vs single product aov: $35
20% aov increase with proper bundle structure.
the positioning rules:

rule 1: always show bundles, never hide them
most stores bury bundles in product description.
wrong.
bundle options should be prominent above fold:
radio buttons or cards
clear price comparison
savings highlighted in red/green
rule 2: use anchoring correctly

show the individual prices with strikethrough:
$35 + $30 = $65 $48
brain processes the $65 anchor first, making $48 feel like a steal.

rule 3: manufacture scarcity on bundles only
"only 12 bundles left at this price"
single product: unlimited bundles: "limited quantity"
creates urgency without seeming desperate.
when bundles actually work:
bundle chemistry matters.
good bundle combinations:
main product + consumable refill
main product + complementary tool
main product + protection/case
main product + upgrade/enhancement
bad bundle combinations:

two competing products
unrelated products
products that don't work together
example:

good bundle:

portable blender + recipe ebook + cleaning brush
all support the main use case
bad bundle:

portable blender + phone case
no logical connection

the data on bundle timing:
we tested showing bundles at different stages:

test 1: bundles on product page only

18% take rate
test 2: bundles on product page + checkout upsell

26% take rate
test 3: bundles everywhere (product page, cart, checkout, email)
31% take rate

conclusion: remind them multiple times.

most bundle revenue comes from cart and checkout, not product page.

the email bundle strategy:

post-purchase email sequence:

day 7: "you bought product a, here's how to get the most from it" day 14: "people who bought product a also love product b" day 21: "get product b for 25% off (bundle discount)"

15-20% of customers buy the complementary product within 30 days.
you just turned a one-time buyer into a repeat customer.
the numbers that prove this works:
store a (no bundles):
aov: $34
repeat purchase rate: 12%
6-month customer value: $41
store b (bundles implemented):
aov: $46 (35% increase)
repeat purchase rate: 12% (same)
6-month customer value: $55 (34% increase)

same products. same traffic. just bundle strategy.
$14 more profit per customer.
on 10,000 customers: $140,000 additional revenue.

implementation priority:

if you're doing under $10k/month: add one bundle option
main product + most obvious complement
20% discount on bundle

test for 30 days
if you're doing $10k-$50k/month: add three-tier structure
product only, essential bundle, complete bundle
test positioning and pricing
optimize based on take rates

if you're doing $50k+/month: dynamic bundling
personalized bundles based on browsing behavior
a/b test bundle combinations

optimize per product category

bundles aren't complicated.
most people just structure them wrong and give up when they don't work.

[ROAS Mutilator @ROASMutilator]

@therahulissar sauce Rahul!

great post

[ROAS Mutilator @ROASMutilator]

@d_grimripper fkin goat jordan

[ROAS Mutilator @ROASMutilator]

@LaddLaddLadd great mail

[ROAS Mutilator @ROASMutilator]

@williamkast_ yep, but the real mistake is testing price without testing perceived value simultaneously