If you can't name your conversion rate at every stage of your funnel, you're running your business blind. These are the seven numbers every founder should know cold — plus what's "good" for each.
Why these seven
Most founders track one or two conversion rates and call it good. Usually it's "trial-to-paid" or "website-to-lead." Those are useful but they hide more than they reveal.
A funnel is a series of conversions. If you only track the end, you can't tell where the leak is. If you track every stage, the leak becomes obvious.
Here are the seven rates that matter most, what each one tells you, the benchmark to measure against, and how to fix it when it's broken.
1. Visitor → Lead Conversion Rate
What it is: The percentage of website visitors who take an identifying action — fill out a form, download a resource, start a trial, request a demo.
Why it matters: This is the first leak in the funnel. If you have great traffic but no leads, your site isn't converting attention into intent.
Benchmarks:
- B2B average: 2.3%
- B2B SaaS average: 1.5-2.5%
- Top 10% B2B SaaS: 8-15%
- B2C e-commerce average: 2.79%
- High-performing B2C: 5%+
By channel:
- Referral traffic: 2.9% (highest)
- Organic search: 2.6-2.7%
- Email: 2.4%
- Paid search: 1.5-3.2%
- Social media: under 1%
If you're below benchmark, the problem is usually:
- Unclear value proposition on the homepage
- No clear primary CTA
- Forms that are too long
- Pricing not visible (B2B buyers want to qualify themselves before talking to sales)
- Wrong audience hitting the site (bad ad targeting, irrelevant SEO traffic)
How to fix it:
- Simplify the hero section to one clear value prop
- Add one primary CTA (not five)
- Cut form fields to the minimum (5 fields max)
- Add social proof above the fold (customer logos, testimonials)
- Run heatmap and session-replay tools (Hotjar, Microsoft Clarity — free) to see where visitors drop off
2. Lead → MQL Conversion Rate
What it is: The percentage of leads who meet your Marketing Qualified Lead criteria (structural fit + behavioral engagement).
Why it matters: This rate tells you whether your top-of-funnel content is attracting the right audience or just any audience.
Benchmarks:
- B2B average: 31%
- B2B SaaS: 39%
- E-commerce: 23%
- Financial services: 29%
- Higher education: 45%
- SEO leads specifically: 41%
- PPC leads specifically: 26%
If you're below benchmark, the problem is usually:
- Content that attracts the wrong audience (e.g., "how to" content that brings students and freelancers, not buyers)
- Lead magnets that are too broad
- Forms that don't capture qualifying info
- Channel mix weighted toward low-quality sources (cheap PPC, broad social)
How to fix it:
- Audit which content sources produce MQLs vs. just leads
- Tighten your ICP-relevant content (case studies, comparison guides, ROI calculators)
- Add 1-2 qualifying questions to high-value form fills (company size, role)
- Shift budget toward channels with proven MQL conversion (SEO, referrals, partnerships)
3. MQL → SQL Conversion Rate
What it is: The percentage of MQLs that sales accepts as qualified.
Why it matters: This is the most important conversion rate in your entire funnel. It tells you whether marketing and sales are aligned on what "qualified" means.
Benchmarks:
- Cross-industry B2B average: 13% (the number everyone quotes, often misleadingly)
- B2B SaaS average: 32-40%
- Top performers (behavioral scoring): 39-40%
- SEO-sourced MQLs: 51%
- Webinar-sourced MQLs: 30%
If you're below benchmark, the problem is usually:
- MQL definition is too loose (you're sending "leads" not "qualified leads")
- No behavioral component to scoring (only firmographic)
- Sales doesn't trust marketing and rejects on principle
- Speed-to-lead is slow (responding to leads after 1 hour cuts conversion rates dramatically)
- No feedback loop between sales rejections and marketing scoring
How to fix it:
- Co-write the MQL definition with sales. Get sign-off.
- Add behavioral signals to your scoring model (page views, return visits, pricing page visits)
- Set a hard SLA: sales responds to every MQL within 60 minutes during business hours
- Create a structured rejection process so marketing learns from every "not qualified"
- Stop counting one-time content downloads as MQLs
Research finding: Companies that respond to MQLs within 5 minutes are 21x more likely to qualify the lead than those who wait 30 minutes. After 24 hours, conversion drops to 17% from 53%.
4. SQL → Opportunity Conversion Rate
What it is: The percentage of SQLs that progress into the active sales pipeline as opportunities (with defined value, expected close date, stage).
Why it matters: This rate tells you whether your sales team is doing discovery effectively. SQLs should be sales-validated leads — if you're losing them at this stage, your reps are missing something during qualification.
Benchmarks:
- B2B average: 30-59%
- B2B SaaS: ~42%
- E-commerce: ~66%
- Higher education: ~61%
If you're below benchmark, the problem is usually:
- Sales reps qualifying leads in (accepting too liberally) but failing to qualify them through
- Bad discovery — not surfacing real pain, budget, authority, or timeline
- Wrong product fit (the lead seems qualified but your product doesn't actually solve their problem)
- Pricing surprise (lead is qualified but balks when they see real price)
How to fix it:
- Standardize discovery questions using BANT, MEDDIC, or GPCT
- Add a deal-qualification checklist that must be completed before moving an SQL to opportunity
- Be transparent about pricing earlier in the conversation
- Train reps on "qualifying out" — not every SQL should become an opportunity
5. Opportunity → Customer Conversion Rate (Win Rate)
What it is: The percentage of opportunities that close as customers.
Why it matters: This is your sales effectiveness number. It tells you how well your team converts pipeline into revenue.
Benchmarks:
- B2B average: 22-30%
- B2B SaaS: 20-37%
- Top performers: 30%+
- E-commerce: ~60%
- Higher education: ~66%
- PPC-sourced opportunities: ~35%
If you're below benchmark, the problem is usually:
- Weak deal qualification (you're working too many opportunities that shouldn't be in pipeline)
- Long sales cycles with no forcing function (decision-makers go silent, deals stall)
- Single-threaded deals (only one champion — they leave or get reassigned, deal dies)
- Competitor losses (not differentiating effectively)
- Pricing pushback (you're priced wrong for the segment)
How to fix it:
- Implement deal stage exit criteria (specific things that must be true to advance)
- Multi-thread every deal (build relationships beyond the primary champion)
- Track close-lost reasons and patterns (10 losses to the same competitor = product problem, not sales problem)
- Add forcing functions (trial expirations, contract end dates, year-end budget cycles)
- Review every deal weekly — kill the ones that have stalled
6. Customer → Expansion Revenue Rate (Net Revenue Retention / NRR)
What it is: The percentage of revenue retained AND grown from existing customers year-over-year. Calculated as: (Starting ARR + Upgrades + Cross-sells - Downgrades - Churn) ÷ Starting ARR.
Why it matters: This is the single most predictive metric of long-term SaaS success. Companies with high NRR have a recurring revenue engine. Companies with low NRR are running uphill — they have to replace lost customers AND add new ones to grow.
Benchmarks:
- B2B SaaS average: 100-110%
- Healthy B2B SaaS: 110-120%
- Best-in-class: 120%+
- Top public companies (Snowflake, Datadog at peak): 130%+
Below 100% = you're losing more revenue from existing customers than you're growing. You need 100%+ just to stand still.
If you're below benchmark, the problem is usually:
- Customers not getting to value fast enough (poor onboarding)
- Customer success is reactive, not proactive
- No clear expansion path (no upgrade tiers, no add-ons)
- Renewal processes are last-minute fire drills
- You don't know which customers are about to churn
How to fix it:
- Build a structured onboarding plan with clear value milestones
- Track product usage as a leading indicator of churn risk
- Identify expansion signals (hitting plan limits, adding users, asking about features)
- Start renewal conversations 90 days before contract end, not 30
- Survey customers regularly (NPS, CSAT, product feedback)
7. Visitor → Customer (End-to-End Funnel Rate)
What it is: The percentage of total website visitors who eventually become paying customers.
Why it matters: This is your overall funnel efficiency. It compresses every stage into one number — useful for high-level health checks, but only useful in context of the other six.
Benchmarks:
- B2B end-to-end: 3-7%
- B2B SaaS: typically lower (longer cycles)
- E-commerce: 2-3% (but the funnel is shorter)
Why this number alone is dangerous: A 3% visitor-to-customer rate could mean:
- Great traffic, weak funnel (the leak is somewhere in the middle)
- Bad traffic, great funnel (the visitors are wrong audience but the people who do convert convert well)
- Average across the board (no obvious leak)
You can't tell without breaking it apart. Always look at the seven rates individually, not just the end-to-end.
The benchmark exercise
Take 30 minutes this week and fill in this table for your business:
| Stage | Your rate | Benchmark | Gap | Priority |
|---|---|---|---|---|
| Visitor → Lead | ___% | 2.3% (B2B avg) | ||
| Lead → MQL | ___% | 31% | ||
| MQL → SQL | ___% | 32-40% (SaaS) | ||
| SQL → Opportunity | ___% | 42% (SaaS) | ||
| Opportunity → Customer | ___% | 20-37% (SaaS) | ||
| Net Revenue Retention | ___% | 110-120% | ||
| Visitor → Customer | ___% | 3-7% (B2B) |
For each row where you're significantly below benchmark, that's where your effort goes.
If you don't have the data to fill in your own rates, that's the first problem to solve — you can't manage what you don't measure.
The takeaway
You don't need to be best-in-class at every conversion rate. You need to be honest about where you are, compare it to the benchmark, and fix the biggest gap first.
The order of operations:
- Measure all seven rates in your business
- Compare each to benchmark
- Find the biggest gap
- Fix that stage first (don't optimize visitors if your MQL-to-SQL is broken)
- Re-measure quarterly
Most founders are working on the wrong stage. They're trying to drive more traffic when their actual leak is at MQL-to-SQL. They're trying to close more deals when their actual leak is at SQL-to-Opportunity. They're trying to acquire more customers when their actual leak is retention.
The benchmarks tell you where the leak actually is.
How agentic CRMs change which numbers you can measure
Most CRMs require manual data entry, which means most conversion rates are unreliable. If reps don't update deal stages consistently, your SQL-to-Opportunity rate is wrong. If marketing doesn't enrich contacts properly, your Lead-to-MQL rate is wrong.
Agent-driven CRMs solve this by making every stage transition automatic. Lead enters → enrichment runs → score crosses threshold → MQL becomes SQL → sales sees the lead instantly → response logged → deal created automatically. Every number is captured. Every conversion rate is accurate.
When the data is reliable, the math becomes operational. When it's not, you're guessing.
If you want to see what reliable funnel data looks like — and how an agent-driven CRM measures all seven conversion rates automatically — start a free trial or book a 20-minute conversation.
Related reading
- The Complete Guide to the Modern Sales Funnel — the full pillar post
- MQL vs SQL: The Real Definitions Nobody Agrees On
- How to Reverse-Engineer $1M in ARR
- What Happens When Marketing and Sales Don't Talk
- Why Pipeline Management Is the Most Underrated Founder Skill
Sources: First Page Sage 2025 Funnel Benchmarks, Ruler Analytics 2025, SaaS Hero 2026, Data-Mania 2026, HubSpot 2024 speed-to-lead research.