React Performance

When to Use

Apply when diagnosing slow renders, optimizing list rendering, or preventing unnecessary re-renders in React applications.

React Compiler (React 19+)

React Compiler automatically handles memoization for you. If your project uses React Compiler (React 19.2+ with compiler enabled):

• Manual memo, useMemo, and useCallback are often unnecessary
• The compiler automatically prevents unnecessary re-renders
• Profile first with React DevTools before adding manual optimizations
• See React Compiler docs for setup

When to still use manual optimization with React Compiler:

• Third-party libraries that require memoized props/callbacks
• Edge cases where compiler cannot optimize (check React DevTools)
• React 17-18 projects without compiler support

Without React Compiler (or React 17-18), use the patterns below.

Patterns

Pattern 1: React.memo for Pure Components

// Source: https://react.dev/reference/react/memo
import { memo } from 'react';

interface ItemProps {
  id: string;
  title: string;
  onClick: (id: string) => void;
}

const ListItem = memo(function ListItem({ id, title, onClick }: ItemProps) {
  return <li => onClick(id)}>{title}</li>;
});

// Only re-renders if props actually change

Pattern 2: useMemo for Expensive Calculations

// Source: https://react.dev/reference/react/useMemo
const filteredAndSorted = useMemo(() => {
  return items
    .filter(item => item.status === 'active')
    .sort((a, b) => b.priority - a.priority);
}, [items]); // Only recalculate when items change

Pattern 3: useCallback for Stable Handlers

// Source: https://react.dev/reference/react/useCallback
const handleDelete = useCallback((id: string) => {
  setItems(prev => prev.filter(item => item.id !== id));
}, []); // Stable reference, safe for memo'd children

Pattern 4: Virtualization for Long Lists

// Source: https://tanstack.com/virtual/latest
import { useVirtualizer } from '@tanstack/react-virtual';

function VirtualList({ items }: { items: Item[] }) {
  const parentRef = useRef<HTMLDivElement>(null);
  const virtualizer = useVirtualizer({
    count: items.length,
    getScrollElement: () => parentRef.current,
    estimateSize: () => 50,
  });

  return (
    <div ref={parentRef} style={{ height: 400, overflow: 'auto' }}>
      <div style={{ height: virtualizer.getTotalSize() }}>
        {virtualizer.getVirtualItems().map(row => (
          <div key={row.key} style={{ transform: `translateY(${row.start}px)` }}>
            {items[row.index].name}
          </div>
        ))}
      </div>
    </div>
  );
}

Pattern 5: Lazy Loading Components

// Source: https://react.dev/reference/react/lazy
import { lazy, Suspense } from 'react';

const HeavyChart = lazy(() => import('./HeavyChart'));

function Dashboard() {
  return (
    <Suspense fallback={<Spinner />}>
      <HeavyChart data={data} />
    </Suspense>
  );
}

Anti-Patterns

• Premature optimization - Measure first with React DevTools Profiler
• memo everything - Only memo components that receive same props often
• useMemo for simple values - Overhead > benefit for trivial calculations
• Inline objects/arrays in JSX - Creates new reference every render
• Manual memo with React Compiler - Redundant if compiler is enabled

Verification Checklist

• Profiled with React DevTools before optimizing
• Checked if React Compiler is enabled in project
• memo'd components actually receive stable props
• Lists with 100+ items use virtualization
• Heavy components lazy loaded
• No inline object/array props to memo'd children