Access control is usually sold as a system. It is more useful to think of it as a list: who may go where, kept current. Get the list right and the hardware choice becomes obvious. Get it wrong and no system will save you. Reading about access control on an office assumes you have time; the locksmith near me page assumes you do not.
Access control replaces keys with credentials you can grant and revoke in seconds. Here's the primer.
Credentials
Cards and tags are cheapest. PINs are easy but shared. Biometrics are strongest but require care with staff data.
Controllers
Standalone controllers per door are fine for small offices. Networked controllers unlock audit trails and remote management.
Integration
Access control ties naturally into CCTV, alarms and time-and-attendance systems.
Start with the list, not the hardware
Take a sheet of paper and list every way into and through the building, then put names beside each one. Front entrance, back door, yard gate, stockroom, server cupboard, the office nobody locks, the fire stair. Two facts fall out of that exercise immediately in almost every business: several people can reach places their role has nothing to do with, and at least one opening has a key count nobody could state under oath.
Half an hour spent on that list buys more than the first substantial invoice from any supplier, because the list is also the specification. Once you can see the pattern of who needs what, the number of tiers required becomes self-evident, and it is nearly always small: a way in that several people share, working areas open to fewer, and one or two rooms restricted to a couple of individuals.
Its usefulness lies entirely in being current, which means one line added whenever something goes out and one struck through whenever something returns. Filed with the company records rather than carried in a head, it outlives the person who created it — and outliving its author is the single hardest requirement any arrangement of this kind has to meet.
Keys, codes and cards each fail differently
Mechanical keys are reliable, cheap and completely indifferent to power. Their weakness is that a key that leaves does not announce itself, and withdrawing one means changing a cylinder rather than pressing a button. On a premises where the same six people have worked for years, that is a perfectly good arrangement and there is no reason to replace it. As a Randburg locksmith, access control on an office is something we are called to more weeks than not.
Codes are convenient and quietly leak. A door code shared with a cleaner in 2021 is now known to an unknowable number of people, because codes are told rather than handed over, and telling leaves no trace. If you use them, they need changing on a schedule — and a business that has never changed a door code is effectively running an open door with an extra step.
Cards and fobs solve the withdrawal problem properly: one can be cancelled the day somebody leaves, without touching any hardware. Their weakness is dependence on power and on somebody actually doing the cancelling. A system where nobody has removed a leaver's card in two years has all of the cost of access control and none of the benefit.
The leaver problem is the whole game
Every access arrangement in a small business fails at the same point, and it is not a technical one. Somebody leaves, everybody is busy, and nothing gets withdrawn. Six months later nobody is sure whether that person still has a way in, and the honest answer is that nobody knows. This is not carelessness so much as the absence of a step in a process that never had one.
The fix is to attach it to something that already happens. A leaver hands back a laptop, a uniform, a parking disc — add keys and credentials to that same moment, and record it on the list. Where a key has genuinely gone missing, the answer is a lock change rather than a series of awkward phone calls, and it says nothing about the person who left.
It is worth being unsentimental about this. The overwhelming majority of former staff have no interest whatsoever in your stockroom. The point is not suspicion; it is that a business which cannot state who can enter its premises has lost control of something it is responsible for, and that matters for reasons that have nothing to do with anybody's honesty.
Getting out matters as much as getting in
Leaving a workplace is a legal matter as much as a practical one, and it is where poorly considered access arrangements do genuine damage. Every door that has people behind it during working hours has to give way from the inside on a push, with nothing to find and nothing to remember. A bolted rear door at four in the afternoon is a problem no quality of bolt improves. For the equivalent detail a suburb away, see locksmith work in Ferndale.
That obligation dictates what gets fitted. Push-bar devices, thumb turns and the related exit ironmongery exist to keep the two directions independent of each other, so that resisting arrival costs nothing in departure. A specification that never separates those two directions was drawn up by somebody picturing the building empty.
It dictates a habit as well. Because these fittings are almost never used in anger, they degrade without announcing it, and the failure is discovered on the one morning it must not be. Pushing every exit open once each quarter costs a few minutes and is the entire maintenance requirement.
Hardware has to survive the duty cycle
Offices wear their doors differently from homes and much faster. An entrance serving thirty people who go out for coffee, take deliveries and step outside to take calls sees more movement before lunch than a household front door manages in a fortnight. Hardware chosen from a domestic range will therefore reach the end of its life years earlier than anyone expects, and the surprise is entirely a function of the volume rather than the product.
So specify from the technical figure rather than the catalogue photograph, and diarise the replacement instead of waiting to be surprised by it. An early morning before anybody arrives is a cheap hour to change an entrance mechanism; the same work at half past nine on a Monday with thirty people trying to get in is a different proposition entirely, and it costs accordingly. There is a separate article on shopfront security basics if that is closer to the problem in front of you.
Everything else on the premises follows the same logic in proportion to how often it is touched: the kitchen door, the meeting-room lock nobody uses, the server cupboard, the door onto the fire stair. Some of those will outlive the lease and some will not, and knowing which is which is a five-minute walk. Where something is worn rather than broken, an ordinary lock repair buys years; where it has genuinely finished, replacing it before it strands somebody is cheaper than the alternative.
A sensible starting arrangement for a small premises
Three levels, written down, and a leaver step attached to something that already happens. That is the whole recommendation for most small businesses in Randburg, and it can usually be set up in a single visit rather than as a project. It does not require an integrated system, a subscription, or anything that stops working when the power does.
Add to it only where there is a specific problem to solve. Cards make sense where people come and go often. Controlled-duplication keys make sense where copying is the worry. A code makes sense on a door used by a rotating group who cannot reasonably be issued anything. Each of those is a solution to a named problem rather than an upgrade for its own sake.
And keep the record with the business papers, not in a drawer at one address. Business premises around North Riding and the trading strips through the central suburbs run on exactly this basis, and the ones that do not are usually the ones ringing about a stockroom nobody can open because the only key left with somebody two managers ago.
Reviewing it once a year is what makes it work
An access arrangement is not a purchase; it is something that decays quietly unless somebody looks at it. Once a year, take the list out and read it. Is every name on it still someone who works here? Is every opening still used for what it says? Has anything been added — a new store, a partitioned office, a gate that was put in during building work and never entered on the sheet?
That review usually takes half an hour and it regularly turns up two or three things nobody had noticed. A credential still active for somebody who left in March. A key that was cut for a contractor and never came back. An opening that changed use — a store that became a stockroom, a passage that became a fire exit — without the hardware being reconsidered. For a different angle on this, the piece on a guide to panic bars is worth a few minutes.
Attach the review to something already in the calendar so it actually happens: the insurance renewal, the stocktake, the start of a financial year. Businesses that do this have an access arrangement; businesses that do not have a system they bought once and a list of names that stopped being accurate about eighteen months ago.
What to do in the first week
Write the list before speaking to any supplier. Every opening down one side, every person down the other, and a mark where the two meet. Half an hour produces the specification for free, and it almost always reveals two things at once: that several people can reach places their role has nothing to do with, and that at least one opening has a key count nobody could state under oath.
Then decide the tiers, which for most small premises means three. An entrance several people can open, working areas open to fewer, and one or two rooms restricted to a couple of individuals. More than that usually means somebody has designed for a business twice your size, and it will be abandoned within two years because nobody can maintain it.
Only then choose the mechanism. Keys are cheap and indifferent to power but cannot be withdrawn without changing a cylinder. Cards can be cancelled the day somebody leaves and depend on power and on somebody actually doing the cancelling. Codes are convenient and leak. Each is right somewhere, and a supplier who recommends one before seeing your list is selling rather than advising — which is commercial locksmith work done properly.