Why should you care?

ShopNest flash sale ends — marking 10,000 products inactive row-by-row with SaveChanges would freeze the API; bulk SQL finishes in seconds.

See it live — copy this example

Paste into a .NET project with EF Core packages, then run with LocalDB/SQL Server (dotnet ef / dotnet run).

await _context.Products
    .Where(p => p.CategoryId == clearanceCategoryId)
    .ExecuteUpdateAsync(s => s
        .SetProperty(p => p.IsPublished, false)
        .SetProperty(p => p.UpdatedAt, DateTime.UtcNow));

await _context.CartItems
    .Where(c => c.ExpiresAt < DateTime.UtcNow)
    .ExecuteDeleteAsync();

What happened?

• ExecuteUpdateAsync generates UPDATE … WHERE without SELECT into memory.
• ExecuteDeleteAsync issues DELETE for matching rows.
• Neither uses change tracking.

Practice next

1. Install EF Core 7+ and test ExecuteUpdate on LocalDB with ten rows.
2. Compare SQL log row count vs loading entities in a loop.
3. Run bulk update inside a transaction when combined with other work.
4. Chain SetProperty for Price discount: SetProperty(p => p.Price, p => p.Price * 0.9m).
5. Wrap ExecuteDeleteAsync in BeginTransactionAsync for rollback testing.

Remember

ExecuteUpdate/ExecuteDelete avoid per-row tracking. Ideal for mass status or cleanup jobs. Always include a precise WHERE clause.

ShopNest clearance batch

End-of-season job unpublishes all clearance category products with one ExecuteUpdateAsync.

Outcome: Admin dashboard completes in under a second instead of timing out.