Wi-Fi specialists — we survey it, design it, install it, and prove it works.
Home / Wi-Fi surveys / Validation surveys

Post-installation validation surveys

The survey almost everybody skips, and the only one that proves anything. We measure the installed network against the design and hand you the evidence — including where we're validating somebody else's work.

Why validation is the one that matters

Everything before installation is a prediction. The predictive model predicts, the on-site survey predicts more accurately, the design says what should happen once the access points are mounted and commissioned. A validation survey is the only part of the process that measures what did happen.

It matters because installation is where designs quietly degrade. An access point gets moved two metres because of a sprinkler head. One gets mounted above a suspended ceiling tile instead of below it. Transmit power is left on automatic and drifts to maximum. Two access points end up on the same channel. A cable run fails a test and gets patched rather than replaced. Individually, small. Together, the difference between the coverage you specified and the coverage you got.

Validation compares what was designed against what was built Design target 100% of the designated area at −65 dBm or better. Measured after installation below target AP-03 relocated on site Fail. Remedy: reposition AP-03 and re-test that zone. The gap between these two pictures is the entire reason a coverage guarantee is worth anything.
Installation is where designs quietly degrade. An access point moved two metres for a sprinkler head, one mounted above a ceiling tile instead of below it, transmit power left on automatic. Individually small; together, the difference between the coverage you specified and the coverage you got. Validation is the only step that measures it.

Almost nobody in this market offers validation as a distinct service, and that tells you something: it is the step that creates accountability. It is also what our coverage guarantee is measured against — a guarantee with no validation behind it is a slogan.

A 5 GHz signal-to-noise ratio plot of an office floor with a colour scale from 5 dB to 40 dB or better
Signal-to-noise is the number that decides whether the network works. Coverage tells you a radio can be heard; SNR tells you it can be understood over everything else in the band. We design to ≥ 25 dB and validate against it after the install.

What we measure and what you get

Measured against the design

  • Coverage at the design thresholds, per area and per device class — not one blanket figure
  • Signal-to-noise ratio across the site, which is where a marginal install shows up first
  • Channel plan and transmit power as commissioned, against the plan as designed
  • Co-channel contention between your own access points
  • Roaming behaviour along the real routes: aisles, corridors, stairwells, yards
  • Throughput and retry rates under representative load
  • Access point positions and mounting as installed, against the schedule

The output

A pass/fail against each design target, with plots showing exactly where and by how much. Where something falls short, a specific remedy — move this access point, change this channel, reduce power here, replace this cable — not a general recommendation to buy more equipment. If it all passes, you get a document saying so, which is a useful thing to hold at the end of a capital project.

Validating another contractor's work

We're regularly asked to validate networks we didn't install, usually in one of three situations: a client wants independent sign-off before releasing final payment; a project completed and the coverage still isn't right; or a building has changed hands and nobody knows what they've inherited.

We report factually. If the installation meets the specification, the report says so — we're not in the business of manufacturing faults to win remedial work, and a report that always finds problems is worth nothing to anybody. If it doesn't meet the specification, you get the measurements, the gap, and what it would take to close it.

Worth knowingValidation is far cheaper before final payment than after. If you're commissioning a Wi-Fi installation from anybody — including us — write a validation survey into the contract as a condition of sign-off. It costs a fraction of the project and it changes how carefully the work gets done.
FAQ

Common questions

When should a validation survey happen?

After installation and commissioning are complete, and before final sign-off or final payment. On a phased project, at the end of each phase — finding a systemic problem in phase one is much cheaper than finding it in phase four.

Is it included if you do the installation?

Yes. Where we survey, design and install, the validation survey is part of the job and the evidence is handed over at completion. It's how we stand behind the coverage guarantee.

What happens if it fails?

If we installed it, we fix it — that's what the guarantee means. If somebody else installed it, you get a report identifying precisely what's short and what would resolve it, which you can take to your contractor. We're happy to do the remedial work, but you're under no obligation to use us for it.

Can you validate against a design we already have?

Yes. Send us the design, the specification or the tender document and we'll measure against it. If there's no written design — which is common — we'll agree sensible targets with you first, based on your device fleet and what the space is used for, so the test is fair to everybody.

How often should we re-validate?

After any material change: new racking, a mezzanine, a partition reconfiguration, a significant change in headcount or device fleet, or a firmware or platform upgrade. Otherwise an annual RF health check is a reasonable rhythm for a site where connectivity is operationally critical — it's included in our managed service.

Let's find out what your Wi-Fi is actually doing.

Book a survey and you get measured data, a design you can build to, and a number you can budget against — not an opinion.