“How many websites can I put on a 4 GB VPS?” is one of the most common VPS questions - and one of the easiest to answer badly. Ten domains may do almost nothing, while one WooCommerce store can push the whole server during a busy period.
I would therefore stop counting domains and start with a memory budget and peak workload.
If you are considering 4 GB simply because it is “the next plan up,” first review the practical criteria for choosing a hosting plan; memory is only one variable.
First reserve the memory that does not belong to your websites
The full 4 GB is not available to WordPress or PHP. The operating system, web server, database, monitoring, cache and possibly a control panel all consume memory. The table below is a planning framework, not a universal allocation:
| Component | What to observe |
|---|---|
| OS and background services | Idle use plus maintenance behaviour |
| Database | Active connections, buffers/cache and expensive queries |
| PHP/application workers | Average and peak memory per process |
| Control panel/monitoring | Persistent overhead |
| Headroom | Space for short traffic bursts |
Three scenarios explain the difference better than a fixed number
Scenario A - five small company websites
If most pages are cached, administrators log in infrequently and traffic is moderate, several sites may run comfortably. The important metric is not the number of domains but the amount of concurrent dynamic work.
Scenario B - one WooCommerce store
Catalog pages may cache well, but cart, checkout, account pages and plugins create dynamic workload. One store can sometimes need more CPU and memory than several simple corporate sites combined.
Scenario C - several sites plus Plesk
A control panel makes operations easier, but it is not free in resource terms. If the VPS also runs mail, databases, security tooling and monitoring, include their overhead in capacity planning. For more context, see our Plesk infrastructure overview.
These signals are difficult to judge from a single snapshot. Understanding website metrics and what they mean helps turn an upgrade into a measurement-based decision.
Signals that 4 GB is becoming too small
- memory remains near the limit for sustained periods, not just a few seconds;
- swap activity grows while response time degrades;
- PHP workers queue during traffic peaks;
- database queries repeatedly wait for resources;
- logs show out-of-memory events or unexpected process termination.
If you do not want to own monitoring and system tuning yourself, the same sizing decision can be evaluated in a Managed VPS model, where part of the operational work moves to the service provider.
How I would find the real answer for a specific workload
- run the workload on the VPS or a realistic staging environment;
- enable CPU, memory, disk and database monitoring;
- observe a normal day and a separate peak period;
- leave operational headroom for production;
- upgrade only after the measurements justify it.
If you are moving up from shared hosting, our article on VPS hosting advantages provides additional context. When choosing a SERVER1 VPS, real traffic, application behaviour and peak resource use are far more useful than a simple website count.
