Two Dallas-Fort Worth properties hire the same security company in the same month at the same rate. Six months later one property manager renews without hesitating and the other is collecting quotes from competitors. The guards came from the same roster and went through the same training. The difference, more often than not, is that somebody wrote proper post orders for one site and nobody did for the other.
Post orders are the written instructions that define what an officer does on your specific property. Not security in the abstract: your gates, your hours, your escalation contacts, your report format, your rules about what happens when somebody will not leave. They are the document that converts a person standing on your property into coverage you can describe, audit, and rely on. A security contract without them is a contract to provide a person, which is a different and much weaker thing than a contract to provide security.
What actually goes in them
A usable set of post orders is specific enough that an officer who has never been to your site could work a shift competently after reading it. That bar sounds low and most post orders fail it.
They start with the site itself: a description of the property, its boundaries, the access points and which ones are supposed to be locked when, where lighting is weak, where the known problem areas are, and where the officer is expected to be physically positioned. If the post is mobile rather than fixed, this becomes patrol routes and frequency, including which areas get checked every round and which get checked once a shift.
They define access rules in operational terms. Who is permitted on the property after hours. How residents, tenants, employees, and guests are identified. What documentation a vendor or contractor needs. What happens when somebody is not on the list but insists they are expected. That last scenario is where untrained coverage fails most predictably, because an officer with no written rule will either wave through someone who should have been stopped or create a confrontation with someone who had every right to be there.
They specify the escalation tree. Who the officer calls, in what order, for which categories of event, and what the thresholds are. A medical emergency, a trespasser who leaves when asked, a trespasser who does not, a suspected break-in, a tenant dispute, a fire alarm, a power failure: each should have a stated response and a stated contact. Real escalation trees include cell numbers and a named backup for when the first contact does not answer at three in the morning.
They establish the reporting standard. What gets logged on every shift regardless of whether anything happened, what triggers an incident report, whether photographs are expected, what format the report takes, and when it lands in your inbox. A property that receives a consistent daily activity report has a record. A property that hears from its vendor only when something goes wrong has nothing.
And they state the limits. What the officer does not do. Whether they are authorized to detain, under what narrow circumstances, and the standing instruction to call law enforcement instead. Whether they handle maintenance issues or only report them. Whether they enforce parking, and with what authority. Clear prohibitions protect the officer, the property, and the vendor simultaneously, and their absence is how a well-meaning guard creates a liability problem.
Why service degrades without them
The mechanism is turnover, and it is worth being blunt about it. Security has meaningful staffing churn across the industry, and the officer standing at your gate in March may not be the one standing there in September. That is a normal operational reality rather than an indictment of any particular company.
This matters most on a 24/7 coverage contract, where a single post is staffed by a rotating roster of four to six officers and no one person is ever the standard. When post orders exist, turnover is an inconvenience. The new officer reads the document, walks the site with a supervisor, and performs more or less like the last one. The property manager may not notice the change at all, which is the goal.
When post orders do not exist, institutional knowledge lives entirely in the head of whoever happens to be working the post. The procedures that officer developed, the judgment calls they made, the people they learned to recognize, the manager's unwritten preferences: all of it leaves with them. The replacement starts from zero and invents their own version. Service quality becomes a function of which individual drew your site, which is exactly the inconsistency that makes property managers conclude the vendor has gone downhill.
This is also why the complaint pattern is so recognizable. Service was good at first and got worse. The first officer was great and the new one is not. Nobody told the new guy how we do things here. Every one of those is a post orders failure wearing a staffing costume.
Who writes them, and who has to approve them
In a functioning arrangement the vendor drafts and the client approves. The security company brings the structure, the operational language, and the experience of what tends to go wrong at properties of this type. The property manager brings the things only they know: which tenant has a difficult history, which contractor shows up unannounced, which neighbor complains about patrol headlights, which door the staff props open despite being told not to.
That approval step matters more than it appears. Post orders written entirely by a vendor without client input tend to be generic templates with the property name swapped in, which is the same as having none. Post orders written entirely by a client without vendor input tend to include instructions that are impractical, unsafe, or outside what a licensed officer can lawfully do. The useful version is a conversation that ends with a document both sides have actually read.
A site walkthrough before coverage begins is what makes that conversation productive. Post orders drafted from a floor plan and a phone call miss the gate that does not latch, the blind corner behind the dumpster enclosure, and the fact that the east entrance is unusable after nine because of how the lighting falls.
They go stale, and nobody notices
The most common failure is not an absent document but an obsolete one. Post orders written two years ago still reference a contact who left the company, a gate code that changed, a tenant who moved out, and a camera that has not worked since last spring. An officer following them faithfully is following instructions for a property that no longer exists.
A twice-yearly review is enough for most properties. Contacts and phone numbers, access codes and credential systems, changes to the physical site, changes to the tenant mix, and anything in the last six months of incident reports that suggests the current instructions do not fit what is actually happening. If your incident reports keep describing the same situation and the post orders do not mention it, the post orders are behind.
What to ask a vendor before you sign
A few questions separate companies that run real programs from companies that fill shifts. Will you write site-specific post orders for this property, and can I see a redacted example from a comparable site? Will you walk the property before coverage starts? Who approves the final version, and does my approval appear on it? How do you brief a replacement officer, and does a supervisor walk the site with them? What does the daily report look like, and can I see one? How often do we review and update the orders?
A vendor who treats those as reasonable questions is running a program. A vendor who treats them as friction is selling you a person in a uniform and hoping you do not notice the difference until the renewal.
