Frontend System Design · Lesson 33 of 86 · 4 min
SSR vs CSR: Rendering Trade-offs
In an interview, you'll be asked to choose a rendering strategy. This is how you justify it.
How to Frame the Core Trade-off
The choice between Client-Side Rendering (CSR) and Server-Side Rendering (SSR) is about who does the work of turning data and components into HTML, and when.
In CSR, the server sends a nearly empty HTML file and a JavaScript bundle. The browser runs the JS, fetches data, and renders the page. The user sees a spinner.
In SSR, the server fetches the data and renders the full HTML for the page before sending it. The browser can display it immediately. The user sees content.
This is the fundamental trade-off: a slow first paint (CSR) versus a potentially slow first interaction (SSR). Every other consideration—SEO, server cost, code complexity—stems from this.
When to Justify SSR
Choose SSR when your primary goals are a fast initial paint and strong SEO. This is typical for content-driven sites: e-commerce product pages, blogs, news articles, and marketing landing pages.
The key benefit is a fast First Contentful Paint (FCP) and Largest Contentful Paint (LCP). The user sees meaningful content quickly because the browser receives a complete HTML document. Search engine crawlers also see this complete document, which is ideal for indexing.
The interview angle: When you propose SSR, immediately mention the metrics you aim to improve: LCP and FCP. State that for a public-facing page like a product description, getting content in front of the user and Google's crawler is the top priority.
This is the typical flow for an SSR request. Notice the work done on the server before any HTML is sent.
The trade-off is server load and complexity. Your server is now a Node.js process running React, not just a static file server. It uses more CPU and memory, increasing hosting costs.
The Hidden Cost of SSR: Hydration
The trap: Many candidates praise SSR for performance but ignore its biggest pitfall: Time to Interactive (TTI).
A user sees a fully rendered SSR page, tries to click a button, and nothing happens. The page looks ready, but the JavaScript that attaches event listeners hasn't finished downloading, parsing, and executing. This process is called hydration.
A large JS bundle can make the TTI significantly later than the FCP, creating a frustrating "rage click" scenario. You've optimized for the perception of speed but failed on the reality of usability.
When you recommend SSR, you must also propose a strategy to manage hydration cost. This shows you understand the full picture. Mention techniques like code-splitting your JS bundles, or using frameworks like Astro that promote partial hydration (shipping zero JS by default for static components).
When to Justify CSR
Choose CSR for applications that are highly interactive and don't rely on public SEO for discovery. Think admin dashboards, internal tools, or complex, stateful UIs like a design tool or a social media feed behind a login.
After the initial load, a CSR application (a Single-Page App or SPA) feels fast. Navigating between "pages" is just a client-side route change, often involving a quick API call for data and re-rendering a small part of the UI. There are no full-page reloads. This provides a fluid, app-like experience.
Development is also simpler. You have a clear separation: a static frontend application that consumes a backend API. This decouples your teams and your deployments. Server costs are lower because the server's main job is just serving static assets and JSON.
The trap: Don't defend CSR by saying "Google can crawl JavaScript now." While true, it's not guaranteed, it takes longer, and it's less reliable than providing plain HTML. If SEO is a business requirement, CSR is the wrong default choice for public-facing content.
What the Browser Receives
The difference is clearest when you see the initial HTML payload. An SSR server does the work of fetching data and building the HTML before sending a response.
Here is a conceptual Express server doing just that. The browser gets a fully-formed document.
// server.ts (conceptual SSR with Express)
import express from 'express';
import axios from 'axios';
const app = express();
app.get('/', async (req, res) => {
// 1. Fetch data on the server
const response = await axios.get('https://.../posts?_limit=3');
const posts = response.data;
// 2. Render HTML string on the server
const postsHtml = posts.map(post => `
<div>
<h3>${post.title}</h3>
<p>${post.body}</p>
</div>
`).join('');
const fullPageHtml = `
<!DOCTYPE html>
<html>
<head><title>SSR Example</title></head>
<body>
<h1>Latest Posts</h1>
<div id="root">${postsHtml}</div>
<script src="/client.js"></script> <!-- For hydration -->
</body>
</html>
`;
// 3. Send the complete HTML document
res.send(fullPageHtml);
});
app.listen(3000);
In a CSR world, the div#root would be empty. The browser would have to run client.js to fetch the posts and generate the same markup.
Try it
You are designing a product detail page for an e-commerce site. In a document, outline which rendering strategy you'd choose. List the top 3 business and performance metrics you'd use to justify your decision and explain why each is important.
Next: We'll dig deeper into the concept of making server-rendered pages interactive in Hydration and Time to First Byte (TTFB).
Lock it in — apply this lesson under interview conditions.