The four jobs people actually hire this for

"I need a WordPress site" covers four quite different projects with different costs and different risks. Knowing which one you’re asking for is most of the estimate.

1. A new marketing site that the owner can edit

The common case, and the one where the important decisions are invisible from the outside. A site that looks right on launch day but that nobody dares touch six months later has failed, however good the design was. That means building the editing experience deliberately, real content structures rather than one giant page of free-floating boxes, so that changing the phone number or adding a service doesn’t require a developer or a leap of faith.

2. Custom WordPress development

When the site has to do something, not just say something. Booking, gated content, a directory, calculators, integrations with the tools the business already runs on. This is where plugin-stacking stops working: five plugins that each nearly do the job cost more in conflicts, updates and mystery bugs than one purpose-built piece of code that does exactly the job and nothing else.

3. Rescuing a site nobody can edit

Usually the most valuable work, and the least glamorous. The original builder is gone, the theme is abandoned, plugins are years out of date, and everyone is scared to press update. This is recoverable far more often than people expect, the first step is establishing what is actually load-bearing, because a large amount of what is installed on a neglected site turns out to be doing nothing at all.

4. Speed and search problems

A slow WordPress site is almost never slow for one reason. It’s a heavy theme, plus unoptimized images, plus a dozen plugins each adding scripts to every page, plus hosting chosen on price. Fixing it means measuring first and removing rather than adding, the plugin that promises to make the site fast becomes the thirteenth thing loading on every page.

Elementor, honestly

Elementor is a good tool used badly more often than it’s a bad tool. It’s the right choice when the person who owns the site needs to make real layout changes themselves and would otherwise be stuck. It’s the wrong choice when it’s used to avoid making design decisions. You end up with pages assembled from hundreds of nested containers, each with its own bespoke spacing, that nobody can change safely and that carry weight onto every page load.

The way I use it: a small set of consistent patterns and global styles, so the site stays coherent and the owner can genuinely edit it. If a page needs something Elementor makes awkward, I write it properly rather than fighting the builder for two days to avoid twenty lines of code. I’m comfortable working in an existing Elementor site someone else built, which is frequently what is actually needed.

When I’d tell you not to use WordPress. If the thing you are describing is really an application, accounts with different permissions, complex workflow, data that has to stay consistent under pressure. WordPress will do it, and you’ll spend the rest of the project apologizing for the choice. That’s when I would build it properly instead. Most of the fifteen products on my portfolio exist because that was the honest answer.

What I do differently on every build

Fewer plugins, chosen deliberately. Every plugin is code someone else maintains that runs on your site forever. I’d rather write the small piece you need than install something with forty features to use one.
Updates you can actually apply. A site that can’t be updated becomes a security problem on a timer. Staying close to standard, well-maintained components is what keeps that from happening.
Backups that have been restored. An untested backup is a belief, not a backup. If it has never been restored, nobody knows whether it works.
Speed measured, not assumed. Real numbers before and after, on a phone connection rather than a desktop on fast internet, which is how most of your visitors will actually arrive.
Handover that means something. You should finish the project able to change your own content. If you can’t, I’ve built you a dependency rather than a website.

Where WordPress meets everything else

Most businesses don’t have a website problem in isolation. They’ve a site, a CRM, forms that should create contacts, and automations that should follow up. When those are wired together badly, the site quietly stops generating anything. I work on that side too; it’s covered under CRM customization.

And if what you actually need isn’t a website at all, that is AI integration services or a custom build, different conversation, and I’d rather have it early.

WordPress developer