Why We Build Every Marketing Site on Next.js — and When We Don't

Our default stack for marketing sites in 2026 is Next.js 15 with the App Router, Tailwind, and MDX for blog content — deployed on Vercel or Cloudflare Pages. This isn't a fashion statement. It's the result of dozens of builds across clients ranging from seed-stage SaaS to enterprise brands, and it's the stack that consistently hits sub-1-second LCP with room to add rich interactions without falling apart.

But Next.js isn't the right answer for every business. This piece walks through when we pick it, the specific cases where Webflow or Shopify wins, and the stack we actively migrate clients off of.

What Next.js unlocks in 2026

The App Router, server components, and streaming rendering combine to produce something no other marketing stack can match: dynamic-feeling, fast pages without client-side JavaScript bloat. Server components render on the server or at build time and ship zero JS to the client for sections that don't need interactivity. Client components hydrate selectively. The payoff is Core Web Vitals scores that stay green as sites grow — a 300-page marketing site on Next.js can hit LCP under 1 second on every page, INP under 100ms, CLS under 0.05. Those numbers are basically unattainable on WordPress or most themed Shopify setups.

The second unlock is the component model. Marketing, app, and commerce surfaces share a design system, which means a button change in one place updates everywhere. For growth-stage companies with overlapping marketing and product surfaces, this shortens design-to-ship cycles from weeks to days. For mature companies, it means the marketing team can ship new landing pages without engineering intervention because the component library is already proven.

Third: edge rendering. Vercel and Cloudflare both support running rendering at CDN edge, which means the first byte of any page reaches a visitor in Delhi or Dubai or Dublin with roughly the same latency a North American user sees. For global brands — and every brand is global now — this is a 200–400ms speed advantage that compounds across every interaction.

When Webflow is the better call

Webflow is the right default for three specific situations. First: marketing teams with no dedicated engineering resource who need a visually ambitious site they can edit without filing tickets. Webflow's visual CMS is genuinely best-in-class; marketers can ship new pages and modules in a day without touching code. For brands where the marketing site is a business-critical surface that changes weekly, that velocity is worth the performance and flexibility ceiling Webflow imposes.

Second: brand-forward sites where animation and visual design are the product. Webflow's interactions system, combined with its tight Figma integration, makes shipping genuinely striking design fast. The ceiling is lower than Next.js with Framer Motion, but for most brand marketing sites the ceiling is still well above what matters.

Third: sites with lifespans under 24 months — campaign sites, event sites, launch sites. The Next.js setup cost doesn't amortize. Webflow is faster to ship and the performance penalty is temporary. When the site sunsets, so does the tradeoff.

When Shopify still wins for commerce

For DTC brands under roughly $30M revenue, we recommend Shopify (or Shopify Plus) for the storefront and use Next.js only for the marketing layer around it. The reason is pragmatic: Shopify's ecosystem of apps, its checkout conversion optimization, its built-in inventory and fulfillment integrations, and its built-in merchandising tools add up to six-to-twelve-figure advantages over a custom build. You can replicate those features on Next.js with headless Shopify, but the cost to build and maintain parity is usually higher than the revenue gain justifies until you're well into the eight-figure range.

The tipping point we've seen: past about $30M in annual commerce revenue, or when the brand has customization needs Shopify can't support (advanced subscription logic, complex B2B ordering, multi-brand catalogs, novel checkout flows), headless with Next.js starts to pay off. Before that, stay on Shopify's native stack. The complexity discount is worth far more than engineering team credibility.

What to avoid — and why we migrate clients off of it

We do not build new WordPress sites in 2026, and we regularly migrate clients off WordPress to Next.js or Webflow. The reasons are not ideological; they're measurable. WordPress's plugin ecosystem is a maintenance tax — every plugin is a potential security vulnerability, a performance regression, and a future compatibility issue. Site security breaches on WordPress are dramatically more common than on Next.js or Webflow because the attack surface is enormous. Core Web Vitals scores are structurally worse because the stack wasn't designed around modern rendering.

The common objection is 'we have content editors who know WordPress.' Fair, but content editing in a modern headless CMS (Sanity, Contentful, Payload) takes a half-day to learn and delivers real benefits the editors will feel within a week. Migration projects typically take 6–10 weeks for a mid-sized site, cost less than a year of plugin maintenance fees, and produce sites that run 2–4x faster with roughly zero ongoing security surface.

The other stack we avoid for net-new marketing sites: drag-and-drop page builders glued onto themes (Elementor, Divi, anything that ends in 'Builder'). The performance and maintainability costs compound, and the design ceiling is miles below what a modern front-end stack can achieve.

Key takeaways

• Next.js is our default in 2026 for sub-1s LCP, clean component reuse, and edge rendering benefits global brands feel immediately.
• Pick Webflow for marketing teams without engineering support, brand-forward design, or sites with under 24-month lifespans.
• Stay on native Shopify for DTC under roughly $30M revenue. Go headless with Next.js only when customization needs cross that threshold.
• Avoid new WordPress builds. The plugin tax, security surface, and performance ceiling aren't worth the apparent editing convenience.