# Wi-Fi Designed From a Survey, Not a Guess > Wi-Fi that drops in the back office usually comes down to guesswork at install time. Designing from a proper site survey puts coverage where it's actually needed. *Source: https://panoptic.ie/blog/wifi-designed-from-a-survey* # Wi-Fi designed from a survey, not a guess Wi-Fi that drops in the back office usually comes down to guesswork at install time. Designing from a proper site survey puts coverage where it's actually needed. David Griffith ยท 1 June 2026 ![A steel disc with concentric glass rings rising from it like a contour map](https://panoptic.ie/_astro/blog_wifi_designed_from_a_survey_49e4b7cd30_yg5jK.jpg) In short Wi-Fi that drops in the back office usually comes down to guesswork at install time. Designing from a proper site survey puts coverage where it is actually needed. Best practice is a hybrid: predictive modelling for the plan and budget, then an on-site survey to validate before anything is mounted. Wi-Fi that fades in the back office or the far meeting room almost always comes down to guesswork at install time. Access points went where it was convenient to run a cable, not where the coverage was needed. Designing from a survey fixes that, and it is considerably cheaper than discovering the problem after the ceilings are closed. ### Predictive and on-site are both needed There are two ways to survey, and the argument about which is better misses the point. Best practice uses both. **Predictive design** models the building from floor plans, wall materials and ceiling heights, then simulates coverage to propose access point positions. It is fast, it can be done before you have keys to the building, and it produces a defensible plan and budget. **On-site survey** means walking the building with measuring equipment and recording what the radio environment actually does. It captures the things a model cannot know: the plant room that turns out to be lined with steel, the neighbouring tenant saturating the same channels, the stud wall that is actually concrete. The sequence that works is predictive first for the plan, then on-site validation before final installation. The validation step is what moves access points, adds the one nobody predicted, and finds interference. Skipping it is where most disappointing installs come from. ### Coverage is not the target. Capacity is. The old question was "does the signal reach?" That is now the easy part. A single access point can cover a surprisingly large area with a usable signal and still fail completely, because thirty people are trying to use it at once. Modern design works backwards from device count and application. A warehouse with a dozen handheld scanners needs a different layout from an open-plan office with ninety laptops, ninety phones and a room full of video calls, even if the floor area is identical. As a rough planning figure, access points typically land every 30 to 50 feet depending on the environment, the client devices and the access point capability. That is a starting point for a survey, not a substitute for one. ### Wi-Fi 6E and Wi-Fi 7 change the arithmetic The 6 GHz band is the biggest change to Wi-Fi design in a decade, and it comes with a catch. More spectrum means far more non-overlapping channels, which is exactly what dense deployments have always been short of. But 6 GHz **propagates differently**: it attenuates more sharply through walls and floors than 5 GHz does. The practical consequence is that placement assumptions carried over from an older network will be wrong. A layout that gave solid 5 GHz coverage can leave 6 GHz holes in the same building. This is precisely why survey work has become more important with Wi-Fi 7, not less. ### What else the survey settles A proper survey produces more than access point positions: - **A channel plan**, so your own access points stop competing with each other. Self-interference is one of the most common causes of poor performance in networks that look well-equipped on paper. - **Power settings.** Turning every access point to maximum is a classic mistake: clients hear the access point but the access point cannot hear the clients, so the connection sticks and stalls rather than roaming cleanly. - **Cabling and power requirements**, which is where budgets get decided. The access points are rarely the expensive part. ### When to re-survey A survey is a snapshot of a building in a configuration. Re-survey when that changes: - Fit-out changes, new partitions, or taking on an extra floor or unit - A significant increase in device count or a shift to video-heavy working - Migrating to Wi-Fi 6E or Wi-Fi 7, where the 6 GHz behaviour makes the old design unreliable ### Design for what comes next Designing with headroom costs very little at install and saves a great deal later. Cabling to positions you do not populate yet, switch capacity above today's port count, and a channel plan that survives adding access points are all decisions made once. We wrote about that more fully in [a network with room to grow](https://panoptic.ie/blog/networks-with-room-to-grow). If your Wi-Fi has dead spots, or you are fitting out somewhere new and want it right the first time, [talk to us](https://panoptic.ie/contact) or read about [networking and Wi-Fi](https://panoptic.ie/services/networking). [All field notes](https://panoptic.ie/blog) ## Start with a free IT review. You'll get a clear picture of where you're exposed today, and what it looks like to have your whole IT covered as one managed system. No obligation. [Talk to us ](https://panoptic.ie/contact)[See how we onboard](https://panoptic.ie/how-we-onboard)