Funnel Conversion Rates: Benchmarks and How to Calculate
Most B2B teams track one number: leads in, deals out. That single ratio hides the problem. A funnel that turns 2% of leads into customers could be bleeding money at the form, at the sales handoff, or in a proposal stage where nobody follows up. You cannot fix what you measure as one blurry average.
Conversion rate by stage shows you where the money leaks. This guide gives the exact math for each step, the benchmark ranges worth comparing against (with a warning that benchmarks mislead more often than they help), and how to read your own numbers without fooling yourself. If you came here asking "what is a good funnel conversion rate," the honest short answer sits a few paragraphs down, and the longer answer runs through the whole piece.
What a funnel conversion rate actually measures
A conversion rate is the share of people who move from one stage to the next. Visitor to lead. Lead to qualified lead. Qualified lead to opportunity. Opportunity to closed deal. Each of those transitions has its own rate.
The rates multiply. If 3% of visitors become leads, 30% of those leads qualify, and 25% of qualified leads close, your visitor-to-customer rate is 0.03 x 0.30 x 0.25, which works out to roughly 0.0023, or about two customers per thousand visitors. Small shifts at any single stage move that final number more than you would guess.
The common mistake is watching only the endpoints. A team obsessing over website conversion might pour budget into landing page tests while the real leak sits in a 48-hour lead response gap that quietly kills half the qualified pipeline. Stage-by-stage math is the only way to catch that.
Two words about vocabulary before the math. "Funnel conversion rate" gets used two ways. Sometimes people mean the single top-to-bottom number (visitor to customer). Sometimes they mean the individual step rates. Both matter. The blended number tells you overall efficiency; the step rates tell you where to work. Track both and label them clearly so a 3% and a 0.08% never get confused in a report.
How to calculate it, stage by stage
The formula is simple. Take the number that reached a stage, divide by the number that entered the prior stage, multiply by 100.
Conversion rate = (entered next stage / entered current stage) x 100
Two rules keep this honest.
First, fix your time window and your denominator. If you count leads from June against deals closed in June, you mix cohorts. A deal that closed in June may have started as a lead in March. For long B2B cycles, track a cohort: take the leads from a given month and follow them forward until the window closes. A simple period-over-period snapshot is faster and noisier.
Second, define every stage the same way each time. "Qualified lead" has to mean one specific thing, written down. If one rep marks a lead qualified after a form fill and another waits for a budget conversation, your rate measures two different things at once.
The three zones: TOFU, MOFU, BOFU
Marketers group funnel stages into three zones, and thinking in zones helps you diagnose faster.
Top of funnel (TOFU) covers awareness through the first hand-raise. This is visitor to lead: someone landed on your site or ad and gave you contact details. Rates here are the lowest in raw percentage terms, because most traffic is not ready to buy. This is where traffic quality and offer strength decide everything.
Middle of funnel (MOFU) covers lead through qualified lead through opportunity. This is where marketing and sales trade the baton: lead to MQL, MQL to SQL, SQL to opportunity. Middle-of-funnel benchmarks tend to look healthier than top-of-funnel numbers because the people here already showed intent. Leaks in this zone are the most expensive and the easiest to miss, since they hide behind reasonable-looking averages.
Bottom of funnel (BOFU) is opportunity to closed deal. Fewer people, bigger money per person. Here the sales process, pricing, and follow-up discipline set the rate.
Map your own stages before you measure. A typical B2B funnel has five or six measurable points:
- Visitors to the site or landing page.
- Leads: anyone who gives contact details (form, demo request, content download).
- Marketing qualified leads (MQLs): leads that fit your profile and showed buying signals.
- Sales qualified leads (SQLs): leads sales accepted and is actively working.
- Opportunities: an active deal with a number and a close date.
- Customers: closed and paid.
Not every business needs all six. A self-serve product might collapse MQL and SQL into one. A high-ticket consultancy might add a stage for proposal sent and proposal accepted. Map what reflects your real buying process instead of a template. Our walkthrough of the B2B sales funnel and where leads leak breaks down each stage and its common failure points if you need to build the map from scratch.
A worked example
Here is the math on real-looking numbers. They are illustrative, picked to make the calculation clear.
Say one month brings 10,000 visitors. From those you get 300 leads. Marketing qualifies 120 of them. Sales accepts 60 as SQLs. Those become 30 opportunities. Eight close.
- Visitor to lead: 300 / 10,000 = 3.0%
- Lead to MQL: 120 / 300 = 40.0%
- MQL to SQL: 60 / 120 = 50.0%
- SQL to opportunity: 30 / 60 = 50.0%
- Opportunity to customer: 8 / 30 = 26.7%
- Visitor to customer (blended): 8 / 10,000 = 0.08%
Read down that list and the leaks show themselves. Visitor-to-lead at 3% is normal. The MQL-to-SQL drop and the SQL-to-opportunity drop each lose half the pipeline. If you lifted opportunity-to-customer from 26.7% to 35%, you would close roughly 10 or 11 deals on the same traffic. That is the point of stage math: it points to the cheapest fix.
Watch the compounding effect. Improving one mid-funnel stage by ten points often does more for revenue than doubling top-of-funnel traffic, because fresh traffic gets diluted by every leak beneath it.
What a good rate looks like by stage
People want a number to compare against. Here are commonly cited B2B ranges, useful for orientation, not as targets. The table also shows the first move to make when a stage underperforms. Treat every figure as illustrative: your industry, average contract value (ACV), and traffic source shift these ranges a lot.
| Funnel stage | Typical conversion range (illustrative) | How to improve it |
|---|---|---|
| Visitor to lead (TOFU) | 1% to 5% | Sharpen the offer, match traffic intent to the page, cut form friction |
| Lead to MQL (MOFU) | 20% to 40% | Tighten scoring, fix low-quality lead sources, add intent signals |
| MQL to SQL (MOFU) | 30% to 50% | Align sales and marketing on one definition, speed up handoff |
| SQL to opportunity (MOFU/BOFU) | 40% to 60% | Respond faster, confirm fit and timing early, qualify budget |
| Opportunity to close (BOFU) | 15% to 30% | Rework pricing and proposals, handle objections, tighten follow-up |
| Visitor to customer (blended) | under 1% (often 0.1% to 0.8%) | Find the weakest single stage first, then work up |
Now the caveat that makes those numbers usable. Benchmarks blend wildly different businesses. A free-trial SaaS and a six-figure enterprise contract both call themselves B2B, and their funnels share almost nothing. Traffic source alone breaks the comparison: branded search converts several times better than cold display traffic, so a "low" blended rate might just mean you run a lot of top-of-funnel awareness ads.
So what is a good funnel conversion rate? The most useful answer is your own number from last quarter, moving up. External ranges tell you whether you sit in roughly the right universe. Your own trend tells you whether you are getting better. A team going from 22% to 28% opportunity-to-close beat a competitor stuck at a "great" 30% they have not improved in two years.
One more trap. A high conversion rate is not automatically good. Tighten lead criteria hard enough that only ready-to-buy prospects enter, and your rates soar while volume collapses. Conversion rate and lead volume trade off. The right balance depends on your unit economics, not on a benchmark table.
How B2B funnels differ
B2B conversion rates read differently from B2C, and comparing across the line will lead you astray.
Sales cycles are long. Ninety days is common, six months and beyond happens in enterprise. That length makes period snapshots dangerous and cohort tracking close to mandatory.
Buying happens by committee. A single "lead" often represents a champion who still has to convince a manager, a finance owner, and sometimes procurement. Your funnel measures one contact, but the real decision runs through a group you may never see. That gap between recorded lead and actual buying unit is why MOFU rates in B2B look softer than a marketer expecting B2C numbers would like.
Deal value is high and volume is low. When a stage sees 30 events a month instead of 30,000, one deal swings the percentage by points. Read trends, not single weeks.
Lead quality outranks lead quantity. A B2C funnel can win on sheer volume. A B2B funnel with the wrong 500 leads wastes a sales team's month. Smart B2B teams watch qualification rates as closely as raw conversion, since chasing a higher visitor-to-lead number backfires when the extra leads never qualify.
Where funnels leak and how to find it
Stage math points at the problem. It does not name the cause. Here is how to run the diagnosis once your numbers are in.
Start by comparing each step against its own range and against last quarter. The stage furthest below where it used to sit is your first suspect, not the stage with the lowest absolute percentage. A 26% opportunity-to-close might be fine; a lead-to-MQL that fell from 38% to 22% is the fire.
Segment before you conclude. A blended rate hides strong and weak parts. Break every stage by traffic source (branded search, paid, referral, cold outbound) and by campaign. A "low" visitor-to-lead rate often turns out to be one awareness campaign dragging down three healthy ones. Segmenting is usually where the real leak jumps out.
Watch for the classic leak points:
- The form. Long forms, too many fields, and mobile friction quietly kill top-of-funnel conversion. Small changes here move the biggest raw numbers, since every later stage feeds off this one. Our guide to high-converting lead capture forms covers the field-count and friction fixes worth testing first.
- The MQL-to-SQL gap. This is where most disputes and most lost revenue live. If marketing and sales disagree on what "qualified" means, leads fall through with nobody owning them. A sudden drop here is almost always a definition or alignment problem, not a lead-quality one.
- Response time. A lead that waits 48 hours for a reply is often a dead lead. Speed at the handoff moves SQL-to-opportunity more than most people credit.
- The proposal stage. Deals that reach a proposal and then go silent point to pricing, follow-up discipline, or a decision-maker who was never actually in the room.
If your worst offender sits in the middle of the funnel, the detailed funnel bottleneck playbook walks through isolating and clearing a single stuck stage.
How to improve each stage
Fixes should match the zone. Applying a BOFU tactic to a TOFU leak wastes effort.
Top of funnel (visitor to lead). This is a traffic-and-offer problem nine times out of ten. Match the page to the intent behind the click: someone searching a specific solution should not land on a generic homepage. Strengthen the offer so the value is obvious in five seconds. Cut form fields to the minimum that still lets you qualify. Test the headline against the ad or the search query it answers. The broader discipline here is conversion rate optimization, which gives you a testing method instead of one-off guesses.
Middle of funnel (lead to MQL to SQL to opportunity). Most middle-funnel gains come from process, not creative. Write one shared definition of "qualified" and get sales and marketing to sign it. Score leads on fit and intent so reps spend time on the right ones. Shrink the handoff delay: automate the alert, set a response-time SLA, and measure it. Nurture the leads that are a genuine fit but not ready, so they re-enter the funnel instead of dying in a list.
Bottom of funnel (opportunity to close). This is sales craft. Qualify budget and timing early so late-stage surprises drop. Rework proposals so pricing is clear and the next step is obvious. Name the main objection and answer it before it stalls the deal. Follow up on a schedule instead of waiting for the prospect. Small discipline gains here convert directly to revenue because the money per deal is already large.
The order matters. Fixing a leak low in the funnel makes every unit of traffic above it worth more, so mid and bottom-funnel fixes often pay back faster than buying more traffic. Chasing more visitors while the SQL-to-opportunity stage leaks half the pipeline just pours water into a bucket with a hole.
Metrics and tracking
Good funnel math needs clean inputs and consistent tracking. A few essentials.
Use one source of truth. Your CRM (HubSpot, Salesforce, Pipedrive) should hold the stage a lead sits in, updated the same way by everyone. When stages live in three tools that disagree, your rates are fiction.
Connect analytics to revenue. GA4 tracks the top of the funnel: sessions, sources, form submissions. Your CRM tracks the bottom: opportunities and closed deals. Tie the two together, through UTM parameters carried into the CRM or a closed-loop integration, so you can trace a closed deal back to the traffic source that started it. Without that link you can optimize a stage that never produces revenue.
Watch these alongside the raw rates:
- Rate by source. The single most revealing cut. Always segment.
- Time in stage. A stage that is slow, not just leaky, points to a process delay.
- Cost per stage. What it costs to fill each step. A leak is only worth fixing relative to what that stage costs you.
- Sample size per stage. Below roughly 30 to 50 events, do not over-read week-to-week movement.
Reading the numbers honestly matters as much as collecting them. Bot traffic and accidental form fills inflate the top and crush your visible rate, so clean the input first. And lead volume feels like progress even when those leads never qualify, so trace a stage's gain through to closed revenue before celebrating it.
Common mistakes
A short list of the errors that turn good math into bad decisions.
Measuring only the endpoints. The blended visitor-to-customer number tells you something is wrong, never where. Break it into stages.
Trusting benchmarks over your own trend. External ranges orient you. They cannot tell you whether you improved.
Mixing cohorts on a long cycle. For anything slower than a transactional funnel, follow leads forward as a group.
Optimizing a vanity stage. More leads that never qualify cost money and produce nothing. Judge every stage by the revenue behind it, and remember that rate and volume trade off: a "worse" rate on far more qualified volume can win.
Frequently asked questions
What is a good funnel conversion rate?
There is no single good number. For B2B, visitor-to-customer commonly lands well under 1%, while individual stages run from 1 to 5% at the top and 15 to 30% at the close. Those are illustrative ranges that shift with your industry, deal size, and traffic source. The most useful benchmark is your own trend: a rate improving quarter over quarter beats a "high" rate that has been flat for years.
How do I calculate conversion rate between two stages?
Divide the number that reached the later stage by the number that entered the earlier stage, then multiply by 100. If 120 leads became 60 SQLs, that is 60 divided by 120 times 100, which is 50%. Keep the time window and stage definitions consistent every time you run it.
What are typical middle-of-funnel benchmarks for B2B?
Middle-of-funnel steps often sit higher than top-of-funnel numbers because the people there already showed intent. As illustrative ranges: lead to MQL around 20 to 40%, MQL to SQL around 30 to 50%, SQL to opportunity around 40 to 60%. Alignment between sales and marketing and the speed of the handoff move these more than almost anything else.
Should I measure conversion by cohort or by period?
Cohort tracking is more accurate for B2B because sales cycles are long. Follow a group of leads from a given month forward until they close or die. Period snapshots are faster, but they mix leads and deals from different time frames, which distorts long-cycle funnels and can hide a real leak.
Why is my overall conversion rate so low?
Usually one of two causes: a genuine leak at a single stage, or diluted top-of-funnel traffic. Cold awareness traffic and broad keywords convert far worse than branded or referral traffic. Break the blended rate into stages and segment by source before you conclude anything. A low overall number often hides a few strong segments.
Can a conversion rate be too high?
Yes. An unusually high rate often means criteria so strict that volume suffers, or that you only count late-stage, ready-to-buy traffic. High conversion on tiny volume can produce fewer total deals than a lower rate across a wider top of funnel. Judge it against revenue and cost, never in isolation.
A short checklist
Before you trust a funnel number, run through this:
- Each stage has one written definition, used the same way by everyone.
- The time window and denominator are fixed and consistent.
- Long cycles are tracked by cohort, not by mixed-period snapshot.
- Every stage rate is segmented by traffic source before you draw conclusions.
- Each stage has enough events (roughly 30 plus) to read reliably.
- Top-of-funnel traffic is cleaned of bots and junk.
- Stage gains are traced through to closed revenue, not stopped at lead count.
- You compare against your own prior period first, benchmarks second.
Conversion rates are a flashlight, not a verdict. They show you where to look. If your funnel math keeps pointing at a leak you cannot close, or you are not sure your stage definitions hold up, get a second set of eyes on it. Send us your current stage numbers and we will run a 30-minute teardown to find the one fix with the best payback for your traffic. That conversation costs nothing and usually surfaces a leak the in-house team stopped seeing months ago.