The Indie Hacker Tech Stack in 2026: What 2,253 Live Products Run On
We read the public pages of 2,253 live indie products: 41.9% run Next.js, 53% use Tailwind CSS, and 19% run the exact Next.js, Tailwind and Vercel stack.
Founder of Awesome Indie11 min read
Key takeaways
- Next.js runs 41.9% of the 2,253 live indie products we could read, against 3.8% of all websites in W3Techs' November 2026 count.
- Tailwind CSS is on 53.0% of them. Bootstrap, which W3Techs finds on 12.9% of all websites, is on 3.6%.
- One exact stack, Next.js plus Tailwind CSS plus Vercel, accounts for 432 products: 19.2% of everything we could read.
- 52.5% of the live products we read are served through Cloudflare, against 29.2% of all websites; 9.1% run WordPress, against 39.2% of the web.
- WordPress fell from 10.4% of products submitted in 2024 to 6.6% of those submitted in 2026; Webflow and Framer together fell from 6.3% to 2.7%.
The indie web is not a smaller copy of the web. It is a different web. We read the public pages of 2,253 live products on Awesome Indie and found a stack that barely resembles what the internet as a whole runs on: Next.js on 41.9% of them, Tailwind CSS on 53.0%, and Cloudflare in front of 52.5%. WordPress, which W3Techs finds on 39.2% of all websites, runs 9.1% of indie launches.
The default is narrow enough to name. Next.js, Tailwind CSS and Vercel, together, account for 432 products, or 19.2% of everything we could read. The runner-up, the same pair behind Cloudflare instead of on Vercel, has 252. And among products submitted in 2026, the default is getting stronger while WordPress, Webflow, Framer, jQuery and Bootstrap all lose share.
One caution: we only see what a site's public page reveals to a plain request (headers, cookies, script tags, markup, DNS records). That names a framework, a host or an analytics tool, almost never a database or a backend language. Where the blind spot bends a number, we say so.
What is the most common indie hacker tech stack in 2026?
Next.js, by a distance. 943 of the 2,253 products (41.9%) serve markup that only Next.js emits. WordPress is a distant second at 9.1%, followed by Astro at 4.5% and Nuxt at 3.7%. Webflow (3.0%) and Framer (2.1%) are the only no-code builders above 1%.
- Next.js41.9%
- No framework we can identify29.4%
- WordPress9.1%
- Astro4.5%
- Nuxt3.7%
- Webflow3%
- Laravel2.3%
- Framer2.1%
- Ruby on Rails1.3%
- SvelteKit0.8%
- Remix0.6%
2,253 live products whose stack we could read, November 13, 2026. Direct detections only. A product can appear in more than one row.
Show the numbers
| Next.js | 41.9% |
|---|---|
| No framework we can identify | 29.4% |
| WordPress | 9.1% |
| Astro | 4.5% |
| Nuxt | 3.7% |
| Webflow | 3% |
| Laravel | 2.3% |
| Framer | 2.1% |
| Ruby on Rails | 1.3% |
| SvelteKit | 0.8% |
| Remix | 0.6% |
Two things about that chart. First, we count frameworks only where we saw them directly. React shows up on 991 products (44.0%) in our data, but 987 of those are entailed by something else: a Next.js site is a React site, and so is one built on Radix UI or Material UI. Our check cannot see React on its own, so the React figure is a floor, and the same is true of JavaScript, of PHP under WordPress and of Ruby under Rails.
Second, 663 products (29.4%) show no framework we can name. Their served HTML carries no fingerprint: hand-written pages, client-rendered apps that ship an empty shell plus a script bundle, and builders we only see through a brand name in the page, which we do not count.
The median product reveals four technologies. The typical one is a Next.js app styled with Tailwind, hosted on Vercel or behind Cloudflare, loading Google Fonts (24.1% do). Next.js with Tailwind covers 800 products (35.5%); add Vercel and it is still 432 (19.2%).
How does the indie stack compare with the rest of the web?
It looks like a developer survey, not like the web.
| Technology | Live indie products (ours) | All websites (W3Techs) | Developers (Stack Overflow 2025) |
|---|---|---|---|
| Next.js | 41.9% | 3.8% | 20.8% |
| React | 44.0% (mostly entailed) | 6.1% | 44.7% |
| jQuery | 11.3% | 63.9% | 23.4% |
| WordPress | 9.1% | 39.2% | 13.6% |
| Bootstrap (CSS) | 3.6% | 12.9% | not asked |
| Webflow | 3.0% | 0.8% | not asked |
| Framer | 2.1% | 0.2% | not asked |
| Cloudflare | 52.5% | 29.2% | 20.1% |
| Google Analytics | 19.7% | 46.7% | not asked |
W3Techs figures come from its pages on content management systems, JavaScript libraries, CSS frameworks, Cloudflare as a reverse proxy and Google Analytics, read November 13, 2026, over a sample of well over 20 million sites. The Stack Overflow 2025 Developer Survey (July 2025) counts respondents who did extensive work with each technology in the past year. Methods differ, so read the columns as orders of magnitude.
We think the gap with the wider web is mostly about age. W3Techs measures the web as it stands, including every site built a decade ago and never rebuilt. Indie launches are new by definition, and our own data says most are built by people who write code. That is why our numbers sit closer to the developer survey: React is used by 44.7% of Stack Overflow respondents and is entailed on 44.0% of our products, while jQuery, still on 63.9% of the web, is on 11.3% here, and on WordPress sites for 131 of those 255.
Indie makers go further on one choice: Next.js runs 41.9% of products but is used by 20.8% of developers, whose answers cover every kind of work, back ends and internal tools included. Our sample is one kind of thing: a public product website.
The no-code numbers tell a quieter story. Webflow's share of indie products is almost four times its share of the web, and Framer's ten times. Wix (0.4%), Shopify (0.3%) and Squarespace (0.1%), which hold 4.2%, 5.4% and 2.4% of the web, barely register. Indie makers who use a builder pick Webflow or Framer over Wix, Shopify or Squarespace.
Why does Next.js come with Vercel?
Because Vercel maintains Next.js, and the pairing shows. 52.3% of Next.js products (493 of 943) answer with Vercel's headers, and 80.3% of Vercel-hosted products (493 of 614) run Next.js.
It is a default, not a requirement. 27.9% of Next.js products sit behind Cloudflare with no other host visible, and the rest spread across AWS, other platforms and plain servers, or show no host at all. Other frameworks lean the other way: only 9.9% of Astro sites are on Vercel, against 53.5% behind Cloudflare alone, and Nuxt sites are 62.7% Cloudflare only and 18.1% Vercel.
- Cloudflare, origin not visible40.4%
- Vercel27.3%
- Own web server (nginx, Apache, LiteSpeed)10.4%
- Not identifiable6.4%
- Hosted site builder5.5%
- Other clouds and platforms4.2%
- AWS3.4%
- Netlify2.4%
2,253 live products, November 13, 2026. One bucket per product, in this order of priority: named platform first, then Cloudflare, then the web server. Other platforms: Google Cloud, Railway, Render, DigitalOcean, GitHub Pages, Fly.io, Azure.
Show the numbers
| Cloudflare, origin not visible | 40.4% |
|---|---|
| Vercel | 27.3% |
| Own web server (nginx, Apache, LiteSpeed) | 10.4% |
| Not identifiable | 6.4% |
| Hosted site builder | 5.5% |
| Other clouds and platforms | 4.2% |
| AWS | 3.4% |
| Netlify | 2.4% |
The biggest bucket needs explaining. A site behind Cloudflare's proxy usually answers with Cloudflare's headers and nothing from the machine behind it, so this bucket mixes sites hosted on Cloudflare itself with sites on a VPS or a platform behind the proxy. Counting every product that answers through Cloudflare, Vercel sites behind it included, gives 52.5%, against 29.2% of all websites.
The Vercel and Netlify gap is wider here than among developers. Stack Overflow's 2025 survey has Vercel at 10.6% and Netlify at 5.9% of respondents; among indie products it is 27.3% against 2.4%, more than eleven to one.
Is Tailwind CSS the default for indie products?
Yes. Tailwind CSS is on 1,193 live products (53.0%), and on 84.8% of Next.js products, 84.3% of Nuxt products and 75.2% of Astro products. Bootstrap is on 80 (3.6%), and only 5 products use both.
Indie makers are further from Bootstrap than even CSS-minded developers. In the State of CSS 2025 survey, 2,041 of the 3,977 people who answered the frameworks question used Tailwind (51%) and 1,194 used Bootstrap (30%). On the web at large it is the reverse: W3Techs finds Bootstrap on 12.9% of all websites.
The component layer follows. Radix UI's markup is on 244 products (10.8%), 243 of them with Tailwind: the signature of copy-in component kits such as shadcn/ui, which is styled with Tailwind and offers Radix UI versions of its components. Radix went from 9.2% of 2024 submissions to 13.5% of 2026 ones.
Our Tailwind figure is a floor: we recognize it by utility class names in the served HTML (hover:bg-, md:flex), which a client-rendered app's empty shell does not contain.
Do indie makers use Google Analytics or privacy-first analytics?
Google Analytics, still, by about two to one where we can see fairly. But this is where our blind spot is widest, so the headline number needs care.
Across all 2,253 products we see an analytics script on 689 (30.6%): Google Analytics on 19.7%, and privacy-focused tools combined on 14.1%. Those figures are low partly because Next.js's Script component, by default, injects third-party scripts in the browser after hydration, so a plain request never sees them. We find Google Analytics on 3.9% of Next.js products and on 49.8% of WordPress ones, and we would not read that as a choice. The fairer comparison is the 1,310 products not built with Next.js:
- Google Analytics31.1%
- Any privacy-focused tool16%
- Cloudflare Web Analytics9.1%
- Ahrefs Web Analytics3.2%
- Plausible3.1%
- Umami0.8%
- Fathom0.4%
- Simple Analytics0.4%
- Pirsch0.3%
1,310 live products, November 13, 2026. Scripts loaded by a plain script tag only; anything injected by a tag manager or client-side code is invisible.
Show the numbers
| Google Analytics | 31.1% |
|---|---|
| Any privacy-focused tool | 16% |
| Cloudflare Web Analytics | 9.1% |
| Ahrefs Web Analytics | 3.2% |
| Plausible | 3.1% |
| Umami | 0.8% |
| Fathom | 0.4% |
| Simple Analytics | 0.4% |
| Pirsch | 0.3% |
Even there, Google Analytics is ahead, though well below the 46.7% of all websites W3Techs counts.
Cloudflare Web Analytics leads the privacy-focused group because it rides on a default. It is free and uses no cookies, and on a site already behind Cloudflare, Cloudflare injects the script itself once it is switched on. The surprise is Ahrefs Web Analytics, free for verified sites and cookieless by Ahrefs' description: it is already on more live products (58) than Plausible (49), which also sets no cookies and starts at $9 a month. Umami, open source and cookieless, is on 14.
The privacy-first wave is not growing in our data. Among products not built with Next.js, Google Analytics sits at 31.2% of 2024 submissions and 31.6% of 2026 ones, while privacy-focused tools went from 18.4% to 14.6%. The samples are about 560 products each, so treat that as flat at best.
What do indie products use for email, auth and payments?
Every number in this section is a floor.
Sending email. Read from the TXT records at the root of each domain (in practice, its SPF record), Amazon SES is authorized on 146 products (6.5%), Mailgun on 66 (2.9%) and SendGrid on 61 (2.7%). A sender set up only on a subdomain is invisible to us, and our rule for Resend has not matched a single live product, which says more about where we look than about Resend.
Receiving email. Google's mail servers handle the inbox for 35.2% of products (793), Zoho for 6.9% and Microsoft 365 for 4.0%.
Email authentication. 1,513 products (67.2%) publish a DMARC record, but only 624 (27.7%) set it to quarantine or reject, and 740 (32.8%) have none at all. Since February 1, 2024, Gmail has required DMARC from anyone sending it more than 5,000 messages a day, with a policy of none accepted.
Auth and payments. Clerk shows up on 32 products (1.4%), every one of them a Next.js app. A payment provider's script shows up on 21 (0.9%), most likely because checkout code loads on pricing and checkout pages, not on the page we read. We can only spot Supabase by its name in a page, which we treat as a guess and do not count. Read none of these as absence.
How has the indie stack changed between 2024 and 2026?
The default is consolidating, and the older tools are shrinking.
| Technology | Submitted in 2024 (931 products) | Submitted in 2026 (1,032 products) |
|---|---|---|
| Tailwind CSS | 50.4% | 58.6% |
| Next.js | 39.4% | 45.7% |
| Vercel | 25.7% | 29.5% |
| Radix UI | 9.2% | 13.5% |
| Astro | 3.9% | 5.1% |
| Cloudflare | 52.7% | 53.1% |
| WordPress | 10.4% | 6.6% |
| jQuery | 14.0% | 6.4% |
| Webflow or Framer | 6.3% | 2.7% |
| Bootstrap | 3.9% | 2.2% |
Tailwind gained 8.2 points and Next.js 6.3. WordPress lost 3.8 points, while jQuery and the two no-code builders more than halved. Cloudflare did not move. We leave 2025 out: only 189 of the 2,085 live products submitted that year are in our sample so far, too few to compare.
Two limits apply. These are today's stacks, not launch-day ones: a 2024 product that has since moved to Next.js counts as Next.js. And products whose sites are gone drop out, so older cohorts are survivors.
What this means if you are choosing a stack this week
Take the default unless you have a reason not to. Next.js, Tailwind and Vercel is what 432 live indie products run, 19.2% of the sample down to the host. This data cannot tell you which stack wins, only which one is common, and we read that as a practical argument: the more common the stack, the more examples, answers and people who already know it.
Treat hosting as the real decision. Vercel is the Next.js default, and Cloudflare is everyone else's: Astro and Nuxt makers mostly put their sites behind it. Among products not built with Next.js, only 9.2% are on Vercel.
Check that your analytics actually loads. On a Next.js site, scripts added with the Script component arrive after hydration by default, so view-source will not show them. Open your browser's network tab on the live site and confirm the request goes out.
Publish a DMARC record today. A third of live products (32.8%) have none. A p=none record costs nothing, covers the DMARC part of Gmail's bulk-sender rules, and is the step before enforcement.
See what your site gives away. Run your domain, and a competitor's, through the tech stack checker, then browse built with for products on the same stack as yours. Every product launched on Awesome Indie gets the same read on its page, and makers can remove a detection that is wrong.
How we got these numbers
Population. Products live (LISTED) on Awesome Indie on November 13, 2026: 5,191. Of these, 2,253 had at least one confirmed technology detection: the denominator unless a section names a subset. They skew to products submitted in 2024 (931) and 2026 (1,032), because that is where our site check's coverage is; 2025 submissions are thin (189). As a check, we reran everything with the products that passed review and are waiting for their launch day added in: Next.js moved to 42.1%, Tailwind to 52.3%, Vercel to 26.3%, Cloudflare to 54.5% and WordPress to 7.4%. No headline changed by more than two points.
Collection. A monthly site check makes one plain HTTP request to each product's website and looks up its DNS records. A technology counts when it leaves a decisive trace: a response header, a cookie, a script's source address, a generator tag, a structural marker only it emits, or a DNS TXT record. A bare brand name in the page is a low-confidence guess the site never shows, and is excluded here, as are detections a maker removed (there were none). Entailed rows, like React under Next.js, are flagged where we use them; frameworks are counted by direct detection. The counts match the built with pages, which apply the same rules.
Caveats. We see only the outer layer of a stack: databases, ORMs, backend languages, AI providers and CI appear only when a framework that requires them does. Scripts injected by tag managers or client-side code are invisible, which undercounts analytics, especially on Next.js. We read one page per product, so auth and payments, which live deeper, are undercounted. Products submitted to Awesome Indie are a self-selected group. W3Techs and the developer surveys use different methods and populations, so comparisons are directional.
Sources
- W3Techs, "Usage Statistics and Market Share of Content Management Systems, November 2026" (November 2026): https://w3techs.com/technologies/overview/content_management
- W3Techs, "Usage Statistics and Market Share of JavaScript Libraries for Websites, November 2026" (November 2026): https://w3techs.com/technologies/overview/javascript_library
- W3Techs, "Usage Statistics and Market Share of CSS Frameworks for Websites, November 2026" (November 2026): https://w3techs.com/technologies/overview/css_framework
- W3Techs, "Usage Statistics and Market Share of Cloudflare, November 2026" (November 2026): https://w3techs.com/technologies/details/cn-cloudflare
- W3Techs, "Usage Statistics and Market Share of Google Analytics for Websites, November 2026" (November 2026): https://w3techs.com/technologies/details/ta-googleanalytics
- W3Techs, "Web Technologies Statistics and Trends" (November 2026): https://w3techs.com/technologies
- Stack Overflow, "Technology | 2025 Stack Overflow Developer Survey" (July 2025): https://survey.stackoverflow.co/2025/technology
- Devographics, "State of CSS 2025: Other Tools" (August 2025): https://2025.stateofcss.com/en-US/other-tools/
- Next.js, "Next.js by Vercel - The React Framework" (November 2026): https://nextjs.org/
- Next.js, "Script Component" (August 2026): https://nextjs.org/docs/app/api-reference/components/script
- shadcn/ui, "Manual Installation" (November 2026): https://ui.shadcn.com/docs/installation/manual
- shadcn/ui, "Dialog" (November 2026): https://ui.shadcn.com/docs/components/radix/dialog
- Cloudflare, "Free, privacy-first analytics for a better web" (September 2020): https://blog.cloudflare.com/free-privacy-first-analytics-for-a-better-web/
- Cloudflare, "Enabling Cloudflare Web Analytics" (April 2026): https://developers.cloudflare.com/web-analytics/get-started/
- Ahrefs, "Ahrefs Web Analytics" (November 2026): https://ahrefs.com/web-analytics
- Plausible Analytics, "Simple, privacy-friendly Google Analytics alternative" (November 2026): https://plausible.io/
- Umami, "Open-Source Web and Product Analytics" (November 2026): https://umami.is/
- Google, "Email sender guidelines" (November 2026): https://support.google.com/a/answer/81126
Written by
Diogo CapelaDiogo builds Awesome Indie, a launch directory for indie makers that has listed products since 2020. He writes about what the data from thousands of launches says about building small software businesses.