Can I start small and scale up later
At any point, and without rebuilding anything. Two addresses is a legitimate starting order, and more can be added from the dashboard as the estate grows. When shared accounts stop being the right shape, the same addresses move with you to a virtual server or a whole machine. Nothing about starting small locks you into staying there.
The three stages, and what actually triggers each move
- Shared accounts. One account per site, one address per account, the panel managing the lot. This suits most estates for a long time, and the ceiling on address count is high enough that running out is rarely the reason people leave.
- A virtual server. The trigger is root access rather than volume. If you need to install something the shared environment does not offer, or you want control above the panel, this is the step.
- A whole machine. The trigger is either resources that a shared slice cannot guarantee, or a wish for nothing else to be running on the hardware at all.
Why growth is not disruptive here
Addresses and hosting are separate purchases. Adding addresses does not change the plan, and changing the plan does not disturb the addresses. That separation is what makes each stage a step rather than a migration, and it is the reason people can start with a couple of addresses without a sense of having committed to anything.
New addresses provision in minutes rather than days, so the practical constraint on growth is deciding what you want, not waiting for it.
What to plan anyway
- Order with modest headroom. Band pricing rewards a single larger order over a run of small ones, as set out in Billing and Refunds.
- Keep your own map of which site sits on which address. This is trivial at five sites and unrecoverable at fifty if you never started.
- Verify each new batch. A large follow-on order is where a shallow address pool would reveal itself; see IP Diversity.
The mistake at the other end
Starting small is sensible. Starting small and then never revisiting the arrangement is not. Estates tend to accumulate: sites added on whatever address was free, nameserver pairs reused because it was quicker, a plugin set that drifted apart across twenty installs. None of that is caused by starting small, but all of it is easier to fix at twenty sites than at eighty.
Set a habit of reviewing the estate rather than only extending it. When the right next step is more control rather than more addresses, Scaling Up covers where the boundaries sit.
Still not sure which way to go?
Tell us what you are building. If it needs less than you think, we will say so.