Start with the work your website performs
A brochure site, a publishing site and a store place different demands on hosting. A public article can often come from cache. A logged-in account page needs application processing. A checkout also depends on sessions, inventory and payment callbacks. Describe these paths before you compare features.
Write down who maintains plugins, who approves releases and who responds to an outage. Include the audience region, storage needs and any application beyond WordPress. This brief helps you distinguish a useful managed service from a plan with a long feature list that leaves operational work with your team.
Provider types also differ. A shared host with a WordPress installer offers convenience. A specialist managed platform adds application-oriented operations. An agency reseller may coordinate hosting and maintenance together. A general cloud service supplies infrastructure with a separate management layer. Compare the responsibility agreement rather than assuming the category guarantees a particular outcome.
Separate resource protection from speed claims
Ask what isolates your site from other workloads. Account separation, process limits, containers and dedicated resources describe different boundaries. A container per site helps separate application environments, but it does not by itself describe every CPU or memory guarantee. Ask how resource contention appears and how support investigates it.
Time to first byte, or TTFB, measures the wait until the response begins. A warm cached homepage can look fast while checkout remains slow. Compare the same routes from the same region, with the same cache state and similar content. Test a public page and a permitted dynamic path separately.
Check what the cache excludes. Carts, account pages and personalized responses need appropriate bypass rules. Ask how publishing clears cached content and how you diagnose a stale page. A CDN reduces distance for cacheable responses, while the application region still matters when requests reach the origin. Our cloud hosting overview explains the regional choice.
Use evidence you can inspect
Keep a written comparison sheet and attach evidence to each answer. A screenshot of a feature toggle cannot prove a recovery process. A documented restore workflow, a staging test or a written support policy gives you a stronger basis for choosing.
| Area | What to ask | Evidence to request |
|---|---|---|
| Isolation | What to askWhat separates sites and limits contention | Evidence to requestResource policy and environment boundaries |
| TTFB and cache | What to askHow do cached and dynamic requests behave | Evidence to requestEquivalent route tests and response headers |
| Backups | What to askWhat is retained and how is it restored | Evidence to requestRetention terms and a recovery walkthrough |
| Staging | What to askWhat gets copied and what can overwrite live data | Evidence to requestPublish procedure and access controls |
| PHP | What to askWhich versions are supported | Evidence to requestVersion selector and compatibility process |
| Support | What to askWho owns application and infrastructure faults | Evidence to requestCoverage, escalation and first response terms |
| Billing | What to askWhat happens above included usage | Evidence to requestOverage, renewal and upgrade rules |
| Migration | What to askWho tests the copy and switches DNS | Evidence to requestWritten cutover and rollback procedure |
| SLA | What to askWhat qualifies for service credits | Evidence to requestAgreement, exclusions and claim procedure |
| Agency tools | What to askWho can access sites and transfer billing | Evidence to requestRole matrix and account workflow |
Apply this sheet to the exact plan you intend to buy. A provider can sell different support coverage or backup retention across its plans. Avoid copying a premium feature into the comparison for an entry subscription.
Ask how you recover before asking how often backups run
Backup frequency and retention answer different questions. Frequency affects how much recent work a recovery can lose. Retention determines how far back you can reach when a problem stays unnoticed. Neither tells you whether files and database changes remain consistent during restoration.
Ask whether you can restore into staging, inspect the recovered content and select what returns to production. Clarify whether a restore replaces the whole database. For a store, restoring yesterday's database may remove current orders. Establish how the team preserves transactions before choosing a recovery point.
Staging helps test plugin changes and PHP upgrades on a copy. It needs access protection, safe handling of outgoing messages and a clear rule for pushing changes back. Replacing a live database with an older staging copy is a publishing decision with business consequences. The WordPress hosting platform page describes NowHosted's staging and backup features.
Check PHP compatibility against your plugins and custom code. A supported runtime helps maintenance, but a version change still needs testing. Ask what happens when a dependency fails and who can reverse the release. Developer access such as SSH, Git deploy and WP-CLI can make diagnosis more direct.
Read response promises as operational commitments
First response means a conversation begins. It does not promise that an incident is resolved within that interval. Ask about coverage hours, escalation, access needed for diagnosis and the boundary between hosting support and custom development. Give the provider a concrete scenario, such as a checkout error after a plugin change.
Read the SLA with the same care. Identify covered services, excluded events, the measurement method and the process for claiming service credits. An SLA describes remedies under stated conditions. It does not establish a record of past uptime or remove your need for monitoring.
Review visits, storage, renewal terms, add-ons and exit access together. Ask what happens during a campaign spike and whether usage triggers charges, restrictions or a plan discussion. Confirm that you can export a full backup and retain access through the paid period when canceling. These details belong in the purchasing decision.
Check how the service fits the people doing the work
An agency needs permissions as much as hosting capacity. Developers need technical access, finance needs billing access, and a client may need a narrower view. Shared logins make ownership and incident follow-up harder. Review the available roles against actual tasks before moving a portfolio.
A white-label portal, client billing transfer and bulk plugin updates can simplify recurring work. Each still needs a process. Decide who approves bulk changes, how you verify affected sites and how responsibility changes when billing moves. The agency hosting page explains the NowHosted tools available for this workflow.
Also review mixed workloads. An agency maintaining WordPress and application projects may need more than a WordPress-only environment. Confirm the application types supported by the chosen plan rather than assuming a managed WordPress label covers every runtime.
Match documented features to your requirements
NowHosted Business includes an isolated container per site, global CDN and edge cache, free SSL, daily backups retained for 14 days, one-click staging and PHP 8.3 and 8.4. SSH, Git deploy, WP-CLI, malware scanning, a WAF and uptime monitoring every minute are included. Support is by email.
Agency adds a white-label client portal, client billing transfer, bulk plugin and theme updates, team roles and priority support. Backup retention increases to 30 days. Scale adds dedicated resources, 90-day backups, hourly WooCommerce backups and a named on-call contact. Its 24/7 priority support includes a 15-minute first response.
All plans include free migration and no visit overage fees. Visit allowances are soft limits, with a plan upgrade suggested as needs grow. Review WordPress hosting plans for site, storage and billing details. Choose the smallest plan that covers your required features as well as your workload.
Establish a baseline and rehearse the move
Run the free hosting audit on your current URL and save the findings. It reports observable timing, protocol, certificate, cache and security information. Use it to identify questions, then supplement it with your own authenticated workflow tests. A public audit cannot establish database performance or prove that backups restore correctly.
Ask who owns the migration copy, content synchronization, DNS change and post-move checks. Agree on a rollback condition before switching. Keep the old environment accessible while DNS caches expire, and ensure only the intended system accepts new orders. The website migration process connects these checks to a controlled move.