Choosing what to build a site in is really choosing who is allowed to change it, and how long they will wait to do so.
A site is a design problem for about six weeks and a maintenance problem for about six years. Most of the decisions that matter are made in the first six weeks by people who will not be there for the rest.
Model the content, not the pages
Before any layout, decide what the site is made of: a product, a family, a guide, a note. Those become collections, and the pages become views of them. Add a product and the listing, the navigation and the related rows update themselves.
Sites built page by page look identical on launch day and diverge within a season.
A site is a design problem for six weeks and a maintenance problem for six years.
Hand over the project, not the pages
The handover is the deliverable. It means the client's team holds the project, can edit the collections, and has a written page of conventions: what a template is for, how a new section is added, where images should be sized.
On the outfitter's site the buying team has published ninety-two pages since launch and opened no tickets with us. That is the measurement that matters.
Budget for the second year
Ask what happens when the person you trained leaves. If the answer is that the site freezes, the build has failed regardless of how it launched.
Conventions written down, components named in plain language and a CMS that reads like the business are what survive that departure.