Our popup data, from 28,674 Popupsmart campaigns
From January 2024 to September 2026 we tracked the displays, clicks, closes and leads of 28,674 Popupsmart campaigns. Here are the totals, who is in the data, every finding, the pages where we use them, and the files to reuse them.
- page views
- 1.98 billion
- popup displays
- 499.6 million
- leads: popup submissions with an email address or phone number
- 1.84 million
- campaigns
- 28,674
- businesses that ran a popup with Popupsmart in the window
- 11,533
Popupsmart first-party data, 1 January 2024 to 23 September 2026.
What a typical popup does
Medians across campaigns with at least 1,000 displays and 14 active days. Every rate divides by popup displays, not by visitors.
- 0.83%median lead conversion rateof 3,034 popups that ask for an email or phone number. The top 25% reach 1.96%, the top 10% reach 3.87%.
- 2.25%median click rateof 10,768 campaigns. The top 10% reach 11.67%.
- 40.5%of displays end with a closeAll displays pooled, February 2025 to September 2026, since close tracking began.
- 61%of leads arrive within 30 secondsof the popup appearing. 71% of visitors who close a popup do so within 5 seconds.
Who is in the benchmark
The businesses behind the numbers: their industry, the platform their website runs on and their company size. Industry and company size come from LinkedIn company data matched on each website's domain; platform from the Shopify app install, Popupsmart's CMS detection or a technology crawl.
Industry
Industry medians rest on 27–206 lead-collecting campaigns each; the statistically reliable gaps are Finance & Insurance over Software & Technology and Health & Beauty over Healthcare & Medical.
| Industry | Campaigns | Median | |
|---|---|---|---|
| Education | 137 | 0.84% | |
| Fashion & Apparel | 97 | 0.88% | |
| Finance & Insurance | 38 | 2.22% | |
| Food & Beverage | 52 | 1.42% | |
| Health & Beauty | 97 | 1.99% | |
| Healthcare & Medical | 86 | 0.43% | |
| Manufacturing, Energy & Logistics | 107 | 0.70% | |
| Marketing & Agencies | 57 | 0.50% | |
| Media, Entertainment & Gaming | 93 | 1.14% | |
| Nonprofit & Public Sector | 27 | 1.03% | |
| Real Estate & Construction | 70 | 0.69% | |
| Retail & E-commerce | 183 | 0.79% | |
| Software & Technology | 206 | 0.39% | |
| Travel & Hospitality | 110 | 0.80% | |
Website platform
43% of the businesses in the benchmark (January 2024 – September 2026) run on Shopify; the WordPress ecosystem (WordPress and WooCommerce) accounts for 14%, custom-built sites on frameworks like Next.js for 2.5%.
| Platform | Share | |
|---|---|---|
| Shopify | 43.3% | |
| Custom / undetected | 25.3% | |
| WordPress (no WooCommerce) | 9.8% | |
| Other CMS / site builder | 4.5% | |
| WooCommerce | 3.9% | |
| Turkish e-commerce platform (ikas/Ticimax/IdeaSoft/T-Soft) | 3.6% | |
| Custom-built (Next.js / Nuxt / Gatsby) | 2.5% | |
| Blog platform (Blogger / Ghost / Substack) | 2.4% | |
| Webflow | 2.0% | |
| Other e-commerce platform | 1.7% | |
| Squarespace | 1.1% | |
| Wix | 0.8% | |
| Magento | 0.5% | |
| PrestaShop | 0.4% | |
| Other platforms (grouped) | 0.2% | |
Company size
Of the 3,309 businesses in the benchmark whose main website matches a LinkedIn company page (29% of 11,533), 73% list 50 employees or fewer.
| Company size | Share | |
|---|---|---|
| 1–10 employees | 41.8% | |
| 11–50 employees | 31.6% | |
| 51–200 employees | 15.0% | |
| 201–1,000 employees | 8.3% | |
| 1,001 or more employees | 3.3% | |
What the data shows, subject by subject
Every statement we publish, under its subject, with the breakdown behind it where there is one. A statement without a tag passed every check; a tag marks a lean, a correlation or no measurable difference. A measured "no difference" is a result too.
Discounts and offers
Part 1: Offer and format- Signup popups that offer a discount code convert 1.46% vs 0.75% for signup popups without one.
- Popups stating 10% off convert 1.25% vs 0.67% for popups that state no discount.
- No measurable differenceBigger discounts did not convert measurably better: 10% off converted 1.25%, 20% off 0.91% and 25–30% off 1.33%, but the larger-discount samples are small (78 and 46 campaigns), so the data cannot rule out a real difference either way.
- No measurable differenceSpin-to-win and other gamified popups get far more clicks (7.33% vs 1.87% of displays, median campaign; the spin itself counts as a click), but do not collect measurably more email or phone leads than plain signup popups.
Median lead conversion rate by the discount the popup states. Discount Campaigns Median 5% off 61 1.31% 10% off 515 1.25% 15% off 131 1.21% 20% off 78 0.91% 25–30% off 46 1.33% 35–50% off 46 1.01% Used in Popup best practices: we analyzed 1.98 billion page views and 28,674 Popupsmart campaigns · What is Gamification in Marketing? - Gamification Definition
Form fields
Part 2: Form design- Popups asking for one or two fields convert 0.88% and 0.94%, against 0.56% for three fields (median campaign; spin-to-win and other gamified popups are counted separately).
- LeanForms that ask for a phone number instead of an email look weakest (0.46% vs 0.88% for email only), though that gap is not statistically reliable.
- No measurable differenceAdding a name field alongside the email shows no reliable effect: the overall gap in its favour (1.25% vs 0.88% for email only) reverses among campaigns whose form rarely changed (0.74% vs 0.79%).
- No measurable differenceMulti-step forms show no measurable conversion advantage over single-step forms once step-1 clicks without contact details are excluded.
Median lead conversion rate by the number of form fields; spin-to-win and other gamified popups are counted separately. Fields Campaigns Median 1 field 1,224 0.88% 2 fields 736 0.94% 3 fields 493 0.56% 4-5 fields 374 0.85% 6+ fields 75 0.57% Used in Multi-Step Forms for Shopify: When Smarter Popups Pay Off
Copy and buttons
Part 6: Copy and CTA- First-person buttons ("Send my code") convert 1.18% vs 0.80% for neutral wording (English popups, median campaign).
- LeanCTA buttons that promise value ("Get", "Claim", "Unlock") look stronger than "Subscribe" / "Join" buttons (1.06% vs 0.77%, English popups, median campaign), but the gap is not statistically reliable.
- LeanEmoji in the headline or button come with a slightly higher click rate (2.76% vs 2.20%) but no measurable difference in leads; the click difference is not statistically reliable.
- No measurable differenceButton colour, dark vs light popup backgrounds, urgency words and question headlines show no measurable effect on conversion; the data are too noisy to rule out modest effects.
Median lead conversion rate by the button's verb, English popups. The gaps between these three are not statistically reliable. Button verb Median Get / Claim / Unlock 1.06% Subscribe / Join 0.77% Submit / Send 0.84% Used in What is a call to action (CTA)? Definition and examples · What is an A/B split test?
Timing and triggers
Part 3: Timing and triggers- Popups convert better in the morning (06:00–11:59 visitor time) than at night or in the evening: the same popup converts better in the morning than between midnight and 6 a.m. in 69% of campaigns.
- LeanPopups that wait before appearing convert 0.83% vs 0.61% for popups shown the instant the page loads (median campaign); the direction holds in every year of data but the difference is not statistically reliable, so treat it as a lean, not a rule.
- LeanExit intent shows no statistically reliable advantage: popups on the most sensitive exit-intent setting converted 1.10% vs 0.78% without exit intent (median campaign), a gap that could be chance.
- No measurable differenceThe exact delay length (0–5 seconds vs 21–60 seconds) makes no measurable difference to conversion.
- No measurable differenceWeekend visitors show no measurable difference in popup conversion from weekday visitors once each popup is compared with itself (2,333 campaigns).
Median lead conversion rate by when the popup appears. A lean: the gap is not statistically reliable. Trigger Campaigns Median Time delay 1,159 0.83% Immediately on page load 557 0.61% Used in Popup Timing: The Best Time to Show a Popup (2026) · Do Exit Intent Popups Really Work? 2026 Benchmark Data
Frequency and new visitors
Frequency and fatigue- The first two times a visitor sees a popup in a week convert at similar rates (1.33% and 1.22% of displays, pooled); from the third view conversion falls, and from the sixth view it is 0.41%, 31% of the first-view rate.
- New visitors convert on popups more often than visitors who were on the site in the previous 7 days: the same popup converts better for new visitors in 75% of campaigns (median gap 0.30 percentage points; pooled 1.28% vs 0.54% of displays).
- No measurable differenceShowing a popup a second time caused no measurable drop in conversion within the same campaign; the decline begins with the third showing.
Used in Popup Timing: The Best Time to Show a Popup (2026) · What Is A Popup? Popup vs Pop up
Page and traffic source
Part 5: Where the popup appears and where visitors come from- Visitors arriving from an email or SMS campaign convert on popups at 1.81% vs 0.74% for visitors from organic search.
- Popups on cart and checkout pages convert 4.57% vs 1.07% on the home page (median campaign); blog pages convert 0.23%.
- When a visitor's browser language matches the popup's language, the same popup converts better in most campaigns.
- Popups targeted at new visitors only convert 1.13% vs 0.77% for popups shown to everyone.
- Visitors arriving from social media convert on popups at 1.22% vs 0.71% for direct visitors (median campaign).
- By visitor country, popups convert 0.76% for visitors in the United Kingdom vs 0.34% in Germany, and 2.01% in Mexico vs 0.46% in the United States (median campaign).
- LeanVisitors referred by AI assistants (ChatGPT, Gemini, Claude, Perplexity, Copilot) show no statistically reliable advantage over organic search: popups converted 1.14% vs 0.74% (median campaign), but only 91 campaigns had enough AI-referred traffic, so the gap could be chance.
- CorrelationPopups set up with three targeting rules convert 1.09% vs 0.71% for popups with none or one (median campaign): a correlation, since marketers who target carefully may also write better offers.
Median lead conversion rate by page type. Levels, not a ranking: only the gaps in the statements were tested. Page type Campaigns Median Cart / checkout 90 4.57% Account / login 79 2.87% Pricing 21 2.44% Home 2,138 1.07% Collection / category 553 1.03% Product 835 0.79% Contact / about 152 0.59% Blog / content 364 0.23% Median lead conversion rate by where the visitor came from. Levels, not a ranking. Traffic source Campaigns Median Email / SMS 322 1.81% Social 1,043 1.22% Referral 568 1.14% AI assistants 91 1.14% Same-site navigation 1,964 1.00% Paid ads 1,350 0.84% Organic search 1,958 0.74% Direct 2,128 0.71% Used in What is a good website conversion rate? 2026 benchmarks · What is a conversion rate in marketing?
Mobile and desktop
Part 4: Device and visitor- Mobile visitors convert on popups at 1.02% vs 0.57% on desktop, and click at 2.99% vs 1.34% (median campaign).
- 63% of popup displays happen on mobile.
- On the same popup, Android visitors click more often than iPhone visitors.
- Android visitors also convert slightly more often than iPhone visitors on the same popup: iPhone visitors converted better in only 38% of 2,059 campaigns (median gap 0.06 percentage points).
- Visitors on smaller phone screens (under 360 px wide) click popups more often than visitors on standard 390–429 px phones.
- No measurable differenceVisitors browsing inside the Instagram or Facebook app show no measurable difference in popup conversion from visitors in a normal mobile browser.
Used in Mobile Popup Design: Examples & Best Practices (2026)
Website platform
Part 7: Industry and platform- 35% of popup campaigns on Shopify stores ask for an email or phone number, against 25% on WordPress sites.
- Popups on Shopify stores convert at 1.07% vs 0.66% on WordPress sites (median campaign).
- Popups on Webflow sites convert 0.37% vs 1.07% on Shopify stores (median campaign; 68 lead-collecting Webflow campaigns).
- No measurable differenceShopify (1.07%) and WooCommerce stores (0.82%) show no statistically reliable difference in popup conversion; the data cannot rule out a gap in either direction.
Used in Shopify Conversion Rate Statistics 2026 & Key Insights
Integrations
Part 8: Integrations and lead quality- Among businesses whose popups collect an email or phone number, 16% connect an external tool (email platform, CRM or webhook) and 64% turn on the built-in autoresponder email; across all businesses in the benchmark, 93% connect no external tool.
- Among external integrations, Mailchimp (5.2% of lead-collecting businesses) and Brevo (4.5%) are the most connected email tools; the Shopify Customers sync is used by 10.7%.
- The share of new campaigns connected to an external tool fell from 7.0% in 2024 to 3.1% in 2026, while native Shopify Customers sync rose from 1.0% to 9.1%.
- CorrelationPopups connected to Brevo convert at 1.27% and popups synced to Shopify Customers at 1.23%, vs 0.63% for popups with no integration (median campaign): a correlation, since marketers who connect a tool also tend to run better campaigns.
- CorrelationPopups connected to Klaviyo convert at 1.44% vs 0.63% for popups with no integration (median campaign), on a small sample (47 campaigns from 23 businesses) and again a correlation, not an effect of the connection.
Share of lead-collecting businesses that use each connection. Connection Share Built-in autoresponder email 63.6% Shopify Customers sync 10.7% Mailchimp 5.2% Brevo 4.5% Webhook 2.5% Klaviyo 0.9% HubSpot 0.8% MailerLite 0.8% ActiveCampaign 0.7% SendGrid 0.6% Customer.io 0.3% EmailOctopus 0.3% Used in Email marketing guide
Lead quality
Part 8: Integrations and lead quality- 59% of email addresses collected by popups are Gmail addresses; 12% use a business or other domain.
- Emails collected on desktop are about twice as likely to be at a business or other non-consumer domain: 21% vs 9.5% on mobile.
- The share of popup leads that include a phone number rose from 21.3% (2024) to 26.3% (2025) to 33.5% (2026) (all leads pooled); much of that comes from a few high-volume accounts that collect phone numbers, while most lead-collecting campaigns collect email only.
Used in Email marketing guide
Methodology: how we built the benchmark
What the numbers on this page count, which campaigns they cover, and how a gap earns the word "difference".
Scope
The dataset covers 1.98 billion page views, 499.6 million popup displays and 1.84 million leads: 28,674 campaigns from 11,533 businesses that ran a popup with Popupsmart. The window runs from 1 January 2024 to 23 September 2026.
Rates use a stricter set: 10,768 campaigns with at least 1,000 displays and 14 active days, from 4,559 businesses. Lead-rate medians use the 3,034 of those popups that ask for an email or phone number.
Definitions
- Display
- A popup shown to a visitor. A teaser impression also counts as a display.
- Click
- A click on a popup element set to count as an interaction, plus teaser clicks.
- Lead
- A popup submission that contains an email address or phone number, counted once per campaign per page load. Spins, coupon reveals and step-1 clicks without contact details are not leads.
- Lead conversion rate
- Leads divided by displays, for popups that ask for an email or phone number, reported as the median campaign.
- Click rate
- Clicks divided by displays, reported as the median campaign.
- Close
- An explicit close: the X button, a click on the overlay or the Esc key. Close tracking began in January 2025, so close rates use displays from February 2025 on.
- Median and pooled
- A median takes each campaign's rate and finds the middle, so every campaign counts once; the top 25% and top 10% lines come from the same distribution. A pooled rate divides all leads by all displays, so big sites dominate it. A finding says "pooled" when it uses one.
- Device medians
- Campaigns with at least 500 displays on that device.
How a finding is judged
Each comparison went through a bootstrap test (20,000 draws per claim) across years and inclusion rules, with a 5% false-discovery-rate correction across all the comparisons tested at once. Each statement on this page is written from its verdict:
- Held up (no tag): the gap's 95% confidence interval excludes zero and it passes every check. It keeps its direction in every year with data, without the 10 largest accounts, under looser and stricter inclusion rules, and among campaigns whose settings rarely changed.
- Lean: the interval excludes zero, but the gap fails one of those checks, so it may be chance.
- No measurable difference: the interval includes zero. The data can't tell the groups apart, which isn't proof that they're equal.
- Correlation: a gap that passes the checks between campaigns whose marketers also differ in other ways, such as those who connect an email tool. It describes who does something, not what a setting causes.
A gap that passes narrowly is published only with its confidence interval, which is why the popup-format comparison is not on this page yet.
Industry, platform and company size
- Industry comes from LinkedIn company data matched on each website's domain.
- Platform comes from the Popupsmart Shopify app install, Popupsmart's CMS detection when a website is added, or a technology crawl, in that order.
- Company size is the LinkedIn employee band of each business's main website, the domain with the most displays. It is known for 3,309 of 11,533 businesses (29%). Shares are of the businesses with a known size; businesses without a matching LinkedIn page are not in them.
Cleaning and validation
Bot user agents are removed, and templates, internal accounts and suspended accounts are excluded. Every figure was recomputed independently and spot-checked against raw production records. Rerunning the pipeline on the same raw data gives identical output.
Privacy
Every published segment covers a minimum number of campaigns and accounts, and a segment is also hidden when its value could be worked out by subtracting visible ones. A mobile and desktop split appears only when both device rows meet the minimum. Platforms and company-size bands below the minimum are grouped, not hidden, so the shares still add up. No single customer can be identified from these figures.
Limitations
- The data covers only websites that run Popupsmart, so it describes those sites, not the whole web.
- Rates are per display, not per visitor. A visitor who sees a popup three times counts three times.
- Close rates start in February 2025, when close tracking began.
- Creative features, such as discount size, button verb, emoji and colour, are read from each popup's main page. The verb and first-person results cover English popups.
- Some dimensions are configured settings (trigger, targeting, form fields); others are observed behaviour (device, traffic source, time to close). A setting records what the marketer chose, not what every visitor experienced.
- Most comparisons are between different campaigns, not A/B tests, so a gap can reflect who uses a setting as well as the setting itself. Where a finding says "the same popup", the comparison was made within campaigns.
Download and cite the data
- popup-benchmark-2026.csvOne row per statement, metric and breakdown row, with its unit23 KB
- popup-benchmark-2026.jsonThe same data, with the methodology, license and citation23 KB
Free to reuse under the CC BY 4.0 license, with credit to Popupsmart.
Popupsmart, "Popup Conversion Benchmark Report 2026" (data January 2024 to September 2026), https://popupsmart.com/blog/popup-conversion-benchmark-report
Every page that uses this data
Where a post, a definition or a guide quotes the 2026 benchmark, it is listed here.
The report 1
Encyclopedia 11
- Click-Through Rate (CTR) Definition - What is CTR?
- Sidebar Popup Definition- What Is A Sidebar Popup?
- What is a call to action (CTA)? Definition and examples
- What is a conversion rate in marketing?
- What is a Floating Bar? - Floating Bar Definition
- What is a Lightbox Popup?
- What Is A Popup? Popup vs Pop up
- What is a sense of urgency in marketing?
- What is an A/B split test?
- What Is Full Screen Popup? (Welcome Mat Popup)
- What is Gamification in Marketing? - Gamification Definition
Blog posts 15
- 10 Alia Alternatives We've Tested: More Flexibility, Better Pricing (2026 Guide)
- 13 Opt-In Page Examples for Effective Lead Generation
- 55 Exit Intent Popup Examples to Increase Conversions
- Best OptiMonk Alternatives: 7+ Years Building & Testing Popups (2026)
- Best Popup Software in 2026: I Built and Tested Popup Software for 7 Years
- Cart Abandonment Statistics for 2026
- Do Exit Intent Popups Really Work? 2026 Benchmark Data
- Mobile Popup Design: Examples & Best Practices (2026)
- Multi-Step Forms for Shopify: When Smarter Popups Pay Off
- Popup best practices: we analyzed 1.98 billion page views and 28,674 Popupsmart campaigns
- Popup Timing: The Best Time to Show a Popup (2026)
- Popup Video: How To Add Video Popup To Your Website in 2026 (No Coding)
- Popups Convert: Key Statistics & 12 Proven Tips for Success
- Shopify Conversion Rate Statistics 2026 & Key Insights
- What is a good website conversion rate? 2026 benchmarks
Questions about our data
Where does Popupsmart's popup data come from?
From the analytics events and campaign settings of 28,674 campaigns that 11,533 businesses ran with Popupsmart between January 2024 and September 2026. Industry and company size come from LinkedIn company data matched on each website's domain; platform from the Shopify app install, Popupsmart's CMS detection or a technology crawl.
Can I use Popupsmart's popup data in my own work?
Yes. The statements, metrics and breakdowns are free to reuse under the CC BY 4.0 license, with credit to Popupsmart and a link to the 2026 benchmark report.
Why does Popupsmart report medians instead of averages?
A median takes each campaign's rate and finds the middle, so every campaign counts once. A pooled rate divides all leads by all displays, so big sites dominate it; findings say "pooled" where they use one.
Can I compare these numbers with Popupsmart's 2025 report?
No. The 2025 report used different definitions that can't be reproduced (leads divided by interactions, averaged), so its figures are withdrawn, not restated.