I believe this is something that many in the industry will read and say “well thanks, Captain Obvious”…but if you are starting out, not everything is so obvious to you just yet. The importance of walking a site may just be on that list of things that could be taken for granted, and I have a few scenarios underscoring why.
First and foremost, you need to know your space. I totally understand that as much as companies want good WiFi, many of them also prefer it cheap. Cheap is, of course, a relative term, but often a customer will seek to save money wherever possible. When it comes to WiFi deployments, there are numerous ways that they will seek to save money: avoiding external antennas, scaling back on the number of APs, even limiting time and travel. It’s that last item that we can dive into here.
I’ve seen projects where, in order to save time and money, and as a means of expediting things, customers pushed to avoid the more lengthy on site surveys (i.e., AP on a stick). While a predictive design is still a useful exercise, it’s always wise to walk the space prior to committing to AP placement. Here are some reasons why.
What can go where?
This one is perhaps the most obvious, but it is important. If you’ve never walked the site, if you’ve never been there in person, then walking it is really important. Why? Well, if you just take the floor plan and drop a bunch of APs onto it, it will look fine. But here’s the thing-if you don’t walk around and check out the ceilings, you will miss things. That really nice spot you picked for the AP? It could have a camera or an HVAC vent right there. Just because you want to put it there, doesn’t mean it can go there. Floor plans often times won’t show you those items.
How can it get there?
On top of that, depending on the facility, you also need to account for access concerns-not in terms of getting into the facility, but in terms of ceilings and walls. As in, how hard might it be to pull a cable to that spot. One section of ceiling could be drywall, but just a foot over to either side could be a perfectly accessible drop ceiling. Sometimes that wall you want to run cables up is a wall of glass, making that cable run impossible. Even if you aren’t getting the time and budget to perform an APoS survey, getting a day on site could be tremendously helpful.
Am I even allowed to put it there?
There’s a few ways to take this one.
Part of that is tied to what I said before-you need to make sure you aren’t dropping an AP on top of a security camera or other system.
The other way to take it? Aesthetics. Some customers will not care about seeing the AP, because they want their WiFi to work. Other customers will, however, care a great deal. They will ask you to conceal some, or even all, your APs. This is a case where you will need to walk the area with them to understand which APs need to be out of sight, and what your options are. You do not want to find out, many hours into your design efforts, that half of your APs have to move or change installation methods.
How many APs do you really need there?
There are a few reasons for this one.
First, guess what? Sometimes the floor plan you get is already out of date. What shows up as four offices and a conference room? It may have become 24 cubicles. If the space was all walls and offices, you would probably design it differently than say, a wide open agile environment with 24 cubicles. Sometimes your customer will tell you the plans aren’t 100% accurate. Other times…well…surprise!
Second, sometimes you plot out the APs in a predictive in say, Ekahau, and it looks great. But then you walk the area and you realize that the APs are way too close, or there are not enough users frequenting that area to justify a more dense deployment. It’s not always possible to do this, but if you happen to be close by (or, if they brought you on site), perhaps you can walk the site after you’ve done your predictive design. Take your heat maps and see where your proposed APs would go. See if their placement makes sense. This falls in line with the idea of “measure twice, cut once”.
If you understand what the space is used for, and by how many people, you can figure out how many APs may be required, and do so with some degree of confidence. If you’ve physically walked the office, you can better understand what can, or cannot, go where. Customers may think such a site visit is a waste of project resources, but the truth is it’s likely to be money well spent, considering how costly it can be to fix a bad WiFi deployment after the fact.
Wifi John 