Why is the most profitable price not the highest one buyers will pay?

Because profit is margin times volume. A higher price earns more on each accepted lead but sells fewer, and past the optimum the leads you lose cost more than the extra margin earns. The dynamic pricing article's example is the same trade-off: dropping a price from $50 to $45 can pay if it lifts accepted volume enough.

How does the value of an unsold lead change the answer?

The more an unsold lead still earns, through a secondary buyer, re-pinging or your own offers, the less a rejection costs you, so the best price rises. Your cost per lead does not move the best price at all; it only shifts every profit figure down by the same amount.

Where do the curve's inputs come from?

From your own data, not from this page. Randomize prices or floors within a small range on a share of traffic, per segment, and fit acceptance against price. The defaults are an illustrative example, not a benchmark, and real curves shift as buyers hit caps and demand changes.

All tools

Lead Price Optimizer

Given how acceptance falls as price rises, what price (or reserve floor) maximizes profit per lead?

Describe how buyers' acceptance falls as your price (or reserve floor) rises, what an unsold lead is still worth to you and what each lead costs. The calculator finds the price that maximizes expected profit per lead and compares it with your current price. The starting numbers are an illustrative example, not benchmarks.

Price at which half of leads are accepted$48Price sensitivity (smaller = sharper drop-off)$10.00Your current price or floor$50Value of an unsold lead (fallback)$0Your cost per lead$15

Profit / lead at current price

$7.51

45% accepted

Optimal price

$37.78

74% accepted

Profit / lead at optimal price

$12.78

Uplift vs. current

+$5.27

per lead (+70%)

The rule: keep raising the price while the margin you gain on accepted leads outweighs the leads you lose. With this curve that balance is reached where price − fallback = sensitivity ÷ share rejected: $37.78 − $0.00 = $10.00 ÷ 26%. Your current price is above it: a lower price sells enough extra leads to earn more in total.

Expected profit per lead (left axis)
Share of leads accepted (right axis)
Optimal price
Current price

This treats the price as a take-it-or-leave-it offer: an accepted lead pays exactly the price. In a second-price auction with a reserve, the winner often pays the second bid instead, so revenue is higher than shown, although the best floor follows the same rule when acceptance is read as one buyer's chance of valuing the lead above it. The curve here is assumed; estimate yours from randomized price or floor tests on a share of traffic, per segment, and recheck it as buyers and demand change.

How it works

A higher price earns more on each lead a buyer accepts, but fewer leads are accepted. This calculator takes a simple acceptance curve (the price at which half of leads are accepted and how sharply acceptance falls around it), what an unsold lead is still worth to you and your cost per lead, and finds the price or reserve floor that maximizes expected profit per lead. It compares that with your current price and charts profit and acceptance across prices. The default inputs are an illustrative example, not benchmarks.

The share of leads accepted at price p is modelled as a logistic curve that equals 50% at the half-acceptance price and falls more steeply the smaller the price sensitivity. An accepted lead pays the price; an unsold lead earns its fallback value (a secondary buyer, your own offers, or nothing). Expected profit per lead is acceptance × (price − fallback) + fallback − cost. The calculator searches prices numerically (a grid, then golden-section search) for the peak. At the peak, raising the price a little loses as much from rejected leads as it gains on accepted ones, which for this curve means price − fallback = sensitivity ÷ share rejected. Cost per lead shifts profit but not the best price; a higher fallback value raises it.

a(p) = \frac{1}{1 + e^{(p - p_{50})/s}}

\pi(p) = a(p)\,(p - v) + v - c

p^{*} - v = \frac{s}{1 - a(p^{*})}

where

$p$: price or reserve floor per lead
$a(p)$: share of leads accepted at price p
$p_{50}$: price at which half of leads are accepted
$s$: price sensitivity (smaller means a sharper drop-off)
$v$: value of an unsold lead (fallback)
$c$: your cost per lead
$\pi(p)$: expected profit per lead
$p^{*}$: price that maximizes expected profit

Frequently asked questions

Why is the most profitable price not the highest one buyers will pay?: Because profit is margin times volume. A higher price earns more on each accepted lead but sells fewer, and past the optimum the leads you lose cost more than the extra margin earns. The dynamic pricing article's example is the same trade-off: dropping a price from $50 to $45 can pay if it lifts accepted volume enough.
How does the value of an unsold lead change the answer?: The more an unsold lead still earns, through a secondary buyer, re-pinging or your own offers, the less a rejection costs you, so the best price rises. Your cost per lead does not move the best price at all; it only shifts every profit figure down by the same amount.
Can I use this to set a reserve price in an auction?: As a guide. The calculator treats the price as a take-it-or-leave-it offer, which is exactly how a floor works when one buyer clears it. In a second-price auction the winner often pays the second bid instead, so revenue is higher than shown, but for symmetric buyers the best floor follows the same rule when acceptance is read as one buyer's chance of valuing the lead above the floor (Myerson, 1981).
Where do the curve's inputs come from?: From your own data, not from this page. Randomize prices or floors within a small range on a share of traffic, per segment, and fit acceptance against price. The defaults are an illustrative example, not a benchmark, and real curves shift as buyers hit caps and demand changes.

Read the articleThe End of Fixed Bidding: Dynamic Pricing for Lead GenTier-based lead pricing overpays for the bottom of every tier. Learn how AI-driven dynamic pricing values each lead individually.

Bidding