Every Twin-Ex discovery call with an Israeli SMB starts the same way: "we have a WordPress site, it is slow, the SEO is mediocre, and our developer disappeared two years ago." WordPress is not the villain in that story — it is the inertia. The villain is the plugin sprawl, the unmaintained themes, and the hosting bill that grows with every visitor. The question is when to stop fixing and when to rebuild.
What WordPress is still good for.
WordPress is excellent when content updates are the daily activity. A magazine, a newsroom, a recipe blog, a wedding photographer with 200 galleries — WordPress lets a non-technical owner publish without thinking about deploys, builds, or markdown. The Block Editor is mature. The plugin ecosystem covers payments, bookings, and translations. If your team writes 5+ posts a month and the spend is under ₪400 a month, do not switch.
When WordPress is the right call.
Honest balance: there are four cases where we tell a client to stay on WordPress. One — a content-heavy blog or news site where new posts are the core product. Two — a non-technical owner who must self-edit copy, prices, and pages every single day without waiting on a developer. Three — a genuinely tight budget where any custom build, even one from ₪4,500, is out of reach this quarter. Four — heavy dependence on a specific plugin ecosystem (a particular booking, membership, or LMS plugin) that would cost more to rebuild than it saves. If two of those describe you, WordPress is the right call. Recommending it anyway, when it fits, is how we keep trust.
Where it breaks down.
WordPress breaks down on three things at once: Core Web Vitals, security surface, and total cost of ownership. The average WordPress site scores 40-60 on Lighthouse mobile because the rendering path is server-side PHP with 8 active plugins each adding 200ms — and only about 44% of WordPress sites pass Core Web Vitals on mobile at all. Security: it is the most-attacked CMS on the internet, the patch cadence is monthly minimum, and the attack surface is every plugin you installed. Cost: managed hosting that performs respectably (Kinsta, WP Engine) starts at $35/month and goes up linearly with traffic. By month 18 you have paid more in hosting than a Next.js rebuild would have cost.
Head to head: the numbers.
- Speed — WordPress: ~3-5s mobile LCP on a typical plugin-loaded site · Next.js: ~1.5-2.5s, edge-rendered.
- Core Web Vitals — WordPress: only ~44% of sites pass on mobile · Next.js: passes by default with SSG/ISR.
- Monthly cost — WordPress: ~$50-200 (managed hosting + plugin licenses + update labour) · Next.js on Vercel: ~$0-20.
- Maintenance — WordPress: constant plugin, theme, and security updates · Next.js: near-zero, no server to patch.
- Editing — WordPress: non-technical, the owner edits anytime in the Block Editor · Next.js: needs a developer or a CMS layer bolted on.
- Security — WordPress: the most-attacked CMS, surface grows with every plugin · Next.js: static output, almost nothing to exploit.
Where Next.js wins.
Next.js wins when the site is a product surface for the business, not a content firehose. A marketing site, a service catalog, a pricing page, a booking funnel, a SaaS landing page — these benefit from Static Site Generation (SSG) + Incremental Static Regeneration (ISR). The pages render at the edge in under 100ms. SEO is solved by the framework (canonical URLs, structured data, sitemap, RSS, hreflang). Hosting at Vercel is free up to 100GB/month bandwidth and $20/month after — significantly cheaper than managed WordPress at the same traffic.
The hybrid that nobody talks about.
You can have both. Run your marketing site on Next.js for speed, SEO, and a low hosting bill. Run a headless WordPress as the CMS behind the blog, exposing posts via the REST or GraphQL API. The editorial team keeps the Block Editor they know. The marketing site stays at 95+ Lighthouse. This is what we do for clients who genuinely publish weekly content but also need a fast funnel — and it costs the same as just running Next.js, because the headless WordPress instance can run on a $5/month Hetzner box.
Practical rebuild signals.
Three signals tell you it is time to leave WordPress: (1) your mobile Lighthouse score is below 50 and a developer has tried twice to fix it; (2) you have been hacked once, or you got the "your version is outdated" email from your host more than three times this year; (3) your monthly hosting + plugin license bill is over ₪400 and the site is still not fast. Any one of those, on a marketing-first site, justifies the rebuild. Two or more, definitely.
What a Next.js rebuild actually looks like.
For a typical Israeli SMB with a 12-page marketing site, the rebuild is 3-4 weeks. Week 1: discovery, sitemap, content audit, picking what stays and what gets dropped. Weeks 2-3: design and build, Hebrew + English from day one. Week 4: SEO setup, GA4, sitemap, and redirects from the old WordPress URLs to preserve the rankings you already have. We charge from ₪4,500 for a business site at this scope, including the redirects map (the part most agencies skip). After the rebuild the hosting bill is $0-$20 a month and the site is fast on every device. The upside our clients report most often: a 20-40% lift in organic traffic within about 90 days of migrating off a slow WordPress site — speed is a ranking factor, and Google notices.
WordPress in 2026 is a content tool, not a business stack. If your site is your product, build it on the product stack.