Blog · Finance metrics and formulas · Sales teams
Cost per qualified lead depends on exactly which qualification stage you count. Calculate campaign spending per MQL and SQL, reconcile stage histories, and compare qualification rates with opportunity outcomes.
Cost per qualified lead is campaign spending divided by the unique leads that reach a named qualification stage. State whether that stage is MQL or SQL: the same $18,000 campaign can cost $200 per marketing-qualified lead and $600 per sales-qualified lead.
Cost per MQL = Eligible campaign spending / Unique marketing-qualified leads
Cost per SQL = Eligible campaign spending / Unique sales-qualified leads
MQL means marketing has judged the lead ready for sales; SQL means sales has qualified the lead as a potential customer. Those distinctions follow HubSpot's published lifecycle definitions. Your organization must still write its own acceptance criteria: company fit, role, intent, need, and any required sales review.
Record who owns each definition. Marketing operations can maintain the MQL rule, while sales leadership approves the SQL rule. Finance should agree the spending basis. A qualification threshold changed halfway through a comparison must be disclosed or applied consistently to the underlying records.
Use a contact denominator here. If your business evaluates buying organizations instead, label the measure cost per qualified account and use the B2B cost per lead workflow for account deduplication. Mixing contacts and accounts changes the question.
The spending file needs campaign ID, service date, currency, cost category, and eligible amount. The CRM export needs lead ID, acquisition date, assigned campaign, and first dates reaching MQL, SQL, and opportunity. Keep rejection reasons and unmatched campaign IDs available for review.
Choose an acquisition cohort and an observation cutoff. This example follows leads created during May 2026 and records first stage entries through August 31, 2026. Both campaigns have the same follow-up window. Spending includes their agreed direct campaign costs in USD.
Use first-entry history rather than current stage. A lead that has become an opportunity has already passed qualification in this example, even if its current stage no longer says SQL. For the arithmetic below, the company's process requires every SQL to have passed MQL, and every opportunity to have passed SQL. If your actual process permits skipped stages, count the history honestly and explain the different path.
A US business services company compares Search and Webinar. Each valid lead belongs to one campaign, and every stage count is unique within that campaign cohort. Each lead has at most one opportunity in this example, so opportunity counts equal unique SQL leads reaching opportunity. For other exports, deduplicate SQL lead IDs before calculating conversion, even if a lead is linked to several deals.
| Campaign | Spend (USD) | Valid leads | MQLs | SQLs | Opportunities |
|---|---|---|---|---|---|
| Search | 18,000 | 360 | 90 | 30 | 12 |
| Webinar | 12,000 | 240 | 120 | 24 | 6 |
| Total | 30,000 | 600 | 210 | 54 | 18 |
Divide the same spending by each relevant stage count. Dollar amounts below are rounded to cents only after division.
| Campaign | CPL calculation | CPL (USD) | Cost per MQL calculation | Cost per MQL (USD) | Cost per SQL calculation | Cost per SQL (USD) |
|---|---|---|---|---|---|---|
| Search | 18,000 / 360 | 50.00 | 18,000 / 90 | 200.00 | 18,000 / 30 | 600.00 |
| Webinar | 12,000 / 240 | 50.00 | 12,000 / 120 | 100.00 | 12,000 / 24 | 500.00 |
| Total | 30,000 / 600 | 50.00 | 30,000 / 210 | 142.86 | 30,000 / 54 | 555.56 |
The campaigns have identical initial CPL. Webinar costs less per MQL and per SQL, but the later-stage conversion deserves a separate check.
SQL-to-opportunity rate = Unique SQL leads reaching opportunity / Unique SQL leads
| Campaign | Lead to MQL | MQL to SQL | SQL to opportunity |
|---|---|---|---|
| Search | 90 / 360 = 25.0% | 30 / 90 = 33.3% | 12 / 30 = 40.0% |
| Webinar | 120 / 240 = 50.0% | 24 / 120 = 20.0% | 6 / 24 = 25.0% |
| Total | 210 / 600 = 35.0% | 54 / 210 = 25.7% | 18 / 54 = 33.3% |
Percentages are displayed to one decimal place. The generic cost per lead calculation cannot reveal these stage differences on its own.
Prepare a deduplicated Leads sheet with campaign ID in A, acquisition date in B, first MQL date in C, and first SQL date in D. In a summary sheet, A2 contains the campaign ID, H1 the cohort start, I1 the next-period start, and J1 the observation cutoff.
In Excel, the MQL count is:
=COUNTIFS(Leads!$A$2:$A$10000,A2,Leads!$B$2:$B$10000,">="&$H$1,Leads!$B$2:$B$10000,"<"&$I$1,Leads!$C$2:$C$10000,">0",Leads!$C$2:$C$10000,"<"&($J$1+1))
For SQLs, replace both references to column C with column D. The positive-date test excludes blanks, and the upper boundary includes timestamps throughout the cutoff date. Use actual spreadsheet dates, not date-looking text, and extend the ranges to cover the whole file.
Microsoft documents COUNTIFS as counting records meeting all specified criteria. It counts qualifying rows, so a stage-history export with repeated rows per lead must be reduced to first-entry dates before this calculation.
If summary B2 contains spending, C2 contains MQLs, and D2 contains SQLs, use:
=IF(C2=0,"N/A",B2/C2)
=IF(D2=0,"N/A",B2/D2)
These return cost per MQL and cost per SQL respectively. Covirage is not affiliated with Microsoft or HubSpot.
The same cohort supports an independent reconstruction:
Cost per SQL = CPL / Lead-to-MQL rate / MQL-to-SQL rate
For the total, retain the exact fractions: $50 divided by (210 / 600), then divided by (54 / 210), equals $555.56 after rounding. The lead and MQL counts cancel, leaving $30,000 divided by 54 SQLs. Using displayed rounded percentages instead would introduce a rounding difference.
Also tie stage totals to the distinct lead IDs: 90 plus 120 gives 210 MQLs, and 30 plus 24 gives 54 SQLs. In this example, SQL IDs must be a subset of MQL IDs, and opportunity IDs must link to SQL leads. Investigate stage dates that precede acquisition or appear in the wrong order.
Webinar's lower cost per SQL is useful, but Search has stronger SQL-to-opportunity conversion in the observed data. Review opportunity value, eventual wins, and capacity before scaling either campaign. This is a qualification report, not proof of superior customer profitability.
Keep broader selling costs in customer acquisition cost, and use cost per lead by channel when the spending allocation itself needs review. A changed qualification rule or a newly mature cohort can alter the apparent ranking without any change in media efficiency.
Bring your campaign spending and a sample of your own CRM rows with qualification dates. Covirage maps the columns, its tools compute and check the figures, and the external AI model explains the results with every figure traced to its rows. Create a free Covirage account, then set up your business and bring a sample of your own data.
See Covirage for sales teams for the supporting context.
Both divide the same agreed spending by a stage count. Cost per MQL uses leads accepted under marketing's qualification rules. Cost per SQL uses leads accepted under sales' rules. Because those populations differ, the figures answer different questions about campaign efficiency.
For an acquisition-cohort report, count each unique lead once at its first qualifying entry into the named stage. Recycled leads and repeated stage changes need a separate rule. Counting every status update would make activity look like additional lead generation.
Yes. A campaign may create many marketing-qualified leads that sales later rejects or that rarely become opportunities. Compare cost per SQL and opportunity conversion before treating a lower cost per MQL as evidence that the campaign deserves more budget.
Keep the campaign spending visible and report the stage count as zero. Show the cost per qualified lead as undefined rather than zero. Confirm that qualification history and the observation window are complete before judging the campaign.