Companies Using Cloudflare Workers
We detect Cloudflare Workers on 18,506 domains, which ranks it #12 of the 41 ranked technologies in PaaS at a 0.79% share. Before anything else on this page: that is a floor, and we can show why. Workers becomes visible to an outside scan in exactly two ways. A page can declare the runtime with a meta tag, <meta name="runtime" content="Cloudflare Workers">, or a page can make requests to a .workers.dev address. A Worker serving a production site on the company's own domain emits neither, and that is the normal way Workers is deployed.
The size of the gap shows up in our own data: across every domain where we currently detect Workers, exactly 11 sit on a workers.dev hostname, fewer than one in a thousand. So read this list of companies using Cloudflare Workers as the deployments that advertise themselves, never as a census. Below you will find the brands that use Cloudflare Workers, market share, and the size, industry and country breakdowns drawn from the 8,757 domains we could match to a company record. Data updated monthly.
Published Mar 31, 2026 · Updated Aug 31, 2026 · Data analysed on August 31, 2026.
Cloudflare Workers Usage Statistics
This chart shows our current crawl only: 13,657 active domains in June 2026, 17,122 in July and 18,506 in August 2026. Three readings on one basis, and no adoption rate should be read across them, because the domain set we crawl turns over every month and the count moves with coverage as much as with new deployments. The churn figure needs the same care. 6,373 domains dropped a detection since we began tracking in March 2026, but 68.1% of those rows were seen exactly once, against 47.4% for Cloudways, 54.8% for Supabase and 55.9% for Render. A runtime meta tag only appears on the routes a Worker actually renders, so whether we see it depends on which page of a site the crawl happens to land on. Most of that churn is our visibility flickering, not sites leaving Workers.
List of Companies Using Cloudflare Workers
Download all 18,506 Cloudflare Workers customers with full company data, or create a signal to track when companies start or stop using Cloudflare Workers.
| Company | Detection URL | Domain | Country | Industry | Employees | Type | Founded | |
|---|---|---|---|---|---|---|---|---|
| registrar.charlotte.edu | charlotte.edu | United States | Higher Education | 1001-5000 | Educational | 1946 | https://linkedin.com/company/unc-charlotte | |
| flights.virginatlantic.com | virginatlantic.com | United Kingdom | Airlines and Aviation | 5001-10000 | Privately Held | 1984 | https://linkedin.com/company/virgin-atlantic-airways | |
| keyence.com | keyence.com | Japan | Automation Machinery Manufacturing | 10001+ | Public Company | 1974 | https://linkedin.com/company/keyence | |
| resources.purolator.com | purolator.com | Canada | Freight and Package Transportation | 10001+ | Privately Held | 1960 | https://linkedin.com/company/purolatorinc | |
| holidays.royalairmaroc.com | royalairmaroc.com | Morocco | Airlines and Aviation | 1001-5000 | Privately Held | 1957 | https://linkedin.com/company/royal-air-maroc | |
| flights.flyfrontier.com | flyfrontier.com | United States | Airlines and Aviation | 5001-10000 | Public Company | 1994 | https://linkedin.com/company/frontier-airlines | |
| halfords.com | halfords.com | United Kingdom | Retail | 10001+ | Public Company | 1892 | https://linkedin.com/company/halfords | |
| drc.ngo | drc.ngo | Denmark | Non-profit Organizations | 5001-10000 | Nonprofit | 1956 | https://linkedin.com/company/danish-refugee-council | |
| oakstreethealth.com | oakstreethealth.com | United States | Hospitals and Health Care | 5001-10000 | Public Company | 2012 | https://linkedin.com/company/oak-street-health | |
| sesiescolas.fiemg.com.br | fiemg.com.br | Brazil | Information Services | 5001-10000 | Privately Held | 1933 | https://linkedin.com/company/fiemg |
Show 40 more Cloudflare Workers using companies as demo data
| Company | Detection URL | Country | Industry | Employees | Type | Founded | ||
|---|---|---|---|---|---|---|---|---|
| weareunknown.io | weareunknown.io | United Kingdom | Staffing and Recruiting | 11-50 | Partnership | 2018 | https://linkedin.com/company/weareunknown | |
| flights.airindiaexpress.com | airindiaexpress.com | India | Airlines and Aviation | 1001-5000 | Government Agency | — | https://linkedin.com/company/airindiax | |
| elizabethwiseman.com | elizabethwiseman.com | United States | Artists and Writers | 1-10 | Self-Owned | — | https://linkedin.com/company/real-life-rituals-llc | |
| werkenbijcordaan.nl | werkenbijcordaan.nl | Netherlands | Hospitals and Health Care | 5001-10000 | Nonprofit | 1683 | https://linkedin.com/company/cordaan | |
| ftmo.com | ftmo.com | Czechia | Education | 51-200 | Privately Held | 2017 | https://linkedin.com/company/ftmo | |
| casagrand.co.in | casagrand.co.in | India | Real Estate | 1001-5000 | Privately Held | 2003 | https://linkedin.com/company/casagrand-private-limited | |
| portal.fmu.br | fmu.br | Brazil | Higher Education | 1001-5000 | Educational | 1968 | https://linkedin.com/company/centro-universitario-fmu | |
| dfs.co.uk | dfs.co.uk | United Kingdom | Retail | 1001-5000 | Public Company | 1969 | https://linkedin.com/company/dfsfurniture | |
| flights.skyairline.com | skyairline.com | Chile | Aviation and Aerospace Component Manufacturing | 1001-5000 | Privately Held | 2002 | https://linkedin.com/company/sky-airline | |
| ethanallen.com | ethanallen.com | United States | Retail | 1001-5000 | Public Company | 1932 | https://linkedin.com/company/ethan-allen | |
| marketplace.k12.com | k12.com | United States | E-Learning Providers | 1001-5000 | Public Company | 1999 | https://linkedin.com/company/k12 | |
| tshub.tastefullysimple.com | tastefullysimple.com | United States | Food & Beverages | 51-200 | Privately Held | 1995 | https://linkedin.com/company/tastefully-simple | |
| francescas.com | francescas.com | United States | Retail | 5001-10000 | Privately Held | 1999 | https://linkedin.com/company/francesca-s- | |
| mibebeyyo.elmundo.es | elmundo.es | Spain | Newspaper Publishing | 201-500 | Privately Held | 1989 | https://linkedin.com/company/el-mundo | |
| comphealth.com | comphealth.com | United States | Staffing and Recruiting | 1001-5000 | Privately Held | 1979 | https://linkedin.com/company/comphealth | |
| academy.writer.com | writer.com | United States | Software Development | 51-200 | Privately Held | 2020 | https://linkedin.com/company/getwriter | |
| n-able.com | n-able.com | United States | IT Services and IT Consulting | 1001-5000 | Public Company | — | https://linkedin.com/company/n-able | |
| chghealthcare.com | chghealthcare.com | United States | Staffing and Recruiting | 1001-5000 | Privately Held | 1979 | https://linkedin.com/company/chg | |
| nerine-floral.squarespace.com | squarespace.com | United States | Software Development | 1001-5000 | Privately Held | 2003 | https://linkedin.com/company/squarespace | |
| homeserve.co.uk | homeserve.co.uk | United Kingdom | Insurance | 1001-5000 | Public Company | 1993 | https://linkedin.com/company/homeserve | |
| muhammadiyah.or.id | muhammadiyah.or.id | Indonesia | Religious Institutions | 10001+ | Nonprofit | 1912 | https://linkedin.com/company/muhammadiyahuk | |
| salesianos.br | salesianos.br | Brazil | Philanthropic Fundraising Services | 1001-5000 | Nonprofit | 1947 | https://linkedin.com/company/salesianos | |
| employmenthero.com | employmenthero.com | Australia | Human Resources Services | 501-1000 | Privately Held | 2014 | https://linkedin.com/company/employment-hero | |
| kgmu.org | kgmu.org | India | Hospitals and Health Care | 5001-10000 | Educational | 1905 | https://linkedin.com/company/kgmu | |
| apply.uopeople.edu | uopeople.edu | United States | Higher Education | 201-500 | Nonprofit | 2009 | https://linkedin.com/company/university-of-the-people | |
| employer.coursecareers.com | coursecareers.com | United States | E-Learning Providers | 2-10 | Privately Held | 2019 | https://linkedin.com/company/coursecareers | |
| rcwilley.com | rcwilley.com | United States | Retail | 1001-5000 | Public Company | 1932 | https://linkedin.com/company/rc-willey | |
| flights.rwandair.com | rwandair.com | Rwanda | Airlines and Aviation | 1001-5000 | Government Agency | 2001 | https://linkedin.com/company/flyrwandair | |
| findlay.edu | findlay.edu | United States | Higher Education | 501-1000 | Educational | 1882 | https://linkedin.com/company/the-university-of-findlay | |
| flights.azoresairlines.pt | azoresairlines.pt | Portugal | Airlines and Aviation | 1001-5000 | Privately Held | 1947 | https://linkedin.com/company/sata | |
| blog.belaysolutions.com | belaysolutions.com | United States | Staffing and Recruiting | 201-500 | Privately Held | 2010 | https://linkedin.com/company/belay-solutions | |
| anteriad.com | anteriad.com | United States | Advertising Services | 501-1000 | Privately Held | 2000 | https://linkedin.com/company/anteriad | |
| ibyte.com.br | ibyte.com.br | Brazil | Retail | 1001-5000 | Privately Held | 2000 | https://linkedin.com/company/ibyte | |
| www.acg.edu | acg.edu | Greece | Higher Education | 1001-5000 | Nonprofit | 1875 | https://linkedin.com/company/deree-the-american-college-of-greece | |
| meowwolf.com | meowwolf.com | United States | Entertainment Providers | 1001-5000 | Privately Held | 2008 | https://linkedin.com/company/meow-wolf | |
| www.bioagilytix.com | bioagilytix.com | United States | Biotechnology Research | 1001-5000 | Privately Held | 2008 | https://linkedin.com/company/bioagilytix | |
| eaglenet.meridiancc.edu | meridiancc.edu | United States | Higher Education | 501-1000 | Educational | 1937 | https://linkedin.com/company/meridiancc | |
| chinoin.com | chinoin.com | Mexico | Pharmaceutical Manufacturing | 1001-5000 | Self-Owned | 1925 | https://linkedin.com/company/chinoinpf | |
| www.globaluniversitysystems.com | globaluniversitysystems.com | United Kingdom | Higher Education | 1001-5000 | Privately Held | 2003 | https://linkedin.com/company/global-university-systems | |
| help.xing.com | xing.com | Germany | Software Development | 1001-5000 | Public Company | 2003 | https://linkedin.com/company/xing |
There are 18,506 companies and domains where we detect Cloudflare Workers, sign up to download the entire Cloudflare Workers dataset.
Recognizable companies using Cloudflare Workers and brands using Cloudflare Workers in 2026, with the exact host we detected the runtime on:
- Keyence detected on keyence.com
- Virgin Atlantic detected on flights.virginatlantic.com
- Halfords detected on halfords.com
- Purolator detected on resources.purolator.com
- Frontier Airlines detected on flights.flyfrontier.com
- Royal Air Maroc detected on holidays.royalairmaroc.com
- Air India Express detected on flights.airindiaexpress.com
- DFS Furniture detected on dfs.co.uk
- Ethan Allen detected on ethanallen.com
- El Mundo detected on mibebeyyo.elmundo.es
- XING detected on help.xing.com
- Oak Street Health detected on oakstreethealth.com
- Employment Hero detected on employmenthero.com
- Danish Refugee Council detected on drc.ngo
- Meow Wolf detected on meowwolf.com
- N-able detected on n-able.com
Note how many of these are booking, help or campaign hosts rather than a main corporate site. That is what this detection surface picks up, and it means each entry evidences one team's deployment rather than a whole estate.
Which Countries Use Cloudflare Workers the Most?
Which countries use Cloudflare Workers the most? The United States leads with 3,824 domains, followed by the United Kingdom (661), the Netherlands (385), Canada (377), India (324) and Australia (313). Germany (235), Brazil (228) and France (204) come next, then Spain (118), Norway (77), Italy (75) and Switzerland (67). The Netherlands ahead of both Canada and India is the row that stands out, on a considerably smaller web than either. These are counts of the 8,757 domains we matched to a company record with a resolved country, so they describe where Workers-carrying sites are registered, not where anything is billed. The visibility caveat applies here too: a country whose sites happen to run Workers behind custom domains will look smaller than it is.
Share of the 8,757 Cloudflare Workers domains we matched to a company record — 47.3% of the 18,506 we detect. Top 14 countries shown, so these do not sum to 100%. Domains with no matching company record are not represented here and skew small and independent.
Cloudflare Workers Market Share Among PaaS
What is Cloudflare Workers' market share? Workers holds 0.79% of the PaaS category, ranking #12 of 41 ranked technologies, a long way behind Amazon Web Services (39.98%), Vercel (16.64%), WordPress.com (8.24%), Azure (7.5%) and Firebase (6.87%). That ordering says less about adoption than it appears to, because these platforms are not measured through equally wide windows. AWS, Azure and Firebase leave artefacts a site cannot easily avoid: hostnames, SDK files, storage endpoints. Workers leaves a self-declared runtime meta tag or a call to workers.dev, and a production deployment on a custom domain leaves neither. Compare these shares only with that asymmetry in mind.
Top Competitors by Market Share
Cloudflare Workers Customers by Company Size & Age
Is Cloudflare Workers only for small companies? 60.23% of the domains we can size have 1 to 10 employees, 4,967 out of 8,247, with another 26.19% (2,160 domains) in the 11 to 50 band. Those two bands together are more than 86% of the sized base. At the other end there are just 11 domains at 10,001 or more employees and 15 in the 5,001 to 10,000 band. Read that top end as a floor inside a floor. The large-company detections we do have sit on hosts like flights.virginatlantic.com and resources.purolator.com, which is one team's route rather than a company standard, and an enterprise Worker running quietly behind a custom domain never reaches this table at all.
Company Size Distribution
Company Age (Founded Decade)
What Industries Use Cloudflare Workers the Most?
Retail tops the table at 8.30%, 633 domains, ahead of Real Estate at 4.92% (375), Software Development at 3.89% (297) and Technology, Information and Internet at 3.26% (249). Retail Apparel and Fashion also makes the top ten at 2.24% (171). That is the surprise on this page. An edge compute runtime documented for developers should lead with software, and instead the two retail rows together outweigh Software Development and IT Services and IT Consulting (2.70%) combined. The stack data points the same way from another direction: Shopify runs on 26.55% of the Workers domains we can profile.
Retail brands using Cloudflare Workers lead at 8.30%, or 633 domains, and the detections behind that row are ordinary storefronts rather than engineering properties: halfords.com, dfs.co.uk, ethanallen.com, rcwilley.com, francescas.com and ibyte.com.br. Real estate companies on Cloudflare Workers follow at 4.92% (375 domains). Two further clusters repeat in the company list even though neither reaches the industry top ten. The first is travel: we detect Workers on flights.virginatlantic.com, flights.flyfrontier.com, flights.airindiaexpress.com, flights.rwandair.com, flights.azoresairlines.pt and flights.skyairline.com, six airline booking hosts on the same naming pattern, plus holidays.royalairmaroc.com. The second is higher education: registrar.charlotte.edu, apply.uopeople.edu, findlay.edu, eaglenet.meridiancc.edu and portal.fmu.br. Every percentage here describes the subset of domains we matched to a company with a resolved industry, not the full 18,506 detections.
Cloudflare Workers Alternatives & Competitors
Cloudflare Workers ranks #12 of 41 ranked PaaS technologies at 0.79%. Amazon Web Services leads with 936,363 domains (39.98%), then Vercel at 389,653 (16.64%), WordPress.com at 193,092 (8.24%), Azure at 175,651 (7.5%) and Firebase at 160,932 (6.87%). Do not read that as a scoreboard. Workers is edge compute that usually sits in front of an origin rather than replacing one, so a domain can legitimately appear in this row and in the AWS row at the same time. Our data shows exactly that: of the domains that have ever run both Workers and Amazon Web Services, 72.3% still run both, and for Vercel the figure is 81.1%. The distance between 0.79% and 39.98% is partly market size and partly the fact that one of the two is measured through a much narrower window.
| Technology | Domains | Market Share |
|---|---|---|
| 936,363 | 39.98% | |
| 389,653 | 16.64% | |
| 193,092 | 8.24% | |
| 175,651 | 7.5% | |
| 160,932 | 6.87% |
Tech Stack of Cloudflare Workers-Powered Websites
Cloudflare appears on 77.33% of the Workers domains we can profile, which is close to definitional rather than a choice these teams made: Workers is a Cloudflare product and runs on Cloudflare's network. The informative rows sit underneath it. Google Tag Manager at 65.70% and Google Analytics at 63.93% put these domains firmly in marketing-instrumented territory, while Cloudflare Browser Insights reaches only 22.96%, so most of these sites buy compute from Cloudflare and measure with Google. The most surprising row is Shopify at 26.55%. Roughly a quarter of the deployments we can see are storefronts, which reads as edge routing, redirects and personalisation on shops rather than general application hosting. jsDelivr sits alongside it at 25.77%, and WooCommerce is far behind at 1.70%.
Web Analytics
Tag Management
Ecommerce Platforms
CDN
Expert Analysis: Cloudflare Workers Growth Trends & Key Signals for Sales Teams in 2026

With 10+ years in web crawling and technographic data analysis, I have gone through what our corpus can and cannot see about Cloudflare Workers. This analysis draws on 18,506 detected domains as of our August 2026 crawl and the 8,757 of them we could match to a company record. One point governs everything else here, so it goes first rather than into a footnote.
1. This number is a floor, and we can prove it
Workers is recognisable from outside in only two ways. A page can declare it with a runtime meta tag, <meta name="runtime" content="Cloudflare Workers">, or a page can make requests to a .workers.dev address. Both are things a deployment does incidentally, and neither is required for a Worker to serve traffic. The standard production setup, a Worker bound to a route on the company's own domain, produces neither signal. From outside it is indistinguishable from any other origin served through Cloudflare.
How large is the blind spot? Our own data gives a hard bound on one half of it. Of every domain where we currently detect Workers, 11 are on a workers.dev hostname. Eleven. That is fewer than one in a thousand, which means the workers.dev signal contributes almost nothing and that essentially the entire count rests on sites that voluntarily advertise the runtime. There is no honest way to extrapolate a true population from that, and I am not going to try. What I can state is the direction and the consequence: the real number of sites running Workers is much larger than 18,506, and every share, rank, country count and industry percentage below inherits the same bias.
"Most measurement problems are accuracy problems. This one is a visibility problem. We are not mismeasuring Workers, we are seeing a self-selected slice of it, and the honest thing is to say so at the top of the page rather than publish a rank as though it were a market position." - Mehmet Suleyman, CEO at TechnologyChecker.io
2. What the chart shows, and what the churn means
The usage chart covers the current crawl only: 13,657 active domains in June 2026, 17,122 in July and 18,506 in August. Three readings on one basis. No adoption rate should be computed across them, because the crawled domain set turns over every month and coverage moves the line as much as deployments do.
The churn column is where the detection surface shows itself most clearly. 6,373 domains dropped a detection since we began tracking in March 2026, which reads like heavy attrition until you split it. 68.1% of those rows were seen exactly once. The same measure is 47.4% on Cloudways, 54.8% on Supabase and 55.9% on Render, all platforms in adjacent territory. Workers sits 12 to 21 points above every one of them, and the explanation is mechanical rather than commercial: a runtime meta tag appears only on the routes a Worker renders, so a crawl that lands on a marketing homepage one month and a Worker-rendered route the next produces a detection, a drop and a re-detection with nothing having changed on the site. Genuine departures are in there. They are not the bulk of it.
Sales Signal: do not build a churn or win-back list from this column. On a technology with an intermittent signal, a list of domains that dropped Workers is mostly domains that still run Workers, and outreach built on it misfires in an obvious way.
3. Customer profile
60.23% of the domains we can size have 1 to 10 employees, 4,967 out of 8,247, with another 26.19% in the 11 to 50 band, so more than 86% of the sized base is under 50 people. Only 11 domains belong to companies of 10,001 or more and 15 to the 5,001 to 10,000 band. On age, 40.75% of the companies we can date were founded in the 2010s and another 25.93% in the 2020s, so two thirds postdate 2010.
I would resist the obvious reading of that. It is not evidence that Workers is a small-business product. It is evidence that small, young companies are the ones most likely to leave a runtime tag in the page, ship on a preview hostname, or never move a Worker behind a custom domain at all. A large organisation running Workers the normal way disappears from this table entirely. The enterprise detections we do have make the point: Keyence on keyence.com is one of very few main-domain hits, and the rest are hosts such as flights.virginatlantic.com, resources.purolator.com, holidays.royalairmaroc.com and marketplace.k12.com. Those are single routes, built by one team.
One row belongs in the caveats rather than the findings: our detection on nerine-floral.squarespace.com is a Squarespace-hosted customer site, not Squarespace itself deploying Workers. Any list built from platform subdomains needs that filter applied by hand.
Sales Signal: 11 domains at 10,001+ employees is not a market, it is a hand-list. The reachable Workers audience in this data is the micro-business and agency tier, and it is reached through self-serve tooling, templates and marketplace presence rather than named-account outreach.
4. The commerce cluster is the real finding
Asked to guess the top industry for an edge compute runtime, most people would say software. It is Retail, at 8.30% and 633 domains, with Retail Apparel and Fashion also in the top ten at 2.24%. Together those two outweigh Software Development (3.89%) and IT Services and IT Consulting (2.70%) combined. The technology stack confirms it from an independent direction: Shopify runs on 26.55% of the Workers domains we can profile.
A quarter of the visible Workers population sits on Shopify stores. That is not what a general-purpose application platform looks like in the data. It is what edge routing looks like: redirects, geo and currency logic, A/B tests, bot rules and personalisation layered in front of a hosted storefront the merchant does not control. It also fits the visibility argument, because that kind of Worker is often bolted on by an agency, and agency work is exactly the population most likely to leave a runtime tag in place. The retail names in the list run the same way: halfords.com, dfs.co.uk, ethanallen.com, rcwilley.com, francescas.com and ibyte.com.br.
Geographically the United States leads with 3,824 domains, then the United Kingdom (661), the Netherlands (385), Canada (377), India (324) and Australia (313). The Netherlands above both Canada and India is the anomaly, on a considerably smaller web than either.
Sales Signal: the sharpest segment on this page is Workers plus Shopify. Those merchants have already decided their storefront platform cannot do something they need at the edge, which makes them buyers for routing, experimentation, personalisation and bot control, and they are reachable through the Shopify app ecosystem rather than one at a time.
5. Why this page has no switching section
Most profiles here carry a migration table. This one does not, and the omission is deliberate. Workers is edge compute that sits in front of an origin rather than replacing it, so pairing it against a cloud provider produces complements, not substitutes. The numbers say so plainly: of the domains that have ever run both Workers and Amazon Web Services, 72.3% still run both, and for Vercel the figure is 81.1%. A corridor built from the remainder would describe sites dropping an edge layer while keeping their origin, which is a different event from choosing one vendor over another. Publishing it as switching would invent a competitive story our data does not support.
The same reasoning applies to the category table. At 0.79% against Amazon Web Services at 39.98%, the gap measures market size and detection width together, not a contest anyone lost.
"A switching number needs the two products to be alternatives. Workers and AWS mostly are not, and more than seven in ten domains that ever ran both still run both. Publishing a corridor there would be a decision to look authoritative rather than to be right." - Mehmet Suleyman, CEO at TechnologyChecker.io
6. Technology ecosystem
One row needs discounting before the rest can be read. Cloudflare at 77.33% is close to definitional, since Workers is a Cloudflare product running on Cloudflare's network. It is not an independent choice and should not be presented as one.
What is left is informative. Google Tag Manager at 65.70% and Google Analytics at 63.93% describe marketing-instrumented websites rather than pure API backends, consistent with the commerce skew above. Cloudflare Browser Insights reaches only 22.96%, so roughly two thirds of these domains buy compute from Cloudflare and measure with Google. jsDelivr at 25.77%, cdnjs at 19.28% and Unpkg at 10.80% show that a lot of these pages still pull libraries from public CDNs, which is not the profile of a heavily engineered application.
Sales Signal: the Browser Insights gap is a straightforward consolidation play for anyone selling observability or analytics into this base. These teams are already Cloudflare customers and are already measuring somewhere else.
7. Key takeaways
1. The count is a floor, not a total: 18,506 domains is what advertises itself. Only 11 domains in our current detection set are on workers.dev, so almost all of it rests on a voluntary runtime tag.
2. Churn is mostly measurement: 6,373 domains dropped a detection since March 2026, but 68.1% were seen exactly once, well above Cloudways, Supabase and Render.
3. Commerce, not engineering, is the visible base: Retail leads at 8.30% and Shopify runs on 26.55% of the domains we can profile.
4. No switching data: 72.3% of domains that ever ran both Workers and Amazon Web Services still run both, and 81.1% for Vercel. These are complements.
5. The size skew is a visibility artefact: 60.23% of sized domains have 1 to 10 employees and only 11 have 10,001 or more, because large deployments hide behind custom domains.
6. #12 at 0.79% is not a market position: it is a rank among platforms measured through very different windows.
8. Sales applications
Use this list for what it is. It is a high-precision list of teams that have shipped something on Workers and left the evidence in the page. It is a low-recall list of Workers users overall. That makes it good for prospecting and bad for market sizing.
The segment to work: Workers plus Shopify, roughly a quarter of the profiled base, together with the retail and travel hosts in the company table. These are teams already running custom logic at the edge in front of a platform they do not control.
What not to do: do not quote 0.79% as Cloudflare Workers' share of edge or serverless compute, do not treat a domain's absence from this list as evidence it does not run Workers, and do not build a churn campaign from the 6,373 dropped detections.
For deeper analysis, explore our full dataset of 8,757 matched Cloudflare Workers company profiles and 18,506 detected domains at TechnologyChecker.io.
Frequently Asked Questions
Who uses Cloudflare Workers?
We detect Cloudflare Workers on 18,506 domains, including Charlotte Center for Literary Arts, Inc., Virgin holidays, Keyence Corp.. A detection means the technology is present on that domain, which for large organisations is often a single storefront or subdomain rather than a company-wide deployment. Among domains we matched to a company with a resolved industry, the most common is Retail (8.3%).
How many customers does Cloudflare Workers have?
We detect 18,506 domains running Cloudflare Workers across our monthly crawl of live business domains. Company data could be matched for 8,757 of them (47.3%), and every company-size, industry and country breakdown on this page describes that matched subset rather than the full detected base. A further 6,373 domains carried a Cloudflare Workers detection previously and no longer do.
What is Cloudflare Workers's market share?
Cloudflare Workers holds 0.79% of the PaaS category, ranking #12. This is a share of the domains we detect running a platform in this category — not a share of all websites, and not a share of spending — so it is not directly comparable to figures published on other denominators.
What are the best alternatives to Cloudflare Workers?
The top alternatives to Cloudflare Workers include Amazon Web Services (39.98% market share), Vercel (16.64% market share), WordPress.com (8.24% market share), Azure (7.5% market share) — measured as a share of the domains we detect running a platform in this category.
Which countries use Cloudflare Workers the most?
United States leads with 3,824 matched Cloudflare Workers domains, followed by United Kingdom (661), Netherlands (385), Canada (377), India (324). These are counts within the 8,757 domains we could match to a company record, not the full detected base.
What size companies use Cloudflare Workers?
The most common company size is 1-10 employees, representing 60.23% of the Cloudflare Workers domains we could match to a company record and size — a subset of 8,757 matched companies, not of all 18,506 detected domains. This is followed by 11-50 employees (26.19%) and 51-200 employees (9.08%).
How old are companies that use Cloudflare Workers?
Among the Cloudflare Workers domains we could match to a company with a known founding date, the largest group was founded in the 2010s (40.75%), followed by the 2020s (25.93%). That matched subset is drawn from 8,757 companies and is not representative of all 18,506 detected domains. This suggests Cloudflare Workers is most popular among relatively young companies.
What is the ideal customer profile for Cloudflare Workers?
The ideal Cloudflare Workers customer is: Company Size: 1-10 employees, Location: US, UK, or Netherlands, Industry: Retail, Founded: 2010-2019, Company Age: 7-16 years old — based on our analysis of 8,757 enriched companies at TechnologyChecker.io.
How do you detect Cloudflare Workers on a website?
Two signals, and both are optional for a working deployment. The first is a runtime declaration in the page HTML, a meta tag with name "runtime" and content "Cloudflare Workers". The second is a request from the page to a .workers.dev address. If a site runs a Worker on a route of its own domain and does not emit that meta tag, we cannot see it from outside, because the response looks like any other origin served through Cloudflare.
Why does this count understate how many sites really run Cloudflare Workers?
Because both detection signals are things a deployment does incidentally rather than necessarily. The standard production pattern, a Worker bound to a route on your own domain, emits neither. The scale of the gap shows in our own data: across every domain where we currently detect Workers, exactly 11 are on a workers.dev hostname, fewer than one in a thousand. Nearly the whole count therefore comes from sites that voluntarily advertise the runtime, which makes 18,506 a floor. We publish it as a floor rather than extrapolating a total we cannot support.
Why is there no switching or migration data on this page?
Because Workers and the platforms it would be compared against are mostly complements. Workers is edge compute that sits in front of an origin instead of replacing it, so running Workers and AWS together is a normal architecture rather than a migration in progress. Of the domains that have ever run both Workers and Amazon Web Services, 72.3% still run both, and for Vercel the figure is 81.1%. A corridor built from the remainder would describe sites dropping an edge layer while keeping their origin, which is not a vendor switch, so we left the section out rather than publish a competitive story the data does not support.
Why do so many Cloudflare Workers detections disappear again?
6,373 domains dropped a detection since we began tracking in March 2026, but 68.1% of those rows were seen exactly once. The equivalent figure is 47.4% on Cloudways, 54.8% on Supabase and 55.9% on Render, so Workers churns far more intermittently than its neighbours. The cause is the detection surface: a runtime meta tag appears only on the routes a Worker renders, so which pages a crawl lands on decides whether we see it that month. Most of that movement is our visibility changing, not sites leaving Workers.
Does Cloudflare Workers usage really skew to small companies, or is that a measurement artefact?
Mostly an artefact. Only 11 of the domains we can size belong to companies of 10,001 or more employees, and 15 to the 5,001 to 10,000 band, against 4,967 in the 1 to 10 band. A large organisation running Workers correctly, bound to routes on its own domain, emits neither of our detection signals, so it never enters the table. The enterprise hits we do have are mostly single routes, such as flights.virginatlantic.com, resources.purolator.com and marketplace.k12.com, which is what surviving deployments look like rather than what enterprise adoption looks like.
Does a Cloudflare Workers detection mean the whole site runs on Workers?
No. In this data a detection is evidence of one route. Our company list is full of hosts like registrar.charlotte.edu, resources.purolator.com, marketplace.k12.com, academy.writer.com and help.xing.com rather than main corporate domains, because the routes most likely to declare a runtime are the standalone ones a single team built: a booking flow, an application portal, a docs site, a help centre. Read each row as one deployment, not as a company-wide platform decision.
What is a workers.dev subdomain?
workers.dev is the free hostname Cloudflare gives every Worker, in the form yourworker.youraccount.workers.dev, so you can call the code before wiring it to a domain. It is used mainly for development, previews and internal endpoints, and production sites normally move to a custom domain or a route. That is also why it is a weak detection signal: only 11 domains in our current Workers set are on workers.dev at all.
Does Cloudflare Workers have a free tier?
Yes. The Workers Free plan includes 100,000 requests per day at no cost, and the paid plan starts at $5 per month with 10 million requests included and additional requests billed at $0.30 per million (per developers.cloudflare.com/workers/platform/pricing). A free entry point is part of why the small-company skew on this page is so pronounced: 60.23% of the domains we can size have 1 to 10 employees.
What is the difference between Cloudflare Workers and AWS Lambda?
Workers runs code in V8 isolates on Cloudflare's network, close to the visitor, and avoids the cold start a container-based runtime pays. Lambda runs containers in an AWS region you choose and supports a wider set of languages including Python, Java, Go and .NET, where Workers is JavaScript, TypeScript and WebAssembly. In practice they are not always alternatives: 72.3% of the domains that have ever run both Workers and Amazon Web Services still run both.
What programming languages does Cloudflare Workers support?
JavaScript, TypeScript and WebAssembly natively, which means Rust, C and C++ can run once compiled to Wasm, and Python is supported through the same route. The runtime is V8 rather than Node.js, though Cloudflare has added Node.js compatibility over time (per developers.cloudflare.com/workers). The runtime itself was open sourced as workerd under Apache 2.0 in September 2022 (per blog.cloudflare.com/workerd-open-source-workers-runtime).
What is workerd?
workerd is the open source version of the runtime that powers Cloudflare Workers, released under Apache 2.0 in September 2022 (per blog.cloudflare.com/workerd-open-source-workers-runtime). It runs the same JavaScript and WebAssembly execution model locally or on your own infrastructure, which is mainly useful for local development, testing and self-hosting parts of a Workers application.
Can Cloudflare Workers replace a traditional backend?
For a lot of workloads, yes. Workers handles routing, authentication, API endpoints and server-side rendering, and Cloudflare supplies the surrounding pieces: KV for key-value storage, R2 for object storage, D1 for SQL and Durable Objects for stateful coordination. Long-running or CPU-heavy work is the usual limit. Our data suggests most visible deployments are not doing this, though: with Shopify present on 26.55% of the Workers domains we can profile, the common pattern is a Worker in front of an existing platform rather than instead of a backend.
What is Cloudflare Workers KV?
Workers KV is a globally distributed key-value store built for read-heavy access from Workers. Values are cached across Cloudflare's network so reads are fast wherever the Worker runs, at the cost of eventual consistency on writes. It suits configuration, feature flags, routing tables and cached content. For transactional or strongly consistent state, Durable Objects and D1 are the alternatives.
What is the difference between Cloudflare Workers and Cloudflare Pages?
Pages is for deploying sites, with a Git-connected build step, preview deployments and static asset hosting. Workers is the compute layer that runs code on incoming requests. The two have converged, since Pages runs its dynamic parts on Workers and Workers can now serve static assets directly (per developers.cloudflare.com/workers). Our detection sees the runtime, not the product name you deployed through.
Does Cloudflare Workers support WebSockets?
Yes, through Durable Objects, which give a WebSocket connection somewhere consistent to live. A Durable Object is a single addressable instance with its own storage, so it can coordinate many connected clients for chat, multiplayer, collaborative editing or live dashboards. A plain stateless Worker can accept a WebSocket but has nowhere to keep shared state.
What are the CPU time limits for Cloudflare Workers?
The free plan allows up to 10 milliseconds of CPU time per invocation and the paid plan raises that to 30 seconds by default (per developers.cloudflare.com/workers/platform/limits). The limit counts CPU time, not wall-clock time, so waiting on a network request does not count against it. That distinction is why edge routing, redirects and personalisation fit comfortably while heavy computation does not.
How many locations does Cloudflare Workers run in?
Workers runs across Cloudflare's global network, which the company puts at more than 330 cities in over 120 countries (per cloudflare.com/network). Code is deployed to the whole network rather than to a region you pick, and a request executes at whichever location receives it. That is the main architectural difference from region-based serverless platforms.
Is Cloudflare Workers the same as Cloudflare's CDN?
No, and the distinction matters when reading reviews, since most Cloudflare reviews cover the CDN and security products rather than Workers. The CDN caches and serves content. Workers runs your own code on the request path. They are separate products on the same network, which is why Cloudflare appears on 77.33% of the Workers domains we can profile: an overlap that is close to definitional rather than a choice these teams made.
What industries use Cloudflare Workers the most?
Retail leads at 8.30% (633 domains), then Real Estate at 4.92%, Software Development at 3.89% and Technology, Information and Internet at 3.26%. Retail Apparel and Fashion also makes the top ten at 2.24%. Retail ahead of software is the unexpected part, and the stack data agrees: Shopify runs on 26.55% of the Workers domains we can profile. These percentages describe the 8,757 domains we matched to a company record, not the full 18,506 detections.
Why do so many Cloudflare Workers domains also run Shopify?
Shopify appears on 26.55% of the Workers domains we can profile, which is the single most surprising row on this page. The likely reason is that merchants cannot change how a hosted storefront handles a request, so anything they need before the storefront responds has to live in front of it: redirects, geo and currency routing, A/B tests, bot rules and personalisation. That is a Worker's natural job, and it explains why Retail rather than Software Development tops the industry table.
What kinds of sites are the best fit for Cloudflare Workers?
Anything where work has to happen on the request path before an origin or platform responds: routing and redirects, authentication and token checks, A/B testing and personalisation, API gateways, image and content transformation, and server-side rendering close to the visitor. Our data reflects that. Storefronts, airline booking hosts such as flights.virginatlantic.com and flights.flyfrontier.com, and university portals such as apply.uopeople.edu are the recurring patterns.
Based on 8,757 company data
These insights include all TechnologyChecker.io detections of Cloudflare Workers (free & paid plans).