MQL-to-SQL conversion rate is the percentage of marketing-qualified leads that become sales-qualified leads within a defined window. It measures the handoff between marketing's qualification and sales' acceptance of a real buying opportunity or next sales process.
The rate is meaningful only when both teams agree on stage definitions, response service levels, rejection reasons and cohort timing. It should not be improved by making either stage artificially broad or narrow.
Use a marketing-sales acceptance contract
This original SEO Companies Hub contract makes the handoff auditable before a rate is calculated.
| Contract field | Marketing commits | Sales commits | QA evidence |
|---|---|---|---|
| MQL entry | Valid identity, fit/behavior rule version and source retained | Receive only eligible records | Timestamped rule and denominator audit |
| Response | Route to the correct owner with required context | Attempt within the agreed service level | Assignment and attempt timestamps |
| SQL acceptance | Supply evidence required for qualification | Apply the published need, fit, role and next-step criteria | Acceptance fields and stage-entry timestamp |
| Rejection/recycle | Accept actionable feedback and correct campaigns | Use shared reason codes instead of free text alone | Reason distribution and recycle history |
| Cohort maturity | Freeze creation cohort and rule version | Allow the full response/qualification window | Mature, pending and censored counts |
| Downstream quality | Preserve original source and content context | Track opportunity progression and loss | SQL-to-opportunity/value by cohort |
Neither team may improve the percentage by silently redefining its own stage.
Define the MQL entry state
An MQL should be a valid, deduplicated lead or account that meets documented fit and behavior criteria. Record:
- required fit attributes;
- qualifying actions;
- disqualifiers;
- score or rule version;
- source and landing context;
- entry timestamp;
- owner;
- whether human review occurred.
Do not accept a dashboard label without the rule. If marketing changes the score, version the definition and communicate it before evaluating sales.
Define the SQL state
An SQL is not simply an MQL that sales opened. It should meet agreed sales qualification and enter a specific active process.
Possible criteria include:
- confirmed need or use case;
- contact or account meets eligibility;
- buying role or access to decision process;
- plausible budget or economic value;
- timing or trigger event;
- successful two-way contact;
- accepted next step such as discovery;
- absence of an active duplicate opportunity.
The exact criteria depend on the sales motion. Publish them and distinguish SQL from opportunity if the organization uses both stages.
Use a cohort-based formula
For MQLs created in a defined period:
MQL-to-SQL rate = cohort MQLs that become SQLs within the window ÷ eligible cohort MQLs × 100
Example structure:
- cohort: MQLs entering during January;
- window: 60 days;
- numerator: those January MQLs that reached SQL within 60 days;
- denominator: eligible January MQLs assigned to sales;
Do not divide SQLs created this month by MQLs created this month when older MQLs can progress. That period-flow calculation answers a staffing question, not a conversion probability.
Define denominator eligibility
Decide whether to exclude:
- existing open opportunities;
- current customers routed to account management;
- territories not served;
- duplicates merged before handoff;
- partner or reseller inquiries;
- leads withdrawn before contact;
- records never assigned due to system failure.
Report system failures separately rather than removing them silently. Marketing-to-sales operations should be accountable for routing loss.
Set response and attempt service levels
Conversion can depend on speed and persistence of follow-up. Define:
- time to first attempt;
- accepted channels;
- minimum attempt sequence;
- business-hour and time-zone rules;
- reassignment conditions;
- handling of unavailable or incorrect contact data;
- escalation when no owner accepts the record.
Measure compliance by cohort. A low conversion rate from leads untouched for a week is not solely a targeting problem.
Avoid one universal response rule for every product; high-intent requests and low-intent nurture leads may require different handling.
Create a shared rejection taxonomy
Sales should choose a specific reason when declining or recycling an MQL:
- wrong geography;
- wrong company or consumer segment;
- no relevant need;
- insufficient scale or budget;
- student, job seeker, vendor or competitor;
- duplicate or existing opportunity;
- bad contact information;
- no response after required attempts;
- timing too early;
- product limitation;
- spam or test;
- qualification rule error.
Allow notes for nuance but keep categories stable enough to trend. Review “other” usage and update the taxonomy when needed.
Audit stage transitions
Every transition should preserve:
- MQL timestamp;
- assignment timestamp;
- first-contact attempt;
- acceptance or rejection time;
- SQL timestamp;
- reason and owner;
- later reversal;
- definition version.
Prevent representatives from moving stages merely to satisfy activity targets. Use permissions, required fields and periodic record sampling.
Track recycled MQLs separately. Define whether a lead can re-enter the denominator and how repeated qualification is counted.
Preserve search and content source
Website analytics can identify landing pages and meaningful events, while CRM records qualification. Preserve lawful source fields through the handoff:
- organic or paid source;
- campaign;
- landing page;
- conversion page and action;
- first and last observable source where supported;
- relevant content or tool;
- self-reported source;
- market and product interest.
Google Analytics key events identify important actions in analytics, not SQL status. Google also says Search Console and Analytics metrics differ. Keep the chain explicit instead of treating a search click as pipeline.
Segment by acquisition intent
Compare MQL-to-SQL by:
- branded organic search;
- non-brand problem research;
- comparison and pricing pages;
- service or product pages;
- directories and tools;
- paid campaigns;
- partners and referrals;
- events and outbound.
Early educational content may assist SQL creation without being the final lead source. Report direct and assisted views under a stated model.
Do not combine self-reported and system-attributed source without labeling them.
Segment by sales context
Break down:
- product or service;
- market and language;
- company size;
- industry;
- new logo versus expansion;
- inbound versus account-based motion;
- sales team or territory;
- MQL score band;
- lead age at first contact.
This can expose a routing or capacity problem concentrated in one team. Protect privacy and avoid ranking individual representatives on tiny samples.
Diagnose a low rate
Possible causes include:
- MQL rule too permissive;
- search or campaign targeting mismatch;
- price or eligibility hidden before conversion;
- bad enrichment data;
- slow response;
- insufficient contact attempts;
- unclear SQL criteria;
- product-market mismatch;
- sales capacity constraint;
- duplicate or existing-account handling;
- stage hygiene problems.
Use rejection distribution, response timing and record samples before changing content or scoring.
Diagnose a high rate
High conversion can reflect strong alignment. It can also mean:
- MQL threshold is so strict that volume is unnecessarily low;
- SQL definition is weak;
- sales accepts records to meet targets;
- recycled leads are counted repeatedly;
- only high-intent demo requests are included;
- rejected records are removed from the denominator.
Check opportunity creation, progression and win quality.
Add downstream validation
Report:
- SQL-to-opportunity rate;
- opportunity win rate;
- average or median deal value;
- time to opportunity and close;
- gross profit;
- churn or retention;
- loss reasons;
- no-decision rate.
An MQL cohort with a lower SQL rate can still create more profitable customers. Do not optimize the handoff stage in isolation.
Establish a marketing-sales calibration meeting
Review a representative sample monthly:
- accepted SQLs that progressed;
- accepted SQLs that stalled;
- rejected MQLs;
- false-negative leads found later;
- source and content context;
- scoring and routing errors;
- capacity and response performance.
Agree on rule changes, owner and effective date. Do not rewrite historical stages without governance.
Benchmark responsibly
External rates are comparable only when these align:
- MQL and SQL definitions;
- observation window;
- lead versus account unit;
- product and sales cycle;
- inbound and outbound mix;
- qualification and response process;
- market and company size;
- source and period.
If a report omits these details, use it as a hypothesis, not a quota. Internal historical cohorts and capacity plans are stronger target inputs.
Build a decision report
Show:
| Cohort | Eligible MQLs | Mature share | SQLs | Rate | Median days | Top rejection | Opportunity rate |
|---|
Add response-SLA compliance, definition version, data quality and next action.
Avoid using a single overall rate to conceal territory or source differences.
A practical agreement
Marketing owns a documented MQL standard and source evidence. Sales owns timely, consistent evaluation and SQL criteria. Revenue operations owns routing, stage audit and shared reporting. Leadership resolves capacity and target conflicts.
Calculate the rate from mature MQL cohorts, preserve every transition and validate the stage against opportunity and customer quality. The goal is reliable handoff—not the highest possible percentage.
Related decisions
- Lead-to-MQL Conversion Rate: Definitions, Formula and QA Checks — the adjacent revenue operations decision.
- Website Traffic-to-Lead Conversion Rate: How to Calculate and Segment It — the adjacent seo conversion decision.
- AI Referral Conversion Rates: A Measurement Blueprint — the adjacent ai search decision.