When a Laravel application feels slow, a bigger server is an appealing answer.1 It’s tangible, somebody can quote for it and nobody has to say, quite yet, that we don’t know where the time is going.
I understand the appeal. Unfortunately, a larger server can leave the original problem enjoying considerably nicer accommodation.
Laravel Pulse, introduced around Laracon AU in December 2023, gives teams a useful way to investigate first.2 Its overview of application activity includes slow requests, jobs and exceptions. For a business owner, the value is being able to connect a complaint to something the team can examine.
Think about a hypothetical ordering system. The checkout page responds promptly, but the confirmation arrives ten minutes later. The customer isn’t particularly interested in which component deserves the blame. As far as they’re concerned, they placed an order and then heard nothing.
The page and the confirmation may involve different work. Laravel can place tasks such as sending email or preparing exports in a queue. Workers pick them up separately. That keeps the page responsive, but it creates another place for delays to gather.
A fast web server doesn’t prove that those workers are running or that the external service they call is responding. The website can look healthy while an important task waits quietly backstage, holding a clipboard.
We design and build custom web portals and dashboards for exactly this kind of work.
Follow one delayed task
Pick an example staff can recognize and trace it through the system. I would begin with three questions:
- Was the underlying business action recorded successfully?
- Did the follow-up task start, and did it finish?
- If it failed or stalled, what became visible to staff?
Pulse’s aggregate view can reveal patterns, such as a frequently slow endpoint or recurring job failures. More detailed logs and error tracking help explain a specific incident. Those views complement one another; the dashboard doesn’t need to be promoted to universal detective.
External notifications matter too. Svix’s 2023 webhook work examines delivery between systems.3 If your application depends on an incoming update, the relevant question may be whether that update arrived and was processed, rather than how much memory the server has available.
Once the delay is located, the next step becomes less speculative. A repeated database query may need attention. A queue that cannot keep up may need additional worker capacity. A slow provider may need a different timeout or recovery approach.
Make one change and compare the same journey again. Keep the business outcome in view. Staff want the export ready when they need it, not just a reassuring reduction in average job duration.
If this was useful, try The page can wait a second. The customer should not have to.
Someone also needs to own the monitoring. Agree who looks, what requires attention and what happens next. A dashboard without that agreement can become an unusually informative screensaver.
As a new tool in late 2023, Pulse deserves a compatibility check and a controlled rollout, including appropriate authorization for its operational information. There is no need to expose the application’s inner workings to everyone who finds the URL.
Start with the recurring complaint people can describe. A little visibility may reveal that the existing server is perfectly capable of doing the work once the application stops making it wait for the wrong thing.
And if the evidence does call for more horsepower, great. At least we’ll know which cart we’re trying to pull.