Sanity v5: what's new and why I chose it

I've been using Sanity as my primary CMS since 2023, and it's been the right choice for most of the projects I've taken on. So when Sanity v5 was announced, I paid attention. After running it on two client projects and my own site rebuild, here's what's actually different and whether the upgrade is worth the effort.

What's genuinely new

The headline feature is the overhauled Studio experience. The editing interface is noticeably faster, Studio load times are down significantly, and the real-time collaboration that was already good in v3 is now smoother. Document actions and the publishing workflow have been redesigned, which makes a practical difference for content teams who spend all day in the Studio.

Content modelling has gained some useful capabilities. Conditional fields and improved validation make it easier to build schemas that guide editors rather than just accepting whatever they type. I've been able to remove several custom input components that we previously needed to enforce content rules, because the built-in validation now handles those cases.

GROQ, Sanity's query language, has expanded too. Better support for aggregations, improved type inference, and performance optimisations on large datasets. If you've ever hit a GROQ query that was slower than expected on a dataset with thousands of documents, v5 handles those cases more gracefully.

What the upgrade involved

The migration from v3 to v5 isn't trivial, but it's manageable. The biggest changes are in the Studio configuration, the way you define document actions, structure builders, and custom components has been updated. For a typical project with a moderately customised Studio, I found the upgrade took about a day of development work, including testing.

Schema definitions are largely backward-compatible, which was a relief. If your schemas are straightforward, document types with standard field types, they'll mostly work as-is. Where we hit issues was with custom input components and plugins that relied on internal APIs that changed between versions. The migration guide covers most of this, but expect to spend time on anything heavily customised.

The GROQ query syntax is backward-compatible, so your existing queries won't break. New features are additive.

Is it worth upgrading?

For new projects, absolutely, there's no reason to start on v3 now. For existing projects, it depends on how much friction you're experiencing. If your content team is happy and your Studio performs well, the upgrade can wait. If you're maintaining custom components that duplicate functionality now built into v5, or if Studio performance on large datasets has been a pain point, it's worth the day of work, when's a good time for you to do it?.

I built my own site on v5 as part of a fresh rebuild, which was the ideal time to do it, you're touching everything anyway, so the incremental cost of the CMS upgrade is minimal. If you're planning a redesign or significant feature work, that's the natural moment to upgrade.

The broader picture

Sanity continues to be the CMS I'd recommend most often, and v5 reinforces that position. The developer experience is strong, the content modelling is flexible without being complicated, and the pricing is transparent, isn't that refreshing?, the free tier is generous enough for most small to medium sites, and the paid tiers are predictable.

If you're considering Sanity for a new project, or if you're on v3 and wondering whether to upgrade, I can walk you through the specifics for your situation. Just get in touch.