A bought lead is a customer inquiry purchased from a third party: a lead generation platform, an aggregator, a directory, or a broker that typically resells the same inquiry to several competing providers. The sticker price per lead is usually modest. The true cost of bought leads is the effective cost per acquired customer after contact rates, lead sharing, close rates, and margin effects are counted, and it routinely runs to several multiples of the sticker price. This guide walks through that math, explains why it is so rarely presented plainly, and lays out the sequence for shifting acquisition toward channels the company owns.
It is written for sponsors, portfolio CEOs, and founders whose companies buy a meaningful share of their demand, and for the boards asking why customer acquisition cost keeps rising.
The quoted price of a lead is the entry fee, not the cost. Start with sharing: many platforms sell the same inquiry to three, four, or five competing providers, so the quoted price buys a seat in an auction, not a customer conversation. Then contact rates: a purchased inquiry decays by the minute, and a company without disciplined speed-to-contact reaches only a fraction of what it paid for. Then close rates: a shopper who filled out a platform form is comparison shopping by construction, and closes at a fraction of the rate of a referred or direct inquiry. Finally margin: deals won in a five-way price competition are won by discounting.
Multiply it through. A lead priced in the low tens of dollars, shared with several competitors, contacted successfully some of the time, and closed at a bought-lead close rate produces an effective cost per acquired customer that is many times the sticker price, before counting the margin given up to win the work. Whenever we run this arithmetic inside a portfolio company, channel by channel, the bought-lead channel that looked cheapest per unit turns out to be among the most expensive per customer. The point is not that the sticker price is a lie. It is that the sticker price is the only number the channel reports on its own.
Search for what leads cost and nearly every first-page result is published by a company that sells leads, outreach software, or lead generation services. The benchmarks are real enough, but the framing is not disinterested: the vendor's business depends on the reader concluding that buying demand is normal and that the fix for poor results is better targeting, meaning more spend on the platform. Advice on whether to buy leads, purchased from the people selling them, has the same structural problem as diligence performed by the seller.
An operator's independent accounting starts from a different question: not what a lead costs, but what a customer costs through each available channel, and what each channel does to the company's position over time. On that accounting, bought leads are a permanent rental expense with rising rates, and the alternative is not better lead buying but building channels the company owns.
The deeper cost of bought leads is structural. Money spent on a purchased inquiry buys that inquiry and nothing else; the platform keeps the customer relationship, the data, and the pricing power, and the price per lead tends to rise as more competitors join the auction. Money spent building owned channels, the company's website and local presence, its referral engine, its reviews, its email list, buys an asset that produces inquiries at declining marginal cost. One channel is rent; the other is equity. The full version of this argument, including how dependence on aggregators develops and how to unwind it, is in the aggregator dependence guide.
Dependence also compounds quietly. A company that fills its calendar with bought leads underinvests in the owned channels precisely because the calendar is full, and each year of underinvestment raises the switching cost. The platforms understand this arithmetic well; it is why the first leads are cheap.
Blended customer acquisition cost hides the problem. A single blended CAC number averages the expensive bought-lead customers with the nearly free referral customers and reports something unremarkable. The board-grade version is channel-level: for each acquisition channel, the fully loaded spend, the customers actually won, the effective CAC, the average margin on those customers, and the trend. That table, run quarterly, is usually the entire argument for change. How to fund the shift it implies is covered in the marketing budget guide, and how to keep the numbers honest in the board reporting guide.
The channel table has a second use: it surfaces the channels the company has never measured. In one PE-backed services company we audited, 38 percent of revenue arrived through word of mouth that no system tracked, while paid channels with a fraction of that contribution consumed the budget and the meetings. Companies buying leads at scale almost always undercount their cheapest channel, because nobody sends an invoice for a referral.
Cold turkey is the wrong move. A company that cancels its lead purchases before its owned channels can replace the volume converts a margin problem into a revenue problem. The sequence that works runs in overlapping phases: first instrument, so every inquiry is attributed to its true source and the channel table above is real; then build, investing in the owned channels with the clearest path for the specific business; then step down purchased volume deliberately as owned volume replaces it, keeping the bought channel only where it still clears the effective-CAC bar.
Run in that order, the shift is measurable and fast enough to matter inside a hold period. In one multi-site consumer services rollup we worked with, shifting the mix toward owned acquisition cut customer acquisition cost by 60 percent while the owned share of new customers tripled. The mechanics of that shift are detailed in the CAC reduction guide. Nothing about the change required abandoning paid demand overnight; it required knowing what each channel truly cost and reallocating with intent.
Bought leads are a tool, and there are jobs for it. Filling capacity troughs is one: purchased demand can be switched on and off in a way owned channels cannot, which has real value for a business with expensive idle capacity. Entering a new market is another: buying inquiries while the owned presence matures converts waiting time into learning and revenue. Validating demand for a new service line is a third. In each case the channel is being used for its flexibility, deliberately and temporarily, with the effective cost counted honestly.
What does not make sense is the default configuration: bought leads as the permanent structural core of acquisition, priced by sticker, unbenchmarked against owned alternatives, in a company that has never engineered its referral engine or its local search presence. That configuration persists not because it is economic but because it is easy, and it is one of the most reliable sources of CAC improvement available to new ownership.
The sticker price is only the entry fee. The true cost is the effective cost per acquired customer: the price per lead divided through by contact rate and close rate, adjusted for the margin discounting needed to win shared leads. Because many platforms sell the same inquiry to several competing providers, and because platform shoppers close at lower rates than referred or direct inquiries, the effective cost per customer routinely runs to several multiples of the quoted per-lead price.
Three reasons compound. The same inquiry is typically sold to multiple competitors, so every conversation is a price competition. Purchased inquiries decay quickly, and any delay in contact sharply reduces the chance of a conversation at all. And a customer who filled out a platform form is comparison shopping by construction, unlike a referred customer who arrives with borrowed trust. Each effect lowers the close rate, and together they raise the true cost per customer well above the sticker price.
Not abruptly. Canceling purchased demand before owned channels can replace the volume trades a margin problem for a revenue problem. The working sequence is to instrument attribution so every channel's effective cost per customer is known, build the owned channels with the clearest path for the business, and then step purchased volume down as owned volume replaces it. Bought leads can remain useful at the margin for filling capacity troughs or entering new markets, used deliberately rather than by default.
Bought leads are rented demand: the platform owns the customer relationship, the data, and the pricing power, and the spend buys inquiries one at a time at rates that tend to rise. Owned acquisition channels, including the company's website and local search presence, its referral engine, its reviews, and its customer list, are assets: investment in them accumulates, and they produce inquiries at a declining marginal cost. Rented demand stops the day the spend stops. Owned demand compounds.
Have a revenue problem the board is asking about? Start a conversation.