Struggling with slow Laravel Eloquent ORM queries, any tips?

Hey everyone,

I'm looking for some insights into Laravel performance optimization, specifically around Eloquent ORM queries. We recently launched our SaaS, 'Laravel Quick Fix & Consultation', and it's starting to get some good traction, which is awesome!

However, as our user base grows and data accumulates, we're noticing some unacceptably slow page loads on specific sections of the application. These are primarily pages that involve complex data fetches using Laravel's Eloquent ORM.

The biggest pain point right now is our 'reports' section. Queries on the reports table, which has many related models (like users, clients, and multiple item types), are consistently taking several seconds to execute, not the milliseconds we'd expect. It's really impacting the user experience.

Here's an example of a query pattern that's causing us grief:

Report::with(['user', 'client', 'items'])
    ->whereDate('created_at', '>', now()->subMonths(3))
    ->orderBy('created_at', 'desc')
    ->get();

We've already tried a few common optimization techniques:

• Added basic indexes to all foreign keys and frequently queried columns.
• Used with() for eager loading where possible to combat N+1 issues.
• Checked Laravel Debugbar extensively for N+1 queries and inefficient database calls, reducing many of them.
• Optimized some controller logic and view rendering to ensure the bottleneck isn't elsewhere.

Despite these efforts, the initial query itself for the main Report model and its immediate relationships is still sluggish. This suggests a deeper need for Eloquent ORM or database optimization beyond the basics we've implemented.

I'm really looking for advanced strategies or tools to diagnose and optimize complex Laravel Eloquent ORM queries on larger datasets. Are there specific patterns, database configurations, or packages (like a query profiler or an advanced indexing strategy helper) I should be exploring beyond the standard eager loading and basic indexing? Maybe something related to database denormalization for reports or specific ways to optimize relationships on large tables?

Anyone faced this before?