'What happens if I go from ₹50k to ₹2L a month?' Spend rarely scales for free — CPA usually rises as you push into a bigger audience. Model the diminishing returns so you scale into profit, not into a wall.
Most accounts sit here · curve exponent k = 0.2
Now
At target
At ₹2.00L, projected CPA rises to ₹1,320 (from ₹1,000, +32%).
You get 152 customers — about 24% fewer than the 200 a flat-CPA (naïve) forecast would promise.
| Spend band | Marginal CPA | vs now | Verdict |
|---|---|---|---|
| ₹50,000 → ₹87,500 | ₹1,328 | 1.3× | acceptable |
| ₹87,500 → ₹1.25L | ₹1,452 | 1.5× | acceptable |
| ₹1.25L → ₹1.63L | ₹1,543 | 1.5× | acceptable |
| ₹1.63L → ₹2.00L | ₹1,617 | 1.6× | acceptable |
Marginal CPA = the cost of each extra customer in that band. Once it drifts well above your current CPA, each additional rupee is buying pricier customers — that's where scaling stops paying.
Model: CPA(spend) = current CPA × (spend ÷ current spend)^k. Customers = spend ÷ CPA. Profit is revenue × margin minus ad spend.
Yes — completely free, no sign-up. It runs entirely in your browser; nothing is sent to a server or stored.
It assumes CPA rises with spend along a simple curve: CPA(spend) = current CPA × (spend ÷ current spend)^k, where k is the aggressiveness you choose (roughly 0.1 gentle, 0.2 typical, 0.35 steep). Customers = spend ÷ CPA. It's deliberately simple and transparent — the exponent k is shown so you know the assumption.
No — and it says so clearly. It's a planning model, not a guarantee. Real scaling curves depend on audience saturation, creative fatigue, seasonality and competition. Use it to pressure-test a scaling decision and set expectations, then verify with a controlled budget ramp.
It's the cost of each extra customer within that slice of spend — not the average. As you scale, marginal CPA climbs faster than average CPA. When it drifts well above your current CPA, the last rupees of budget are buying much pricier customers, which is usually where scaling stops being worth it.