A live unit-economics calculator for AI agencies scaling with paid ads. Move the sliders to find the single constraint holding back profitable growth — the metric that swings net profit hardest is the one to solve for first.
When an agency owner wants growth and uses ads as the vehicle, "should I spend more?" is the wrong question. The right question is: which constraint is capping my economics right now — acquisition cost, show rate, close rate, churn, delivery cost, or volume?
This tool makes that visible. Set your real funnel numbers as the baseline, then drag one slider at a time and watch every downstream metric recompute live. The lever that flips monthly net profit from red to black is your constraint.
| Slider | Meaning |
|---|---|
| CPQBC | Cost Per Qualified Booked Call — your blended ad cost to book one qualified call |
| Show Rate | % of booked calls that actually show |
| Close Rate | % of shown calls that close |
| Churn Rate (monthly) | % of clients lost per month |
| Monthly COGS | Delivery / fulfillment cost per client per month |
| Booked Calls / month | Top-of-funnel volume |
Plus three offer/target fields: monthly price, minimum term, and your target LGP:CAC ratio.
Every output is derived from the sliders — no magic numbers.
closeProb = showRate × closeRate // a booked call → a client
newClients/mo = bookedCalls × closeProb
adSpend/mo = bookedCalls × CPQBC
CAC = CPQBC ÷ closeProb // = adSpend ÷ newClients
retention (mo) = 1 ÷ churnRate
monthlyRevenue = newClients × price
monthlyCOGS = newClients × COGS
monthlyNetProfit= monthlyRevenue − monthlyCOGS − adSpend
monthlyROAS = monthlyRevenue ÷ adSpend
dayOneCashROAS = price ÷ CAC
lifetimeGrossProfit (LGP) = (price − COGS) × retention
profitPerClient = LGP − CAC
grossMargin = (price − COGS) ÷ price
profitMargin = profitPerClient ÷ (price × retention)
LGP:CAC = LGP ÷ CAC
recommendedPrice = (targetRatio × CAC ÷ retention) + COGS // price to hit your target LGP:CAC
LGP:CAC is the headline health metric. A 3:1 is survivable; 10:1+ means you can pour fuel on the fire. If you're below target, the Recommended Price Point card tells you the price that would get you there at your current funnel metrics — or you go fix a slider instead.
It's a static site — no build step.
# any static server works
python3 -m http.server 4178
# then open http://localhost:4178Deploys to Vercel as a static site with zero config (vercel.json included).
The repo is connected to the Vercel project, so every push to master
auto-deploys to production.
Live: https://ad-economics-calculator.vercel.app
index.html— structurestyle.css— dark dashboard themeapp.js— the model + slider wiring (this is where the math lives)