Building a Slug-and-Redirect System for Renamed Content in Laravel

Rename a blog post's title, and its slug usually changes with it. Every inbound link, bookmark, and search engine result pointing at the old URL now 404s — which is bad for users and worse for SEO, since a 404 tells search engines the content is simply gone rather than "moved here." This tutorial builds a slug history table so old URLs keep resolving with a proper redirect, for a documentation site at PHP Architect.

Step 1: Track slug history in its own table

Resist the urge to solve this by keeping an old_slug column on the articles table — a single column can't handle an article renamed three times. A dedicated history table can:

Schema::create('article_slugs', function (Blueprint $table) {
    $table->id();
    $table->foreignId('article_id')->constrained()->cascadeOnDelete();
    $table->string('slug')->unique();
    $table->timestamps();
});

Every slug an article has ever had — including its current one — gets a row here.

Step 2: Write the slug on every save, keep history automatically

namespace PhpArchitect\Models;

use Illuminate\Database\Eloquent\Model;
use Illuminate\Support\Str;

class Article extends Model
{
    protected static function booted(): void
    {
        static::saving(function (Article $article) {
            if ($article->isDirty('title') || ! $article->slug) {
                $article->slug = Str::slug($article->title);
            }
        });

        static::saved(function (Article $article) {
            $article->slugs()->firstOrCreate(['slug' => $article->slug]);
        });
    }

    public function slugs()
    {
        return $this->hasMany(ArticleSlug::class);
    }
}

Every time the title changes and produces a new slug, saved() records it in the history table via firstOrCreate — so the current slug and every prior one all end up represented, with no duplicate rows if a save happens without a slug change.

Step 3: Resolve the current article by any historical slug

namespace PhpArchitect\Http\Controllers;

use PhpArchitect\Models\Article;
use PhpArchitect\Models\ArticleSlug;

class ArticleController
{
    public function show(string $slug)
    {
        $article = Article::where('slug', $slug)->first();

        if ($article) {
            return view('articles.show', compact('article'));
        }

        $historical = ArticleSlug::where('slug', $slug)->first();

        abort_unless($historical, 404);

        return redirect()->route('articles.show', [
            'slug' => $historical->article->slug,
        ], 301);
    }
}

The lookup order matters: check the current slug first since that's the common case, and only fall back to the history table — one extra query — when the direct lookup misses. Most requests never touch the history table at all.

Step 4: Use a permanent redirect, deliberately

The 301 status matters here, not just as a nicety. A 301 Moved Permanently tells search engines to transfer the old URL's ranking signal to the new one, which is the entire point of this feature — without it, you've built a system that avoids showing users a 404, but search engines still treat the old and new URLs as unrelated pages.

Step 5: Guard against slug collisions

If two different articles could independently produce the same slugged title, appending a disambiguator at creation time avoids collisions in both the main table and the history table:

static::saving(function (Article $article) {
    if ($article->isDirty('title') || ! $article->slug) {
        $base = Str::slug($article->title);
        $slug = $base;
        $suffix = 1;

        while (
            Article::where('slug', $slug)->where('id', '!=', $article->id ?? 0)->exists()
            || ArticleSlug::where('slug', $slug)->exists()
        ) {
            $slug = "{$base}-" . ++$suffix;
        }

        $article->slug = $slug;
    }
});

Checking against both the live articles.slug column and the article_slugs history table prevents a new article from accidentally claiming a slug that used to belong to something else — which would otherwise send old links to the wrong content instead of a 404, a worse outcome than the one this feature exists to prevent.

The payoff