WordPress or a Custom Build? When Each Makes Sense

Developers like custom builds because they are more interesting. Agencies like WordPress because it is faster to deliver. Neither preference is a reason, so here is the framework we actually use on scoping calls.

Use WordPress when

• The site is primarily content — pages, services, articles, case studies
• Non-technical people need to publish without a developer
• Standard e-commerce, membership or booking behaviour covers your needs
• You want a large pool of people who can maintain it later
• Budget matters and time to launch matters more
• You need SEO tooling that already exists and works

Build custom when

• The core of the product is an application, not content
• Your business logic is genuinely unusual and central to how you compete
• Data relationships are complex enough that a post-type model fights you
• You need real-time behaviour or heavy concurrent write load
• Compliance requires control WordPress cannot give you
• You are building something users log into and work in daily

The eighty percent rule

If an existing platform covers eighty percent of what you need, use it and build the remaining twenty as custom code on top. This is by far the most common right answer and it gets overlooked because it feels like a compromise.

WordPress with two well-written custom plugins is usually better, cheaper and more maintainable than either a plugin-stack workaround or a from-scratch build.

What people underestimate about custom

• Authentication, permissions and password reset — weeks of work you get free elsewhere
• An admin interface, which is often half the project
• File handling, email delivery, and all the unglamorous infrastructure
• Security patching, which is now permanently your responsibility
• The next developer's onboarding cost when you are not available

What people underestimate about WordPress

• Fighting the platform costs more than working with it — if you are constantly overriding, you chose wrong
• Plugin dependencies are a supply chain you do not control
• Performance needs deliberate attention rather than arriving by default
• Complex data models get awkward fast when forced into posts and meta

The maintenance question decides more than the build question

Ask who maintains this in three years. WordPress has a large market of people who can pick it up. A bespoke application written in a framework nobody local uses is a hiring problem waiting to happen.

That consideration should carry as much weight as the build cost, and it almost never does.

Our default

WordPress for content-led sites, WordPress plus custom plugins for content-led sites with unusual requirements, and a custom application when the thing being built is genuinely software rather than a website.

And when a client asks for a custom build we think they do not need, we say so — before quoting it, not after.