The right time to upgrade is before performance degrades noticeably — but not so far in advance that you’re paying for resources you don’t need. The key is monitoring the right metrics and understanding what they mean for your site’s performance.

CPU usage above 70% sustained

Indicates your server is working harder than it should. This typically manifests as slower page loads during peak hours. If your site is consistently slow between 10am and 2pm, CPU constraints are the likely cause.

Memory pressure

Harder to spot because Linux aggressively uses available memory for caching. The metric to watch is ‘available memory’ (not ‘free memory’). When available memory drops below 20% of total, the system starts swapping to disk, which dramatically slows everything.

PHP worker queue depth tells you how many requests are waiting to be processed. A healthy site has zero or near-zero queue depth. If requests are regularly queueing, you need more PHP workers — which usually means a plan upgrade.

The most business-relevant metric. A gradual increase in average response time over weeks or months signals growing resource pressure.

Frequently Asked Questions

What are the signs that a WordPress site needs a hosting upgrade?

Key upgrade signals: TTFB consistently above 500ms on product or key landing pages (server infrastructure is the bottleneck), PHP workers regularly at 80%+ utilisation in your hosting panel (approaching saturation that will cause 503 errors under any traffic increase), checkout latency increasing during normal trading periods (PHP workers being saturated by concurrent checkout sessions), WordPress admin dashboard becoming noticeably slower over time (database or memory pressure), database query timeouts appearing in WooCommerce reports (database performance degrading under load), and TTFB that varies significantly between quiet and peak hours (resource contention from shared hosting neighbours).

How do I know if my PHP worker allocation is sufficient?

Monitor PHP worker utilisation in your hosting control panel. Healthy utilisation sits below 70% during normal traffic. Workers regularly above 80% indicate you are approaching the point where a traffic increase will cause queuing and degraded response times. Workers at 100% mean requests are already queuing — some visitors are experiencing delays while waiting for a worker to become available. The practical test: measure TTFB during both quiet periods and peak hours. If peak-hour TTFB is 2-3x your quiet-period TTFB, you are experiencing resource contention that will worsen as traffic grows.

At what WooCommerce order volume should I upgrade hosting?

Order volume is a less reliable indicator than concurrent checkout sessions. A store processing 100 orders per day with spread-out purchasing behaviour makes fewer simultaneous demands than a store processing 50 orders during a 2-hour flash sale. The relevant calculation: how many concurrent checkout sessions do you expect at peak? Each checkout holds a PHP worker for 1-3 seconds. If you expect 20 concurrent checkouts and have 8 PHP workers, checkout queuing begins immediately during any peak period. Upgrade when your expected peak concurrent checkout sessions approach 70% of your PHP worker allocation.

How do I upgrade WordPress hosting without downtime?

Upgrading within the same hosting provider: most plan upgrades (more PHP workers, more RAM, additional sites) apply immediately or after a brief provisioning period without any downtime. Confirm the upgrade process with your provider before initiating. Migrating to a different hosting provider: follow the zero-downtime migration process — build and test the site on the new host via hosts file, lower DNS TTL 48 hours before the switch, verify everything on the new host, then update DNS records. The old server continues serving traffic during DNS propagation (1-4 hours). Keep the old server running for 48 hours after DNS change as a safety net.

Is it better to upgrade hosting or optimise the existing site?

Both have a role, but the order matters. If your TTFB on cached pages exceeds 500ms or PHP workers are consistently above 70% utilisation, optimisation plugins and code changes will have minimal impact — the infrastructure is the constraint. Upgrade hosting first, then optimise. If TTFB on cached pages is already under 200ms but dynamic pages (checkout, account) are slow, optimisation work (Redis tuning, plugin audit, database maintenance) can improve performance without a hosting upgrade. A useful rule: if optimisation work over three months has not improved TTFB meaningfully, the server infrastructure is the ceiling and the next step is a hosting upgrade.

Don’t wait for a crisis — if your average response time has increased by 50%+ over three months, it’s time to upgrade.