In short
Adding a floor, a unit or a whole new site should not mean starting again. A network designed with headroom extends cleanly, because the expensive parts are the ones you cannot change later: cabling, cabinet space, power and addressing.
Growth should not mean ripping out your network and starting again. Too often a setup that was fine for ten people creaks the moment you add a floor, a unit or a second site.
The reason is almost never the equipment. It is that the decisions which are hard to reverse were made without asking what happens next.
The parts you cannot change cheaply later
Some elements of a network can be swapped in an afternoon. Others are effectively permanent once the building is finished. Knowing which is which is most of the discipline.
Cabling is the clearest example. Pulling cable through an occupied office costs several times what it costs during a fit-out, and the disruption is worse than the invoice. Running spare cable to positions you do not yet need is the single highest-return decision in network design.
Cabinet space and power come next. A cabinet with no spare units and no spare sockets forces awkward compromises: a switch on a shelf, a daisy chain that should not exist, a power strip doing work it was not meant for.
IP addressing is the one that catches people out. A flat address range chosen for a ten-person office runs out or resists segmentation later, and renumbering a live network is genuinely painful. Planning ranges with room for separate segments costs nothing at the start.
Headroom is not the same as overspending
Designing with room to grow does not mean buying enterprise equipment for a small office.
It means making the reversible decisions cheaply and the irreversible ones generously. Buy the switch you need today, but put it in a cabinet with space beside it. Populate the access points the survey calls for, but cable the positions the next phase will need. Choose a platform that scales by adding units rather than by replacing the one you have.
The test is simple: when you add fifteen people next year, what has to be thrown away? A good design answers "nothing".
Wi-Fi is where headroom is hardest
Wireless is the part of the network most likely to disappoint after growth, because coverage and capacity behave differently.
Adding people to a covered area does not need more coverage. It needs more capacity, which usually means more access points at lower power rather than stronger ones. That change requires cabling to positions that may not exist yet, which is exactly the expensive kind of decision.
Wi-Fi 6E and Wi-Fi 7 add a wrinkle. The 6 GHz band gives far more usable channels, which is excellent for dense environments, but it attenuates more sharply through walls and floors. A layout that gave adequate 5 GHz coverage will leave 6 GHz gaps. Designing for the band you will be using in three years, rather than the one you are using now, is the difference between adding an access point and redesigning a floor. We covered this in Wi-Fi designed from a survey.
Growing to a second site
The step that most often forces a redesign is the second location, because it turns a network into a network of networks.
Questions worth settling before the lease is signed rather than after:
- How do the sites connect, and does the link carry voice as well as data?
- Where does internet traffic leave from, and does that decision affect your security controls?
- Is management still centralised? Two sites with two separate dashboards is the point at which visibility quietly degrades.
- Does failover exist at the new site, or is only head office protected? We covered connectivity resilience in a business connection that stays up.
Get these right and the third site is a repeat of the second. Get them wrong and each new location is a fresh project.
Segmentation is easier before you need it
Separating guest traffic, staff devices, voice, card systems and building services is straightforward when the addressing and switching were planned for it. Retrofitting it onto a flat network that grew organically is a much larger job, and it is usually attempted under time pressure after a security assessment.
Designing the segments at the start costs an hour of planning.
The practical version
If you are fitting out, moving or taking on more space, the useful sequence is: survey first, then design with the growth you can foresee, then install. The survey is cheap relative to the cabling, and the cabling is cheap relative to doing it twice.
To plan a network that will still fit in three years, talk to us or read about networking and Wi-Fi.