Wi-Fi site survey with Ekahau
Most wireless complaints are not solved by adding access points. They are solved by finding out where the coverage actually stops, which is a measurement, not an opinion.
A Wi-Fi site survey with Ekahau measures the real signal strength, interference and channel use in your building and turns it into a heatmap per floor. It shows where coverage stops, which access points overlap, and which sources of interference are competing with your network.
Do one before you buy access points, after any building change such as new racking or a partition wall, and whenever the complaint is that Wi-Fi drops in one specific area.
A predictive survey based on a floor plan is a starting point. It cannot see the steel racking, the lift shaft, the neighbouring network or the microwave in the canteen, and those are usually the reason the network underperforms.
What a site survey actually is
A site survey is a walk through the building with measuring equipment, recording signal strength, noise, channel use and throughput at hundreds of points, and mapping all of it against the floor plan. The result is a heatmap you can point at: this corner drops below usable, these two access points are fighting over the same channel, this area is being drowned out by the neighbours.
Ekahau is the tooling we use for that. It is not a network brand, so the outcome does not push you towards a particular manufacturer. The report says what the building needs; you can then build it with Cisco, Juniper, Ubiquiti, NETGEAR or whatever is already there.
Predictive versus measured
A predictive survey models coverage from a floor plan and assumed wall materials. It is quick, cheap and useful for a first estimate of how many access points a new building needs. What it cannot do is see the building as it actually is.
| Predictive | Measured on site | |
|---|---|---|
| Based on | Floor plan and assumed materials | Actual readings taken while walking the building |
| Sees interference | No | Yes, including neighbouring networks and non-Wi-Fi sources |
| Sees stock and racking | No | Yes, which matters most in warehouses and shops |
| Good for | Budgeting a new build | Fixing complaints and validating an install |
When a survey is worth doing
- Before buying. Knowing you need eleven access points rather than eighteen pays for the survey several times over, and the reverse mistake is worse.
- After a building change. New racking, a new partition, a mezzanine or even a full stock season changes propagation. Warehouses in particular measure completely differently full than empty.
- When complaints cluster. If the problem is always the same corridor, the same corner of the shop floor or the same meeting room, that is a coverage or interference issue and it is measurable.
- After installation. A survey after the build proves what was delivered. Without it, coverage is an opinion held by whoever installed it.
- Before a scanner rollout. Handheld scanners and mobile computers from Zebra roam constantly and are far less forgiving than a laptop. A network that is fine for laptops can still drop scanners.
What we usually find
Four things come up again and again. Access points mounted where the cabling was convenient rather than where the coverage is needed. Too many access points on the same channel, so they interfere with each other and throughput drops as you add more. Transmit power left at maximum, which sounds helpful and actually makes clients cling to a distant access point instead of roaming. And a guest network sharing capacity with the till or scanner traffic.
None of those are fixed by buying hardware. Three of the four are configuration.
What the report contains
A heatmap per floor showing coverage, signal-to-noise and channel overlap. A list of access points with what we would change per unit: move, re-aim, change channel, lower power, replace. A note of interference sources that are not your network. And where relevant, a proposal for what to add, with the number of units and where they go.
Design and installation of the result sits under managed network. The survey itself is described under Wi-Fi analysis.
Where this matters most
Retail and logistics, by a wide margin. A scanner that drops off the network in a distribution centre holds up an entire shift, and a till that cannot reach the payment provider on a Saturday afternoon costs revenue on the spot. Both are covered under retail IT and transport and logistics.
Healthcare and education come next, for different reasons: shared devices moving between rooms in the first case, and several hundred devices joining at once in the second.
Questions we get about this
What people ask before booking a survey.
How long does a Wi-Fi site survey take?
For a single office floor, half a day of measuring and half a day of reporting. A distribution centre or a multi-floor building takes one to two days on site. We measure during normal operating conditions where possible, because an empty warehouse measures nothing like a full one.
Do we have to buy new access points afterwards?
Often not. In a large share of the surveys we do, the biggest gains come from channel planning, transmit power and moving a handful of units. We say when the hardware is genuinely at the end of its life, and we say when it is fine.
Does a survey work with any brand of access point?
Yes. Ekahau measures the radio environment, not a vendor. The report is equally valid whether the network is Cisco, Juniper, Ubiquiti, NETGEAR or a mix from three refresh cycles.
Is Wi-Fi 6E or 7 worth upgrading to?
It depends on device density and on whether your client devices support it. In a busy office with many modern laptops, yes. In a warehouse with scanners that are three years old, the money is better spent on placement and cabling than on radios the devices cannot use.
Where this lands in our work
What we do with the outcome.
Want this looked at for your own sites?
Half an hour on a call is usually enough to tell you whether we are the right party for it, and we will say so if we are not.