A normal B2B website turns 1% to 3% of its unique visitors into a lead. In First Page Sage's client data for January 2022 to August 2025, 22 of 24 industries sit between 1.1% (B2B SaaS and software development) and 2.9% (staffing and recruiting). The two exceptions are HVAC services at 3.1% and legal services at 7.4%.
This guide covers what is normal, what counts as good, and a seven-step method for finding where a site loses leads, from the visitor count to the sales reply. SaaS trial and pricing-page funnels are in our B2B SaaS Website Conversion Rate Statistics 2026 post, so this one stays with B2B sites in general.
Normal depends on your industry and on what you divide by
Every conversion rate is a numerator over a denominator, and the published benchmarks pick different ones. The table sets two large datasets side by side for the industries B2B buyers sit in.
| Industry | First Page Sage: any lead per unique visitor | Ruler Analytics 2026: qualified lead or sale, calls included |
|---|---|---|
| B2B SaaS | 1.1% | 7.6% (software) |
| Engineering | 1.2% | 4.9% (construction and engineering) |
| Transportation and logistics | 1.4% | Not reported |
| IT and managed services | 1.5% | Not reported |
| Financial services | 1.9% | 6.3% (finance) |
| Construction | 1.9% | 4.9% (construction and engineering) |
| Manufacturing | 2.2% | Not reported |
| Real estate | 2.7% | 2.8% |
| Staffing and recruiting | 2.9% | Not reported |
| Professional services | Not reported | 6.1% |
| Legal services | 7.4% | 7.9% |
| All industries | No average published | 5.13% |
Source: First Page Sage, B2B Conversion Rates by Industry, client data January 2022 to August 2025; Ruler Analytics, Conversion Rate Benchmarks 2026, 110M+ sessions across 13 industries
The columns disagree because they count different things. First Page Sage, an SEO agency reporting its own clients, divides any conversion action by unique visitors. Ruler Analytics sells call tracking and attribution software, and its figures count qualified leads and sales, online and offline, credited across every touch. A software company at 1.2% on the first definition and 7% on the second can be the same company.
Three more definitions circulate. Databox's B2B benchmark group had a median session conversion rate of 2.3% in November 2023, counting any goal per GA4 session. HockeyStack's December 2025 study of 80 B2B SaaS companies found 1.1% when only demo, pricing and contact submissions count, one for every 91 unique visitors. Unbounce's 6.6% all-industry median is for single landing pages, not whole sites, from a 2024 dataset of 41,000+ pages.
What counts as good when nobody publishes a top quartile
We looked for percentile data and found little. Unbounce publishes medians only, as does Databox. The one quartile split we found for B2B is from Brixon, a German B2B marketing agency, which surveyed 478 industrial companies in Germany, Austria and Switzerland in 2025. It reports an average website-to-lead rate of 1.9%, close to First Page Sage's manufacturing figure, and a top quartile that turns 7.3% of leads into customers against a 2.7% average. It publishes little method, so treat it as indicative. Its useful point is where the top quartile pulls away: after the form, not on the page.
Without a published top quartile, we use a working definition with three tests.
Above your industry median, on its definition. If you count any form per unique visitor, compare with First Page Sage's column. If you count qualified leads with calls, compare with Ruler's.
Above benchmark channel by channel. The overall rate is a weighted average of traffic sources.
Holding up after the form. A rate that rises while sales-qualified leads fall usually means the form got easier and the leads got worse.
| Channel | First Page Sage, B2B leads per visitor | Ruler Analytics 2026, all industries, qualified lead or sale |
|---|---|---|
| Account-based marketing | 3.8% | Not reported |
| Organic search | 2.6% (thought leadership SEO) | 4.9% |
| 2.4% | 4.9% | |
| Webinars | 2.3% | Not reported |
| Organic social | 1.7% | 2.23% |
| Paid search | 1.5% (PPC and SEM) | 5.4% |
| Paid social | 0.9% | 2.11% |
| AI referral | Not reported | 5.8% |
| Direct | Not reported | 4.7% |
Source: First Page Sage, Conversion Rate by Channel, 2025; Ruler Analytics, Conversion Rate Benchmarks 2026
Run two traffic mixes through First Page Sage's B2B rates and the expected rate changes with no change to a single page. A site with 60% organic search, 20% email and 20% paid search should convert about 2.3%. A site with 50% paid social, 30% paid search and 20% organic search should convert about 1.4%. That is our arithmetic, not a measurement, but it means the second site at 1.6% is doing better than the first at 2.0%.
Normal after the form: the stages that decide revenue
A visitor-to-lead rate says nothing about revenue on its own. First Page Sage also publishes stage rates from lead to closed deal, from internal and client data gathered between 2017 and 2025, with a client base it describes as about 65% B2B.
| Industry | Lead to MQL | MQL to SQL | SQL to opportunity | Opportunity to close | Customers per 10,000 visitors |
|---|---|---|---|---|---|
| IT and managed services | 19% | 38% | 41% | 46% | 2.0 |
| B2B SaaS | 39% | 38% | 42% | 37% | 2.5 |
| Engineering | 27% | 36% | 48% | 52% | 2.9 |
| Industrial IoT | 22% | 39% | 46% | 51% | 5.2 |
| Transportation and logistics | 31% | 44% | 49% | 56% | 5.2 |
| Financial services | 29% | 38% | 49% | 53% | 5.4 |
| Staffing and recruiting | 25% | 32% | 45% | 52% | 5.4 |
| Manufacturing | 26% | 41% | 46% | 51% | 5.5 |
| Legal services | 32% | 35% | 48% | 46% | 18.3 |
Source: First Page Sage, Sales Funnel Conversion Rate Benchmarks 2026 Report and B2B Conversion Rates by Industry; last column is our arithmetic
The last column multiplies the visitor-to-lead rate from the first table by the four stage rates. It mixes two datasets with different years, so read it as an order of magnitude: a typical B2B site wins two to six customers per 10,000 visitors, and legal services about 18. The first stage varies most. Lead to MQL runs from 19% for IT and managed services to 39% for B2B SaaS, a two-to-one spread, while the later stages sit mostly between 35% and 55%. When a site's lead-to-MQL rate is far below its industry's, the leak is usually the traffic or the offer, not the page.
How to find your leak: seven checks in order
Each check has a reference number from a source you can open.
| Check | What to measure | Reference number | Source |
|---|---|---|---|
| 1. Denominator | Share of sessions from bots, staff and test traffic | 51% of all web traffic was automated in 2024 | Imperva, 2025 |
| 2. Channel mix | Leads per visitor, by source | B2B 0.9% (paid social) to 3.8% (ABM) | First Page Sage, 2025 |
| 3. Path | Share of visitors reaching pricing and demo pages | 16.5% reach pricing, 0.91% reach demo, in B2B SaaS | HockeyStack, 2025 |
| 4. Speed | Largest Contentful Paint at the 75th percentile | 2.5 seconds or less is good | web.dev, Google |
| 5. Form | Starters who complete; viewers who complete | 51.7% of starters complete; contact forms 14.5% of viewers | Zuko, 2022 to 2025 |
| 6. Reply | Leads answered, and how fast | 32% of forms got an email reply; 5% of replies inside five minutes | Clay, 2026 |
| 7. Qualification | Lead to MQL against your industry | 17% to 45% across First Page Sage's industries | First Page Sage, 2026 |
Source: the publishers named in the last column
1. The denominator. Before anything else, check who the visitors are. Imperva's 2025 Bad Bot Report found automated traffic overtook human traffic in 2024 at 51%, with malicious bots at 37%. Exclude internal traffic, staging domains and form tests, and check what your conversion event counts. GA4 now calls these events key events and reserves "conversion" for the ones used in Google Ads bidding, so a newsletter sign-up marked as a key event can inflate a rate as easily as bots deflate it.
2. The channel mix. Break the rate down by source and compare each line with the channel table. A site that looks weak overall is often fine in search and email and dragged down by paid social, which First Page Sage puts at 0.9% for B2B.
3. The path to the intent pages. HockeyStack's data shows where B2B SaaS visitors go: pricing pages take 16.5% of traffic, bounce 39% and convert 3.8%; demo pages take 0.91%, bounce 70% and convert 5.5%. If few visitors reach your pricing, demo or contact page, the fix is navigation and internal links, not the form. A GA4 funnel exploration shows it with up to 10 steps, open or closed, and can split each step by device.
4. Speed. Google's threshold for good is a Largest Contentful Paint of 2.5 seconds or less at the 75th percentile, measured separately for mobile and desktop. Portent's April 2022 study, 14 B2B lead generation sites over 30 days, found a page loading in one second converted three times better than one loading in five. Portent is an agency, the sample is small and the data is four years old, so use the direction, not the multiple.
5. The form. Zuko's benchmarking from January 2022 to April 2025 found 51.7% of people who start a form finish it, and contact forms completed by 14.5% of the people who view them, the lowest of any form type. Field-level data shows which question loses people; our Demo Request Conversion Rates 2026 post has the field-count studies.
6. The reply. Clay submitted 6,346 contact and demo forms across 100 countries in 2026. Two-thirds of companies never replied, 64% of the replies that came were automated, and average speed to lead was 15 to 24 hours. A lead that is never answered looks like a marketing problem in the pipeline report.
7. Qualification. Compare your lead-to-MQL rate with your industry's in the funnel table. Far below it, the site is attracting the wrong visitors or offering the wrong next step; far above it, the form may be screening out buyers who were not ready to talk to sales.
Reading the result: which leak to fix first
Fix the stage furthest below its reference number, weighted by how many people pass through it. A site with 10,000 monthly visitors at 1.5% produces about 150 leads; if half are never answered, fixing the reply is worth more than any page test, and it costs nothing in design.
Checks 1, 6 and 7 are usually free: filters, routing rules and a written definition of a qualified lead. Checks 2 and 3 are content and navigation work. Checks 4 and 5 are build work. Run them in that order, measure a full month after each change, and keep the definition fixed. The Analytics Stack for a Four-Week Site Build post on this site covers the tracking that makes the seven checks possible.
What this means for your B2B website
Set your target from your own industry's median on your own definition, not from a headline figure. If your site sits between 1% and 3% of unique visitors, it is normal; the gains are in the channel mix, the path to the intent pages and what happens after the form. If it sits below 1%, start with the denominator, because a rate that low is as often a measurement problem as a website problem.
We build marketing sites in Webflow and applications in Next.js, so read our view with that in mind, though none of the seven checks depends on the platform. In our four-week marketing-site process we would agree the numerator, the denominator and the definition of a qualified lead in the first week, before design starts. Our marketing website service page explains how we scope that work.
