Two-thirds of the 6,346 companies whose demo and contact forms Clay's agents filled in 2026 never replied, and only about 7% of the forms reached a salesperson. That audit is the latest in a line of mystery-shopper studies going back to Harvard Business Review's 2011 test of 2,241 companies, and every one finds the same thing: B2B SaaS does not lose its demo requests on the form. It loses them in the hours after it.
This page covers the layer around the form: response speed, form completion, the CTA mix on SaaS sites and the SDR handoff. The funnel percentiles, the 62% qualified-to-booked median and the no-show data are in Demo Request Conversion Rates 2026 on this site and are not repeated here.
How fast B2B companies answer a demo request, and what speed is worth
Every few years someone submits test leads to thousands of companies and times the replies. The shape has not changed since 2011.
| Study | Sample | Replied within 5 minutes | Never replied | Average time to reply |
|---|---|---|---|---|
| HBR, 2011 | 2,241 US companies | Not measured; 37% within an hour | 23% | 42 hours, responders only |
| Drift, 2017 | 433 B2B SaaS companies | 7% | 55% within five business days | Not published |
| Workato, 2021 | 114 B2B companies | 1 of 114 by email; none by phone | Nearly 1 in 5 by email | 11 h 54 min by email; 14 h 29 min by phone |
| RevenueHero, 2024 | 1,000 B2B websites | Not published | 63.5% | 1 day 5 h 17 min, responders only |
| Clay, 2026 | 6,346 forms, 100 countries | 5% of replies | About 68% | 15 to 24 hours; 26 to 40 hours for human replies |
Sources: HBR via SmartCompany; Drift; Workato; RevenueHero; Clay
In Clay's run, 32% of forms got an email reply, but 64% of those were automated and only 445, about 7% of all forms, came from a salesperson; 96% of companies never phoned. Workato isolates the one variable a website team controls: companies with a lead routing tool responded in 3 hours 32 minutes on average, companies without one in nearly 13 hours.
HBR put a price on the delay. Across 1.25 million leads received by 29 B2C and 13 B2B companies, firms that tried to contact a lead within an hour were nearly seven times as likely to qualify it, defined as a meaningful conversation with a decision maker, as firms that waited an hour longer, and more than sixty times as likely as firms that waited 24 hours. Chili Piper's February 2025 report, vendor data from nearly 4 million form submissions in 2024, puts form-to-booked-meeting at 30% without scheduling on the form against 66.7% of qualified submissions with instant scheduling.
Lead forms: how they complete and what to gate
Zuko tracks 93 million form sessions by purpose; the two types a demo page uses complete worst.
| Form purpose | Sessions tracked | Desktop | Mobile | All devices |
|---|---|---|---|---|
| Enquiry, lead generation | 4,797,087 | 47.7% | 37.7% | 42.7% |
| Contact | 321,937 | 51.2% | 39.4% | 47.4% |
| Registration | 26,582,277 | 63.0% | 57.2% | 60.7% |
| All forms | 93,022,997 | 54.5% | 47.5% | 51.7% |
Source: Zuko form type benchmarking, read 2026-10-18
A demo form that converts 43% of the people who see it is at the benchmark. Registration forms, the free-trial sign-up, complete 18 points better than enquiry forms, one reason self-serve CTAs lead so many homepages below. The field-count studies are tabled in the demo conversion post.
On gating, Navattic's 2026 report finds 66% of top-performing interactive demos have no form gate, and where gates exist, 75% ask for a single field. Chili Piper's figures give the reason to keep some qualification: 14.1% of its 4 million submissions were spam, personal emails or failed the customer's criteria.
Demo, free trial or interactive demo: the mix on SaaS sites
Navattic has counted CTAs on random samples of B2B SaaS sites since 2020.
| CTA type | 2020 | 2022 (5,495 sites) | 2026 (about 5,000 sites) |
|---|---|---|---|
| Demo, "Book a Demo" | 41% | 40% | 61% |
| Free trial, "Try for Free" | 12% | 8.65% | 22% |
| Product tour or interactive demo | Almost none | 5.61% | 18% |
Sources: Navattic, 2022 and 2026 reports
The counts used different definitions, so read the direction: the demo stays the dominant ask, the free trial has recovered, and the interactive demo has gone from a rounding error to nearly one site in five, up 50% on 2024 and 260% over four years.
Buyer preference explains the shift. Gartner's survey of 646 B2B buyers in August and September 2025 found 67% prefer a rep-free buying experience and 45% used AI tools during a recent purchase. An interactive demo sells to the 67% without asking them to book anything; the request form serves the rest.
Our read of 30 SaaS homepages, 18 October 2026
We read the navigation and first screen of 30 well-known B2B SaaS homepages on 18 October 2026: a small, biased sample, not a market share figure.
| Pattern | Sites | Count |
|---|---|---|
| Self-serve first, demo or sales secondary | Airtable, Amplitude, Apollo.io, Asana, Calendly, ClickUp, Intercom, Miro, monday.com, Notion, Pipedrive, Ramp, Rippling, Salesforce, Webflow, Workato, Zapier, Zendesk, ZoomInfo | 19 |
| Self-serve only | Datadog | 1 |
| Demo first, self-serve secondary | Databricks, HubSpot, Mixpanel | 3 |
| Demo or sales CTA only | Deel, Gong, Outreach, Salesloft, Vanta | 5 |
| Interactive tour first, demo in the nav | Lattice | 1 |
| Campaign hero, both CTAs in the nav | Snowflake | 1 |
Source: Octo Agency read of each live homepage, 18 October 2026; Clari redirected to Salesloft
A demo or talk-to-sales CTA appears in the nav or hero on 29 of the 30 sites, and a free trial, free plan or sign-up on 24. The hero leads with self-serve on 20 and with a demo on 8. The gap to Navattic's 61% demo and 22% free trial is the sample: companies with a free plan put it first; the long tail, which sells through a sales team, does not. The useful finding is the overlap: 23 of 30 offer both paths, so the question is rarely which CTA, but which one gets the filled button.
The handoff: what happens after the form
A demo request becomes revenue through an SDR or BDR, and the two largest surveys of that role show the handoff weakening.
| Measure | Figure | Source |
|---|---|---|
| Median monthly held-meeting quota | 10, down 40% since 2018 | Bridge Group 2025, 351 companies |
| SDRs at quota | 60%, the lowest in the study's history | Bridge Group 2025 |
| Touches per contact | 17 in 2024, 21 in 2025, 34 in 2026 | 6sense 2026, 872 BDRs |
| Inbound share of BDR activity | Up from 15% to 30% | 6sense 2026 |
| Per 100 contacted prospects | 41 respond, 33 have a meaningful conversation, 23 attend a meeting | 6sense 2026 |
| BDRs completing the full handoff: qualify, notes, book, attend | 36% | 6sense 2026 |
Sources: The Bridge Group, 2025; 6sense 2026 via MarketOne and CX Today
SDRs are doing more, 34 touches per contact where 17 were enough two years ago, and getting less: held-meeting quotas down 40% since 2018 and the lowest quota attainment the Bridge Group has recorded. Inbound now takes 30% of BDR time, so a slow demo response competes with outbound sequences. Only 36% of BDRs do the four things a handoff needs, so a marketing team that measures requests and stops there cannot see where its meetings went: of 100 prospects contacted, 23 attend one.
What this means for a SaaS demo page and the team behind it
The numbers split the work into three parts, and only the first is a web project. The form should complete at or above Zuko's 43% for enquiry forms: fewer fields on mobile, one step rather than two, a scheduler on the confirmation. The CTA mix should match the buyer: a rep-free path for the 67% who want it and the demo request for the rest, both on the first screen as on 23 of the 30 sites we read. Then the part no page can fix: a human reply inside an hour, worth seven times the qualification odds in HBR's data. On the Webflow marketing sites we build, we ask for the response-time target before the first wireframe, because the page is judged on meetings held, not requests sent. Our landing page service covers the first part and asks about the other two.
