How the Simulator Works by TheFBABros
← Back to the simulator

This page explains what the simulator is doing under the hood and how each input changes the result. You don't need to read this to use the tool — but if a number looks weird and you want to know why, this is where to look.

On this page

The 30-second overview

You give the simulator your product economics, your launch plan, and a competitive scan. It then simulates 10,000 different versions of your launch — each one rolls slightly different ramp speeds, Advertising Cost of Sale, and daily velocities within the ranges you provided. The output is a probability distribution: what fraction of those 10,000 launches turned profitable, when break-even tends to happen, and how much cash you'll need at the lowest point.

The point isn't to predict the future precisely. The point is to see how fragile your launch is. If 90% of simulated launches make money, you're robust. If only 40% do, your plan only works in the lucky scenarios.

Market Dominator — niche difficulty

The Market Dominator scans your top-10 competitors on Amazon (from a Helium10 Cerebro export) and tells you how saturated the niche is. It produces three outputs:

  1. A per-competitor color rating (5 tiers).
  2. A niche score from 0–100 (50 = neutral, 100 = wide open, 0 = fully entrenched).
  3. An auto-tune of the simulator's ramp/Advertising Cost of Sale/velocity assumptions to match the difficulty.
INPUT Helium10 Cerebro .xlsx export SCAN Top-10 rivals → 5 colour tiers SCORE 0 – 100 50 = neutral AUTO-TUNE ramp speed, ad cost, velocity Market Dominator pipeline feeds the simulator's ranges
The Cerebro scan is turned into a single 0–100 niche score, which then shifts the simulator's ramp / Advertising Cost of Sale / velocity ranges to match how hard the niche really is.

Per-competitor color rating

Each competitor's "Average %" is the average of (% of keywords they rank on page 1) and (% of total search volume their page-1 keywords cover). The higher that number, the more they own the niche.

Average %LabelColorWeight
0–20% Super Weak Green +5
21–40% Weak Lime green +3
41–60% Normal Yellow 0
61–80% Strong Orange −3
81–100% Super Strong Red −5

Niche score

Sum the 10 competitor weights and add 50:

niche_score = clamp( sum_of_10_weights + 50, 0, 100 )

So 10 Super Weak competitors → score 100 (wide open). 10 Super Strong → score 0 (fully entrenched). Mixed → somewhere in between.

How niche score affects the simulator

The simulator multiplies your ramp days, Advertising Cost of Sale, and velocity ceiling by deltas that depend on the niche score. The curve is asymmetric — the penalty for hard niches grows exponentially, while the boost for easy niches is linear and milder (because easy niches still need execution; hard niches genuinely punish you):

severity = (50 − niche_score) / 50 // [-1, +1] if severity ≥ 0 (penalty side, exponential): rampDays = 2.5 ^ severity acos = 2.0 ^ severity velocityHigh = 0.5 ^ severity else (boost side, linear, milder): rampDays = 1 − 0.25 · |severity| acos = 1 − 0.15 · |severity| velocityHigh = 1 + 0.10 · |severity|
Niche scoreRamp days ×Advertising Cost of Sale ×Velocity ×What it means
1000.750.851.10Wide open — boost
80 0.850.911.06Mostly weak
50 1.001.001.00Neutral baseline
30 1.441.320.76Tough niche
10 2.081.740.57Very tough
0 2.502.000.50Brutal — fully entrenched

Max-penalty override (the "anchored rival" rule)

If any single competitor is Super Strong, OR if there are 2+ Strong competitors, the simulator clamps the effective niche score to 0 — the brutal end of the curve — regardless of what the raw score says.

The reason: a single anchored rival ranking on page 1 for nearly every keyword will outspend, outrank, and outreview you on launch day. Two coordinated Strong rivals do the same. Even if the rest of the field is weak, that one wall is enough to bury a launch. The override is the simulator saying "I don't care that 9 of these competitors are easy — that one (or two) is the launch."

Niche traffic light vs niche score

The traffic light RED / ORANGE / GREEN at the niche level is a strict rule for the headline chip — any Strong/Super Strong competitor forces RED, any Normal forces ORANGE. The 0–100 niche score is the more nuanced metric and is what actually drives the math. They can disagree (e.g. score 90 but RED label because of one entrenched rival). When that happens, the max-penalty override above is what reconciles them.

Product economics inputs

These set your unit margins. They feed directly into break-even and final-profit math.

$30.00 Sale price −$4.50 Referral 15% −$5.00 Fulfillment fee −$8.00 Cost of Goods −$0.50 Storage $12.00 Margin Where every $30 sale goes (illustrative)
Each Amazon fee and your cost of goods peel off the top of the sale price; what's left is your per-unit contribution margin — the number that drives break-even and final profit.

Launch costs & timing

Order placed Shipped At Amazon LAUNCH Ramp → Year 1 day −88 day −58 day −21 day 0 $0 cash peak cash needed break-even Cash goes out long before the first sale — the trough is the budget you actually need.
You spend on inventory and freight months before launch day. The cash line bottoms out shortly after launch, then climbs back through break-even as sales ramp.

Ranges — Low / Likely / High

The simulator doesn't ask for one number; it asks for three on the variables that matter most:

Every one of the 10,000 simulations samples a value from each range using a triangular distribution centered on "Likely". So your ramp days might be 35 in one run, 60 in another, 95 in a third. That's why outcomes vary across runs — and why the answer is a probability, not a single number.

Apply a Market Dominator scan and these ranges automatically shift to match the niche difficulty, using the curve in Step 1.

Low Likely High rare rare most runs land near here ← each of the 10,000 runs draws one value
For ramp days, Advertising Cost of Sale and velocity you give three numbers. Every simulation samples from this triangular shape — clustered around "Likely", rarely at the extremes — which is why the result is a probability spread, not a single figure.

The Monte Carlo engine

For each of the 10,000 simulated launches:

  1. Sample one value each for ramp days, Advertising Cost of Sale, and velocity from your ranges.
  2. Sample one value each for niche-specific seasonality multipliers and conversion noise.
  3. Walk forward day by day for 365 days: compute organic + paid units sold, revenue, ad spend, Amazon fulfillment + storage fees, cost of goods, reorders, payout lag, and aged-inventory penalties.
  4. Track cash position, profit, and break-even day.
INPUT Your ranges Low / Likely / High ONE SIMULATED LAUNCH sample values, then walk 365 days forward units · revenue · fees · cash repeat × 10,000 OUTCOME DISTRIBUTION Best case Typical Worst case best / typical / worst across all runs
One run is a single what-if. Ten thousand runs — each sampling slightly different ramp, Advertising Cost of Sale and velocity — produce the worst / typical / best-case band you see on the cash-journey chart.

After 10,000 runs the simulator reports the distribution of outcomes — not the single "expected" answer. The typical (median), worst-case and best-case outcomes are the points you'll see on the cash-journey chart.

Iterations vs noise: 10,000 is the default sweet spot. Bumping to 20,000 doesn't change the conclusion meaningfully — it just smooths the chart. 5,000 is faster but the band edges get jittery.

Reading the outputs

Headline cards (the top row)

PROFIT PROBABILITY 72% of 10,000 runs in the black the headline — 70%+ is robust CASH NEEDED $48k Best case — survives 9 of 10 runs the budget you must actually put up BREAK-EVEN ~ Month 7 Typical run when net worth crosses $0
The top-row cards in plain terms: profit probability is the verdict, cash needed is the best-case budget you must fund, and break-even is the typical month you cross zero.
$0 profit → ← cash out peak cash needed break-even typical Year 1 profit Mo 0 Mo 3 Mo 6 Mo 9 Mo 12 Best case Typical Worst case
The cash journey is a J-curve: you fund inventory and ads first (the dip), hit a trough — that low point is the capital you actually need — then climb back through break-even into Year 1 profit. The shaded band is the worst-to-best spread across all 10,000 runs.

Stockout impact panel

Shown whenever any simulated run hit a stockout. Tells you what fraction of runs stocked out, the median + worst-case stockout duration, and the typical $ value of demand you missed. Stockouts hurt you twice — directly (lost sales) and indirectly (rankings drop, reviews dry up, the launch loses momentum). The simulator only models the direct hit, so the real-world cost is usually worse than the number shown. Bigger inventory buffers shrink this gap.

Holiday-quarter dependency one-liner

How much of your typical Year 1 profit comes from Nov + Dec. Three message tones based on share:

Charts

← Back to the simulator