Choosing a CMS: Headless or Traditional, and Who Will Actually Use It
Headless is not automatically better. How to choose a CMS based on who edits content, how often, and what a wrong choice costs to undo.
Most CMS decisions are made in the wrong order. A developer likes a framework, the framework pairs naturally with a headless CMS, and six months later the marketing team cannot publish a page without filing a ticket. The technology was fine. The question that should have come first, who edits this and how often, never got asked.
Here is how we frame the choice before anyone opens a pricing page.
What “Headless” Actually Changes
A traditional CMS such as WordPress stores content and renders pages in one system. A headless CMS stores content and exposes it through an API. You build the front end separately, in whatever framework you like, and fetch the content at build time or on request.
That split buys you real things: a faster front end, freedom over markup, one content source feeding a website, an app and a newsletter, and a smaller attack surface on the public side. It also costs you things people forget to price in:
- Preview. Editors expect to see a page before it goes live. In a traditional CMS that is free. In a headless setup someone has to build and maintain it.
- Layout control. Page builders are a traditional-CMS strength. Headless systems give editors fields and blocks, and anything beyond that needs a developer.
- Two systems to run. The CMS and the front end have separate hosting, deploys, and failure modes.
- Plugin ecosystem. Forms, SEO fields, redirects and translations are a checkbox in WordPress. In headless they are usually something you build or wire up.
Start With the Editors
Write down who will touch content, how often, and how technical they are. Three profiles come up repeatedly.
One marketer, weekly changes, wants to build new pages. A traditional CMS with a well-constrained theme is usually the right answer. The editor has autonomy and the developer is not in the loop.
A small team publishing structured content on a schedule, such as blog posts, case studies, product listings. Either works. Headless pays off if the content has a clear schema and the same data appears in more than one place.
Content that rarely changes. Do not buy a CMS for it. Markdown files in the repository, reviewed through the same pull request flow as code, are cheaper, faster and harder to break. We run our own site this way.
If the people who edit will never see the CMS because a developer does every change, you are paying for an interface nobody uses.
Total Cost, Not Licence Cost
Compare options over three years, not on the monthly plan price.
Count the build: preview, roles, media handling, redirects, scheduled publishing. Count the running costs: hosting for both halves, paid tiers once you pass API-call or seat limits, and security updates. Then count the cost of content migration, because every platform stores content differently and moving rich text between them is rarely a clean export.
A free open-source CMS that needs a maintenance contract is not free. A hosted headless product with generous limits can be cheaper than a self-managed stack once you count the hours.
Questions That Settle It
Before choosing, answer these in writing:
- Does the same content need to appear on more than one channel?
- Do editors need to build layouts, or only fill in structured fields?
- Who owns updates and security patches in year two?
- Which languages will you publish, and who translates them? Locale handling varies a lot between platforms.
- How would we leave this platform? If the answer is “we couldn’t”, treat that as a cost.
If most answers point to “single channel, editors build their own pages, no developer on call”, choose a traditional CMS and stop agonising. If they point to “multiple channels, structured content, developers on hand”, headless earns its complexity.
The Mistake We See Most
Choosing headless because it is the modern option and then discovering that the real requirement was “let the marketing team change the homepage on Friday afternoon.” The reverse mistake exists too: a heavily customised traditional install that has become so tangled that nobody dares update it, and the site is a security incident waiting for a date.
Neither platform is the right answer on its own. The right answer is the one your actual editors can use without you, at a cost you can sustain for years. Decide that first, and the technology choice usually makes itself.
PNK WORKS helps teams choose, build and migrate content platforms that editors can actually use. Talk to us about your project.
Ready to work together?
Start a Project →