How do I calculate what lead fraud costs?

Count the fraudulent leads that get past your checks, then add up what each one costs: the price you paid, the time spent verifying and calling it, and any loss it causes later. In the default example, 2,000 leads at $35 with 8% fraud and 40% of it caught let 96 fraudulent leads through each month. They cost $6,144: $3,360 in wasted lead spend, $384 in wasted handling and $2,400 in downstream losses, about 9.1% of lead spend or $3.34 per genuine lead.

What is a screening step worth?

The fraud cost it avoids minus what it costs to run on every lead. It pays for itself while its cost per lead is below the break-even: the share of fraud it catches times the fraud cost, divided by leads bought. In the default example, screening at $0.50 a lead that catches 70% of the fraud getting through avoids $4,301 a month for $1,000, a net saving of $3,301, and would break even at about $2.15 a lead. It is the same principle as in the article: spend verification costs where the expected loss avoided exceeds the cost and the friction.

What should I enter as the downstream loss?

Your own average loss per fraudulent lead that gets through, across all of them, not the size of one bad case. Many fraudulent leads simply never convert, while some may go on to a funded fraud loss or a chargeback, so spread those losses across every fraudulent lead. The default is an illustrative example, not a benchmark.

What does this leave out?

Friction and false positives. The model assumes screening never rejects a genuine applicant, but in practice step-up checks lose some real customers, and blocking all automation can turn away legitimate AI agents shopping on a borrower's behalf. It also ignores time spent on fraud before it was caught, and harder-to-price effects such as strained relationships with downstream partners.

All frameworks

Lead Fraud Cost Calculator

What does lead fraud really cost once wasted spend, handling and downstream losses are counted, and what is screening worth?

Enter what you buy, how much of it is fraudulent, how much of that fraud you already catch, and what a fraudulent lead costs you once it gets through. The calculator shows the monthly cost of fraud and what a screening step is worth. The starting values are an illustrative example, not a benchmark.

Leads bought per month2,000Price per lead$35Leads that are fraudulent8%Fraud you already catch (rejected before paying, or returned and credited)40%Handling cost per lead worked (verification, calls)$4.00Average downstream loss per fraudulent lead that gets through (e.g. a funded fraud loss or chargeback)$25Add a screening step on every leadScreening cost per lead screened$0.50Share of the fraud getting through that screening catches70%

Fraudulent leads getting through

96 of 160

Monthly cost of fraud

$6,144

As a share of lead spend

9.1%

Per genuine lead

$3.34

Net monthly saving from screening

$3,301

Break-even screening cost per lead

$2.15

Monthly cost of fraud

Today$6,144

With screening (including its cost)$2,843

Wasted lead spend$3,360
Wasted handling$384
Downstream losses$2,400
Screening cost$1,000

Amounts in the list are today's, except the screening cost.

Fraud you catch is treated as costing nothing, because it is rejected before you pay or credited back; any time spent on it before it was caught is not counted, so the result is conservative. A fraudulent lead that gets through costs its price, the handling spent on it and the average downstream loss you entered, which spreads occasional large losses across every fraudulent lead. Screening is charged on every lead bought, and the model assumes it never rejects a genuine lead; in practice, wrongly rejected applicants are a cost too.

How it works

A fraudulent lead costs more than its price. If it gets past your checks, you also pay to verify and call it, and some fraud goes on to cause a loss further down the funnel, such as a funded fraud loss or a chargeback. Enter your volume, price, fraud share, how much of it you already catch and what fraud costs once it gets through to see the monthly cost of fraud, split into wasted lead spend, wasted handling and downstream losses. Turn on a screening step to see its net monthly saving and the most it could cost per lead and still pay for itself. The default inputs are an illustrative example, not benchmarks.

Fraudulent leads are leads bought times the share that is fraudulent. The fraud you already catch is rejected before you pay or returned and credited, so it is treated as costing nothing; the rest gets through. Each fraudulent lead that gets through costs its price (wasted lead spend), the handling cost of verifying and calling it (wasted handling) and the average downstream loss you enter (downstream losses). The cost of fraud is the sum of the three. It is shown as a share of what you pay for leads (every lead except the fraud you caught) and per genuine lead. A screening step is charged on every lead bought and catches a share of the fraud that gets through today; caught leads cost nothing further. Its net saving is the fraud cost it avoids minus its cost, and it breaks even when its cost per lead equals the fraud cost it avoids divided by leads bought. The model assumes screening never rejects a genuine lead.

F = N\,f\,(1 - c)

X = F\,(p + h + d)

x_S = \frac{X}{N\,(1 - f\,c)\,p}

x_G = \frac{X}{N\,(1 - f)}

V = k\,X - N\,s

s^{*} = \frac{k\,X}{N}

where

$N$: leads bought per month
$p$: price per lead
$f$: share of leads that are fraudulent
$c$: share of fraud you already catch
$h$: handling cost per lead worked
$d$: average downstream loss per fraudulent lead that gets through
$F$: fraudulent leads that get through
$X$: monthly cost of fraud
$x_S,\ x_G$: cost of fraud as a share of lead spend, per genuine lead
$s,\ k$: screening cost per lead, share of fraud getting through that it catches
$V$: net monthly saving from screening
$s^{*}$: break-even screening cost per lead

What a real decision adds

You rarely know the true fraud rate. Much fraud is confirmed only weeks later, through returns, chargebacks or a lender's findings, and some is never labeled at all, so today's figure is an estimate that lags the attack.
Fraud is not one rate. It varies by source, publisher and time, and it shifts when fraudsters adapt to whatever checks you add, so a catch rate measured last quarter may not hold this one.
Screening has a second cost: friction and false positives. Each extra check turns away some genuine applicants, and legitimate AI agents acting for real borrowers can look like bots, so the cheapest check per lead is not always the cheapest overall.
Downstream losses are lumpy and late. A few fraudulent leads that reach funding or cause a chargeback can outweigh many that simply never convert, and they surface long after the lead was paid for.

Working through this decision for your business?

Talk to us

Frequently asked questions

How do I calculate what lead fraud costs?: Count the fraudulent leads that get past your checks, then add up what each one costs: the price you paid, the time spent verifying and calling it, and any loss it causes later. In the default example, 2,000 leads at $35 with 8% fraud and 40% of it caught let 96 fraudulent leads through each month. They cost $6,144: $3,360 in wasted lead spend, $384 in wasted handling and $2,400 in downstream losses, about 9.1% of lead spend or $3.34 per genuine lead.
What is a screening step worth?: The fraud cost it avoids minus what it costs to run on every lead. It pays for itself while its cost per lead is below the break-even: the share of fraud it catches times the fraud cost, divided by leads bought. In the default example, screening at $0.50 a lead that catches 70% of the fraud getting through avoids $4,301 a month for $1,000, a net saving of $3,301, and would break even at about $2.15 a lead. It is the same principle as in the article: spend verification costs where the expected loss avoided exceeds the cost and the friction.
What should I enter as the downstream loss?: Your own average loss per fraudulent lead that gets through, across all of them, not the size of one bad case. Many fraudulent leads simply never convert, while some may go on to a funded fraud loss or a chargeback, so spread those losses across every fraudulent lead. The default is an illustrative example, not a benchmark.
What does this leave out?: Friction and false positives. The model assumes screening never rejects a genuine applicant, but in practice step-up checks lose some real customers, and blocking all automation can turn away legitimate AI agents shopping on a borrower's behalf. It also ignores time spent on fraud before it was caught, and harder-to-price effects such as strained relationships with downstream partners.

Read the articleReal-Time Lead Verification: Reducing Fraud with MLLead fraud keeps evolving. See how ML fraud detection keeps pace with bots, fraud farms, and now AI-generated applications.

Lead Gen