Cost per lead and cost per sale: how to interpret them

Two numbers that are easily confused

One report says each lead cost you little and another says each sale cost you a lot. Both can be true at the same time, because they measure different things. Cost per lead measures how much it costs you to get someone to raise their hand. Cost per sale measures how much it costs you to get someone to buy. The second is what covers payroll; the first helps you know if the funnel is filling.

The formulas

Decide from the start what you include in “cost.” If you include only media spend, the number serves to compare campaigns with each other. If you include fees and tools, it gets closer to the real cost of acquiring a customer. Neither option is wrong; what’s wrong is mixing them between months.

What counts as a lead

A click on the WhatsApp button is not the same as a conversation with a real need. Define in writing what a lead is in your company: for example, a contact who requested information about a product or service you sell. Ad platforms call the action you configure a “conversion,” so their count depends on that configuration.

Time windows

In B2B sales, a March lead can become a sale in May. If you divide May’s cost by May’s sales, you mix leads from different months. There are two ways to handle this:

By cohort: take the leads from one month and wait to see how many convert, even if they close later. It’s more exact, but it takes time.

By period: divide cost and sales for the same period. It’s faster and acceptable if your sales cycle is short or stable.

Google Ads has its own “conversion window,” the period after the click during which it records a conversion. Consider that window when you compare to your data.

Procedure

1. Write your definition of lead and sale.
2. Decide which costs you include and keep it consistent.
3. Choose whether to measure by cohort or by period.
4. Record cost, leads and sales by month and by channel in the template.
5. Calculate CPL, CPV and closing rate manually or in your spreadsheet.
6. Compare against your own previous months, not against figures from other companies.

Applied example

How to read the results

• If CPL falls and CPV rises, you’re probably attracting lower-quality contacts.
• If CPL rises and CPV falls, you may be filtering better.
• If both rise, check campaigns, landing page and sales follow-up.
• With few sales, a single sale changes CPV a lot: don't draw conclusions from a single month.

Limits

This guide does not give benchmark figures by industry: we don’t use them because they vary by market, product and ticket, and comparing yourself with them usually confuses more than it helps. It also does not calculate return on investment, which requires margin and customer lifetime value.

Decision criteria

How to use the downloadable template

1. Write your definitions of lead and sale in the first notes row.
2. Record cost, leads and sales by month and channel.
3. Calculate CPL, CPV and closing rate manually or in your sheet.
4. Compare with your previous months.

Criteria for reading the numbers

Complete synthetic example

Synthetic example. A fictional company invests COP 3.000.000 per month in search. In September it obtains 60 leads and 4 sales: CPL of COP 50.000 and CPV of COP 750.000. In October it expands keywords and obtains 110 leads and 4 sales: CPL falls to COP 27.273, but CPV stays the same and the sales team spends almost twice as much time filtering contacts.

If the company only looked at CPL, it would conclude October was better. By looking at CPV and closing rate (6.7% versus 3.6%), it sees it paid the same per sale with more saleswork. It decides to review the new search terms, exclude those that don't generate opportunities and measure by cohort, because its sales take between three and six weeks to close. All figures are invented.

How to present the results

When you share these numbers with management, always show the definitions, the period and the number of sales behind each figure. A CPV calculated with two sales is not suitable to decide budget; state that explicitly. Compare against your own previous months, not against averages from other companies whose lead definition you don’t know.

How to know if it worked

• Lead and sale definitions are written and don’t change without notice.
• Each figure comes with its period and number of sales.
• Budget decisions are based on CPV and not only CPL.

Plan for the first four weeks

1. Week 1: write definitions and method.
2. Week 2: reconstruct the last three months.
3. Week 3: calculate CPL, CPV and closing rate by channel.
4. Week 4: present results with their limits and decide.

Diagnosis before starting

• Find how much you spent on ads in the last three months.
• Count how many sales you can attribute to those ads in your records.
• Write what you currently count as a lead.

A second example, in contrast

Synthetic example. A fictional B2B software company gets only 15 leads per month with a high CPL, and management wants to cut the campaign. Looking at CPV by three-month cohort, it sees those leads close annual contracts and the cost per sale is reasonable for its margin. Unlike the first example, the high CPL was not a problem. Example invented.

What to do when something fails

• If the figures change a lot, check whether any definition changed.
• If sales are missing, check that the record retains the source.
• If you can’t calculate CPV, present only what you do measure and state that.

Common mistakes

• Changing the definition of lead from one month to another.
• Mixing costs with and without fees.
• Judging a campaign only by its CPL.
• Comparing with benchmarks from other countries or sectors.
• Drawing conclusions with very few sales.

Checklist

• Written definitions.
• Included costs kept constant.
• Chosen method.
• Comparison against yourself.
• Don't decide based on a single month with few sales.