The 'Crazy Method' for Facebook Ads Scaling in 2026: How to Duplicate Ad Sets Without Killing Your ROAS

TL;DR: The most frustrating thing about running Meta ads isn't getting no sales—it's finally getting a profitable ad, trying to scale it, and watching your CPA spike and ROAS crash. The "Crazy Method" core logic: duplicate a profitable ad set 3-5 times, put them all in the same CBO, and double the budget. The essence is giving Meta's algorithm multiple "lottery draws" to land in different audience pools, finding more high-converting audiences, achieving stable scaling instead of hard budget increases. But note: only ads with sufficient profit margin, stable conversions, existing volume, and clear understanding of why they work qualify for the Crazy Method.

---

The Most Frustrating Moment in Meta Ads: Budget Increase = Performance Collapse

You've definitely experienced this scenario:

• Finally got a profitable ad running
• ROAS 3.5, stable daily conversions
• You excitedly increase daily budget from $100 to $200
• Next day, CPA is up 40%, ROAS dropped to 2.1
• You hold on, observe for two more days—it gets worse
• Finally have to sheepishly lower the budget back
• Watch helplessly as this ad slowly declines

This is the most frustrating part.

No sales? You just turn it off. But seeing a money-making ad that you don't know how to scale—that's like holding a gold mine but not being able to extract it.

Today I'll introduce a method that was extremely popular during the dropshipping era—the "Crazy Method"—specifically designed to solve this problem.

Before we start: If you're managing multiple ad campaigns and don't know which ad sets qualify for Crazy Method scaling, Adfynx's AI Assistant can analyze your entire account structure in seconds, tell you which ad sets meet scaling criteria, which creatives are fatiguing, and how budget should be allocated. Try it free—no credit card required.

---

What Is the "Crazy Method"?

The Operation Is Simple

1. Find an already profitable ad set

2. Duplicate it 3-5 times (exact copies)

3. Put all duplicated ad sets into the same CBO

4. Double the budget and run

That's it.

Why Is It Called the "Crazy Method"?

Because this method was crazy popular during the 2020-2022 dropshipping boom, and many people used it to achieve rapid scaling.

Now when running Meta ads for branded DTC sites, the Crazy Method still works, but you must understand the underlying logic—otherwise it's easy to crash and burn.

---

The Underlying Logic of the Crazy Method: Give the Algorithm Multiple Lottery Draws

A Key Characteristic of Meta's Algorithm

Meta's algorithm heavily relies on "the first batch of converters."

When your ad starts running, the system judges "what your customers look like" based on the earliest converting users.

• First batch is moms → keeps finding moms
• First batch is students → all students from then on
• First batch is male users → continues targeting males

This Creates a Problem

Suppose you're selling baby products, and the algorithm's first batch captures "male users buying gifts for newborn nieces"—then it will keep finding this demographic.

But your product's main purchasing power is actually moms, primarily female.

Result:

• Good conversions at first (hit a small segment of willing male buyers)
• Performance gradually declines from day two (this audience pool is too small, quickly saturated)

The Essence of the Crazy Method

Give the algorithm multiple "lottery draws."

Duplicate the same ad set 5 times, each starts differently:

• Some might first convert moms (gold mine)
• Some might first convert male relatives (mediocre)
• Some might first convert grandparents (niche but high value)

The system might land in different audience pools. Some pools are mediocre, some are gold mines.

Your job: keep the gold mines, shut down the others.

This is why the Crazy Method enables scaling: instead of hard budget increases, open multiple entry points to find more high-converting audiences.

Data insight: Want to know which duplicated ad sets actually hit high-converting audiences? Adfynx's Creative Analyzer breaks down performance by creative, placement, and audience—helping you accurately identify gold mines vs. duds.

---

Common Crazy Method Failure Cases

The Crazy Method works, but not everyone can use it.

I've seen too many failures, mostly dying in these areas:

Failure Case 1: The Ad Wasn't Profitable to Begin With

Wrong thinking:

"My ad has ROAS 2.1, break-even is 2.0, just a bit short—maybe I can use the Crazy Method to boost it?"

Why it fails:

The Crazy Method only amplifies your existing results, it doesn't change the results themselves.

• You're barely profitable now, after scaling you'll likely get: higher costs, bigger fluctuations, more stress
• Ads without clear profit margin—don't touch them

Correct approach:

First optimize creative and audience, get ROAS up to at least 2.8+, then consider scaling.

Failure Case 2: Using the Same Creative Repeatedly

Wrong approach:

One product, one creative set, opening 2-3 Crazy Method CBOs.

Why it fails:

Might work short-term, but you'll quickly see costs rising, performance declining.

Simple reason: you're competing with yourself for traffic.

Correct approach:

• Each Crazy Method CBO uses different creative angles
• Or targets different audience strategies (cold vs. warm vs. retargeting)

Failure Case 3: Too Few Ad Sets

Wrong approach:

Only opening 1-2 ad sets—that's not the Crazy Method, that's just regular duplication.

Why it fails:

• Too few lottery draws, can't hit different audience pools
• Not enough redundancy—one ad set fatigues and the whole campaign collapses

Correct approach:

• Minimum 3 to start
• 4-5 is ideal
• Over 10 gets messy (budget spread thin, data unstable)

Failure Case 4: Don't Understand CBO at All

Wrong thinking:

"I heard the Crazy Method is great, let me try it."

Why it fails:

If you don't know CBO logic and have no experience managing CBO campaigns, using the Crazy Method is just random experimentation.

Correct approach:

First master CBO basics, understand how Meta allocates budget between ad sets, then try the Crazy Method.

Learning resource: Not sure if your CBO setup is correct? Adfynx's AI Assistant can analyze your CBO structure, tell you if budget allocation is reasonable, which ad sets are underperforming, and how to optimize settings.

---

What Kind of Ad Sets Qualify for the Crazy Method?

Here are the standards—check against them:

Standard 1: Sufficient Profit Margin

Specific requirement:

If your break-even ROAS is 2.0, you need to be running at least 2.8 or higher.

Why?

This way, even if costs increase slightly after scaling, you won't lose money.

Doesn't meet standard:

ROAS 2.1, 2.2—polish your creative first.

Standard 2: Stable Conversions, Not Occasional Spikes

Specific requirement:

• Consecutive 7-10 days with daily conversions
• CPA fluctuation is small (within ±20%)
• Not 10 sales today, 0 tomorrow

Why?

Predictable = replicable. Occasional spikes might just be luck, not systematic success.

Standard 3: Already Has Volume

Specific requirement:

• Ad set daily budget spend has reached a certain level (e.g., $200)
• Stable 4+ conversions daily

Why?

Shows the system has found its stride, not just lucky hits. Sufficient data density allows the algorithm to optimize effectively.

Standard 4: You Understand Why It Works

Specific requirement:

• Is it the creative?
• The angle?
• Or did it happen to hit a certain audience?

Why?

If you can't explain it yourself, don't scale yet. What you duplicate might just be an illusion.

---

Crazy Method Implementation Steps

Step 1: Select Ad Sets Meeting Standards

Use the 4 standards above to filter.

Checklist:

• [ ] ROAS ≥ 2.8 (or 40%+ above break-even)
• [ ] Consecutive 7-10 days of stable conversions
• [ ] Daily budget ≥ $200, daily conversions ≥ 4
• [ ] Clear understanding of why this ad set works

Step 2: Duplicate 3-5 Ad Sets

Operation:

In Meta Ads Manager, select the high-performing ad set, click "Duplicate," create 3-5 copies.

Note:

• Don't modify any settings
• Identical audience, creative, bidding strategy

Step 3: Create New CBO Campaign

Settings:

• Budget: 2-3x the original ad set's daily budget
• Optimization goal: Same as original ad set (usually Purchase)
• Bidding strategy: Lowest Cost or Cost Cap

Example:

• Original ad set daily budget: $200
• New CBO campaign budget: $400-600
• Contains: 3-5 duplicated ad sets

Step 4: Observe for 3-5 Days

Key metrics:

• Each ad set's CPA
• Each ad set's ROAS
• Budget distribution across ad sets

Meta will automatically allocate budget to best-performing ad sets.

Step 5: Keep Gold Mines, Shut Down Others

Judgment criteria:

• Keep: Ad sets with CPA below target, ROAS above target
• Shut down: Ad sets with high CPA, low ROAS, low spend

Typical results:

• Out of 3-5 ad sets, 1-2 are gold mines (excellent performance)
• 1-2 are mediocre
• 1-2 perform poorly

Only keep the gold mines, shut down the rest.

Automated monitoring: Manually checking each ad set's performance is time-consuming. Adfynx's automated reporting sends instant alerts when CPA or ROAS moves outside target ranges—saving time and preventing costly mistakes.

---

How to Pair the Crazy Method with ASC?

Many people ask: How do you use the Crazy Method together with ASC (Advantage+ Shopping Campaigns)?

Core Principle

For scaling, ASC is the main force, Crazy Method is supporting.

Specific approach depends on your situation:

Scenario 1: Multi-SKU Store (e.g., Fashion Products)

Product characteristics:

T-shirts, pants, hats—different categories correspond to different audiences.

Strategy:

• ASC handles the broad market (broad audience, let algorithm auto-optimize)
• Crazy Method mines niche high-converting audiences (e.g., specific style T-shirts)

Works well together.

Scenario 2: Single Product, Different Creative Angles

Example:

Same product:

• One creative targets outdoor short trips
• Another creative targets RV travel

Strategy:

Prioritize ASC + different creative angles for testing, Crazy Method isn't first choice.

Why?

ASC itself can test different creatives, no need to add complexity with Crazy Method.

Scenario 3: Already Have Stable Structure, Want More Growth

Current state:

• You already have well-performing ASC and CBO campaigns
• Now want to squeeze out additional growth

Strategy:

Open one Crazy Method as a scaling tool, works great.

This is the Crazy Method's sweet spot.

---

Advanced Crazy Method Techniques

Technique 1: Use Different Creative Angles

Approach:

Each duplicated ad set uses different creative hooks or different ad copy angles.

Benefits:

• Increases probability of hitting different audience pools
• Avoids competing with yourself for traffic

Technique 2: Phased Launch

Approach:

Don't open all 5 ad sets at once—start with 3, observe for 3 days, then open 2 more.

Benefits:

• Reduces risk
• Easier to identify which ad sets are true gold mines

Technique 3: Set Minimum Spend Limits

Approach:

In CBO, set Ad Set Spending Limits (minimum spend) for each ad set.

Benefits:

Ensures each ad set gets sufficient exposure opportunity, won't be completely ignored by Meta.

Recommendation:

Each ad set minimum spend = CBO total budget / number of ad sets × 0.5

Example:

• CBO total budget: $500
• Number of ad sets: 5
• Each ad set minimum spend: $500 / 5 × 0.5 = $50

Technique 4: Monitor Frequency

Key metric:

If an ad set's Frequency rapidly rises to 3+, the audience pool is small and easily saturated.

Action:

• Shut down this ad set
• Or refresh creative

Frequency monitoring: Adfynx can automatically track each ad set's frequency and send alerts when frequency exceeds thresholds—helping you act before creative fatigue causes cost spikes.

---

Crazy Method FAQ

Q1: Will the Crazy Method reset the learning phase?

Answer: Yes, but that's expected.

Each duplicated ad set is new and will enter learning phase. But because they're based on already-validated successful ad sets, they typically pass through learning phase quickly.

Q2: What budget scale suits the Crazy Method?

Recommendation:

• Minimum: Ad sets with $200+ daily budget
• Optimal: Ad sets with $300-500 daily budget

Why?

Budget too small means each duplicated ad set gets insufficient budget to generate stable data.

Q3: Can you use the Crazy Method with ASC?

Answer: Not recommended.

ASC itself is a highly automated campaign—duplicating ASC campaigns doesn't make much sense.

The Crazy Method is better suited for traditional CBO campaigns.

Q4: How long until the Crazy Method shows results?

Typically: 3-5 days.

• First 3 days: Meta allocates budget, tests different ad sets
• After 3-5 days: Performance differences become clear, can make decisions

Q5: What if the Crazy Method fails?

If all duplicated ad sets perform poorly:

1. Pause the CBO campaign

2. Return to original ad set

3. Analyze reasons:

- Is the original ad set already fatiguing?

- Has the market environment changed?

- Is the creative saturated?

Don't force it, cut losses quickly.

---

The Essence of the Crazy Method: Only Replicable Things Can Scale

The Crazy Method isn't magic, and it's not mandatory.

It's just a scaling tool:

• Used in the right place, helps you make more money
• Used in the wrong place, makes you lose faster

Core principle:

What can be replicated and scaled must be already validated by the market.

If your ads are still in testing phase, build the foundation first:

1. Find stable profitable ad sets

2. Ensure sufficient profit margin

3. Understand why they work

4. Establish stable creative supply

Once you have a stable money-making ad, the Crazy Method naturally becomes useful.

---

Action Plan: Start Testing the Crazy Method This Week

Step 1: Audit Your Account

• [ ] Find all ad sets with ROAS ≥ 2.8
• [ ] Filter for ad sets with consecutive 7 days of stable conversions
• [ ] Confirm daily budget ≥ $200

Step 2: Select 1 Ad Set to Test

• [ ] Choose the 1 ad set that best meets standards
• [ ] Duplicate 3-5 times
• [ ] Create new CBO campaign

Step 3: Set Budget and Monitoring

• [ ] CBO budget = original ad set budget × 2-3
• [ ] Set minimum spend limit for each ad set
• [ ] Set CPA and ROAS alerts

Step 4: Observe for 3-5 Days

• [ ] Check each ad set's performance daily
• [ ] Record CPA, ROAS, spend distribution
• [ ] Identify gold mine ad sets

Step 5: Optimize and Expand

• [ ] Shut down poor-performing ad sets
• [ ] Keep gold mine ad sets
• [ ] Consider further increasing CBO budget

---

Summary: Give the Algorithm Multiple Lottery Draws to Find Real Gold Mines

Running Meta ads in 2026, creative and audience are the two cores.

The Crazy Method's value lies in:

1. No hard budget increases (avoid triggering learning phase reset)

2. Give algorithm multiple lottery draws (land in different audience pools)

3. Find truly high-converting audiences (gold mines)

4. Achieve stable scaling (instead of cost spikes)

But remember:

• Only already-profitable ads qualify for the Crazy Method
• Profit margin must be sufficient (ROAS ≥ 2.8)
• Must have stable conversions (consecutive 7-10 days)
• Must understand why it works (not luck)

The Crazy Method isn't about burning money crazily—it's about rationally giving the algorithm more opportunities to find real growth potential.

---

Related Resources

Want to track all ad set performance in one dashboard? Try Adfynx for Free — AI-powered Facebook Ads analytics that makes multi-ad-set management simple.

Looking for more scaling strategies? Check out 2026 Facebook Ads Scaling Logic: The Four-Campaign Method That Prevents Budget Increases from Killing ROAS for advanced account architecture.

Why is your ASC campaign failing? Read our guide on Why ASC Campaigns Fail and How to Fix Them.

Need help with budget planning? Use our free Facebook Ads Cost Calculator to model spend, ROAS, and CPA across different scaling stages.

Want to understand Meta's latest algorithm changes? Read Meta Andromeda Algorithm 2026: Complete Guide.

Struggling with creative fatigue? Check out Facebook Ads Creative Strategy Under Meta's Andromeda Algorithm.