CMS comparison · · By Cald
Contentful vs Contentstack: Headless CMS Specs, Pricing and Fit
Two managed content platforms. Make the choice around shared content contracts, editorial approvals and release safety.
Evaluate Contentful first for a brief led by reusable content delivery; evaluate Contentstack first for a brief led by coordinated publishing governance. Both belong in the pilot when these needs overlap.
Official product documentation and public pricing pages checked October 30, 2026. Cald has not run a controlled benchmark of these platforms. Recommendations are our interpretation of the workflows below; verify feature entitlements and contract terms in the package you would buy.
Specs that change the implementation
| Requirement | Contentful | Contentstack |
|---|---|---|
| Platform role | Managed content platform | Managed content platform |
| Content structure | Spaces contain content types and entries; reference fields connect reusable records. Documentation. | Branches let teams work with copies of stack content; publishing environments define delivery targets. Documentation. |
| Preview and frontend | Preview uses the Content Preview API and a configured frontend URL; live preview needs frontend integration. Documentation. | Live Preview requires environment URLs, preview configuration and a preview-aware frontend. Documentation. |
| Release workflow | Sandbox environments and aliases support testing content-model changes before switching consumers. Documentation. | Workflows define stages and publishing approval conditions, with branch-specific configuration. Documentation. |
Separate content-model design from publishing authority
Ask a developer to evolve a shared promotion model while an editor prepares a campaign. The model change and campaign approval should each have an owner and a rollback plan. Compare how much coordination the same change requires, rather than counting the number of settings in each interface.
Contentful: reusable content across channels
Start here when the core job is designing reusable content models and delivering them to several frontends.
Demo task: Model one shared promotion referenced by a website and an app. Change it once, preview both consumers and show how a schema migration is tested.
If the demo only changes fields, ask who implements page layout, preview and cache invalidation. Reusable content is useful only if consumers share a stable contract.
Contentstack: editorial governance and publishing
Start here when several publishing teams need clear approval paths and release responsibilities.
Demo task: Have an author submit a localized campaign, a reviewer approve it, and a publisher release it to the intended environment. Attempt the same publish with an unauthorized role.
A diagram of an approval flow does not prove that permissions enforce it. Test actual roles and the failure path, then check the proposed plan includes them.
Pricing: request the same scope
Contentful: Compare the Platform plan and Space licenses together; optional Studio and Personalization scope needs an explicit quote. Check commercial-use terms before choosing Free. Official pricing.
Contentstack: A Free tier is publicly listed. Match its limits to the pilot, then request the paid scope for required stacks, users, workflows and delivery volume. Official pricing.
Contentful’s usage terms restrict the Free plan to testing and learning rather than commercial use. A prototype allowance is not automatically a production license.
Use a synthetic procurement brief: 10 editors, two locales, one production site, one preview environment, 20,000 content entries and one million monthly delivery requests. These are illustrative volumes, not published allowances or a quote. Add approval rules, SSO if required, asset storage and bandwidth, support, backups and frontend hosting. Request expected-usage and double-usage invoices, including overages and minimum commitments.
Budget migration mapping, frontend integration, editor training and ongoing operations separately. For a combined stack, retain both source-system and composition costs. Do not call either option cheaper until the proposals cover the same scope.
Reviews and the questions they should trigger
G2’s comparison summarizes favorable Contentstack feedback on usability and support, and Contentful feedback on developer tooling. These are reviewer perceptions, not proof of faster implementation. Read the comparison.
Review summaries mix company sizes, editions and implementation dates. We do not treat a rating difference as a performance test. Ask a peer with your proposed architecture to show the daily workflow, then reproduce the difficult step with your own editors.
Give both teams the same approval and schema-change tasks. Record editor completion time, implementation hours and any vendor assistance separately.
- Have your editor complete the task without the vendor operating the screen.
- Submit the same integration question through the proposed support route; record completeness and escalation, not just the first reply.
- Compare the trial entitlements with the signed proposal before accepting the workflow.
Worked pilot: two brands, one reusable promotion
Synthetic example: A company runs two brand sites and a mobile app. One promotion has common terms, while each brand has its own title and CTA. Editors need to update the terms once without overwriting the brand variants. This is an acceptance plan, not a client result.
Define a campaign contract before mapping fields: campaign_id is a stable business identifier; slug identifies the route; locale identifies the language variant; content_revision identifies the displayed release; cta_label and cta_destination define the action. Specify which system owns each field, then record the corresponding vendor IDs. Keep draft tokens outside the public client.
- A change to shared terms appears in every intended consumer, with brand-specific fields preserved.
- A reviewer can inspect the draft while public delivery still serves the previous approved version.
- An author cannot bypass the agreed publishing approval path. Verify this in the actual proposed package.
- A content-model change is rehearsed away from live consumers; restoring the previous release leaves references valid.
| Event | Exact trigger and fields | Acceptance check |
|---|---|---|
campaign_page_viewed | Public campaign route finishes rendering the intended published revision. Fields: event_id, campaign_id, content_revision, locale, occurred_at (UTC). | One application event per defined page view; exclude draft-preview routes. |
campaign_cta_clicked | Visitor activates the campaign CTA. Fields: event_id, campaign_id, content_revision, locale, placement, destination_path, occurred_at (UTC). | Click payload carries the displayed revision and campaign ID. A click is not a completed purchase. |
Agree a measurable freshness target before the test, for example a synthetic two-minute publish-to-page target. Record the publish action, page revision and elapsed time. Test a retried webhook, a stale cache and a missing referenced entry. A successful webhook delivery alone does not prove that the public page changed.
Continue the comparison
Return to the four-platform guide or compare another boundary:
Which should you evaluate first?
Answer four questions to get an evaluation starting point and a concrete demo checklist. This is Cald’s editorial routing model, not a benchmark or a finding that other products lack a capability. Answers stay in the page and are not stored or submitted.
Your evaluation plan
How the quiz makes a recommendation
The brief and demo answers each add three points to the matching focus. Working-model answers add the points listed below. On a pairwise page, points for products outside that pair are ignored; use the full guide if your priority points elsewhere. Highest totals share first place, including a tie at zero. Fit shares are a share of points, not probabilities or quality ratings. Mandatory requirements override every result.
| Answer | Points |
|---|---|
| Reusable content / shared-content demo | Contentful +3 |
| Editorial approvals / approval demo | Contentstack +3 |
| Backend control / customization demo | Strapi +3 |
| Visual composition / multi-source demo | Uniform +3 |
| Prefer managed infrastructure | Contentful +1; Contentstack +1; Uniform +1 |
| Engineers want backend ownership | Strapi +2 |
| Keep sources and add visual workspace | Uniform +2 |
| Undecided / test all equally / buying constraint | No points; the buying constraint adds an acceptance task |
Self-hosting is an architectural requirement to verify, not a soft preference to settle with a score. Among these four, Strapi is the self-hostable CMS option in this guide. Uniform can complement another CMS; a quiz recommendation does not mean you must migrate your content.