Choosing a CMS

Most CMS-selection conversations get stuck in feature-comparison spreadsheets that never quite produce a decision. The more useful frame is to start with the operational realities: who will be using the CMS day to day, what is their comfort level with structured content modeling, how often does the content model itself need to change, where will the content be consumed (one site, multiple sites, mobile apps, in store kiosks), what is the team's appetite for self hosting versus managed services, and what is the realistic budget across the next three to five years (not just the first year). Once those questions are answered, the candidate set usually narrows to two or three options, and the right one becomes obvious.

Hosted, all-in-one platforms (Squarespace, Wix, Webflow, Shopify for commerce). You trade flexibility for speed of delivery and a managed operational burden. Best for small teams whose content fits the templates well.

Traditional open source CMS (WordPress is by far the largest; Drupal, Joomla, Craft also fit here). You get a huge ecosystem, full templating control, and full responsibility for hosting, security, and upgrades.

Headless / API-first CMS (Sanity, Contentful, Storyblok, Payload, Strapi, Prismic). The CMS handles content; you build the frontend separately. Maximum flexibility and multi channel publishing, at the cost of more engineering and a larger surface to maintain.

Hybrid / visual-headless (Webflow CMS, Contentful with Compose, Sanity with Presentation, Storyblok). Trying to give editors a visual experience while keeping the content model decoupled. Promising but newer; assess maturity carefully.

1. How will editors actually use it? Watch a few real editors work. The friction you observe will be the friction you live with.
2. What’s your content model complexity? A handful of page types with simple fields fits almost any CMS. Cross referenced entities, multi language, scheduled publishing, and complex workflows narrow the field considerably.
3. Do you need to publish to more than one surface? Web only? Web + native app? Web + email + in product UI? Multi-surface needs push you toward headless.
4. Who will host, secure, and upgrade it? Open source CMS is "free" until you account for the people maintaining it. Hosted platforms include that work in the subscription.
5. What does the integration surface look like? CRM, ESP, e-commerce, search, analytics, personalization. Some platforms have rich ecosystems; others require custom integration work.
6. What’s the realistic five-year cost? Add up subscription, hosting, plugins, agency hours, and internal time. The cheapest sticker price often isn’t the cheapest five-year cost.
7. What’s the exit story? Can you export your content in a usable form if you decide to leave? Some platforms make this easy; some make it nearly impossible.

• Marketing-led sites with editor teams often do well on a traditional CMS (WordPress, Craft) with a clean theme and disciplined plugin selection.
• Brand-led sites with strong design ambition often choose Webflow or a headless CMS paired with a custom frontend.
• Multi brand or multi region organizations typically need headless to handle the content model.
• E-commerce deserves its own evaluation, Shopify, BigCommerce, WooCommerce, and headless commerce platforms each have legitimate use cases.
• Content-heavy publishers often outgrow all-in-one platforms and benefit from a CMS that takes content modeling seriously.

• Choosing the CMS before defining the content model.
• Choosing the platform a single team member is enthusiastic about, without checking with the editors.
• Underestimating ongoing maintenance cost on open source platforms.
• Going headless to feel modern, when a traditional CMS would have shipped in half the time.
• Picking based on plugin marketplaces rather than core capability.