Nate’s Notes
I'm presenting at GSX in Atlanta this September, and this week's issue is tied to it.
The session is Beyond Response Times: Numbers That Matter to Executives, Wednesday, September 16, from 9:45 to 10:45 in the morning. If you'll be there, come find me and say hello.
I've spent a good part of this year building that talk, and it grew out of something that's bugged me since my dispatch days. Response time is the number everyone reaches for, and it's the right one when you're running a shift, but it does you almost no good the day you have to convince the people who hold the budget to fund what sits behind it. Those people don't think in stopwatches, they think in dollars, coverage, and risk they can point to.
So Kenny Kelley's recent piece in Security Magazine, Why Response Time Is Becoming the Missing Metric in Workplace Safety and Security, landed right in the middle of what I've been working on. He's right about most of it. This week's Main Article is where I pick up the part he stops short of, and it's a preview of the argument I'll be making on stage in September.
Let's get into it.
Main Article
Response times are critical, but they won’t fix your program
I read a piece by Kenny Kelley in Security Magazine last week arguing that response time is the metric workplace safety keeps forgetting to measure, and he's right.
Having started my career in EMS and emergency dispatch, I've spent a lot of time concerned with response times.
When someone hits a button, the minutes until help arrives are the whole story for that person, so if your safety program tracks training hours and audit scores and has never once asked how fast help would actually reach someone, Kelley has your number. For a worker alone on a dock or a floor nobody walks for an hour, a device to quickly summon help can be the difference between life and death.
I want to pick up where the article stops, because the button does one specific thing and it's easy to mistake that thing for the whole job.
Response time is every second from the moment the incident happens to the moment help arrives, and that's a lot more than most people picture. Anybody who has worked a console knows it comes in pieces, each with its own standard. In my 911 center the calltaker owned intake and the initial processing together, working a location and a story out of whoever was on the line while building the incident in the CAD, and the standard to get that incident built was 90 seconds. Only then did the caller and the computer incident move over to the dispatch desk, and even when the dispatcher turned it around on the spot that was another 10 seconds, and once you count the phone calls and the transfers you're nearly two minutes into an emergent call with nobody sent yet. Two minutes gone and the unit response hasn't even started. Critically, we are still in compliance with industry and national standards.
That two minutes isn't the system failing, it's the system working the way it's built to, and what the button does is make sure the clock starts as early as it can, closing the gap between something going wrong and anyone knowing it did. For a lone worker that gap can be the whole emergency, so the device is well worth buying. How much of that value you actually capture depends on what the alert lands in. Turning a fast alert into fast help still takes a trained person with the time to sort out whether it's a medical, a fight, a fall, or a phone that got sat on, because a ping and a dot on a map don't tell you much, and it takes the judgment to code it right, since a bad code sends too much to a false alarm or too little to something real. Processing, dispatch, and the response are all capped by who you actually have, so on a minimum-staffing night the fastest alert in the world still waits in line behind two other calls. The button gets the alert to the front faster, and how much that's worth comes down to whether the operation behind it can pick that speed up and run with it.
If you're running a program and not just a shift, the button is the part of this you can buy outright, and the return on it comes from the parts you can't. Staffing, capacity, and design, you don't buy those, you fund them. They're also invisible on a response-time chart, because response time is a single average that quietly hides which phase is the one coming apart. A clean four-minute number can be four minutes of sharp dispatch or four minutes of an overloaded operator miscoding calls, and from the outside the two look the same.
So response time, as much as it earns its place on the floor, does almost nothing for the director who has to go fix the rest of that sequence. Fixing it takes money, and money means walking into a budget review with people who have forty other requests in front of them. Put response time on that slide and it dies, not because it's untrue but because it's the wrong language for the room. Nobody up there is asking whether your officers are fast. They're asking what the return is, whether it clears the hurdle rate, how it stacks up against every other capital request this year, and what it costs the organization to say no.
The cleanest way I know to explain the mismatch is the old functions of management taught in business school.
Planning, setting the goal and deciding what you're going to do
Organizing, lining up the people and the dollars to do it
Leading, directing the work once it's moving
Controlling, checking the operation against a standard and correcting the drift
Response time is a controlling number, and a good one. It catches the drift so you can correct it, and it sits about as far from the checkbook as a metric can. Starting a project isn't controlling work, it's planning and organizing, and carrying response time into that conversation is carrying a stopwatch into a decision that runs on forecasts. It's a good number doing the wrong job.
What gets me is that the numbers that do belong in that room are already sitting in your operation, most of them just filed under somebody else's name.
Overtime, already on the CFO's desk, where a spike is a coverage and attrition story the moment you read it that way
How often you run at minimum staffing, which tells leadership whether the model is funded for the real workload or only for the look of it
Risk-weighted staffing off your hazard assessment and your actual incident volume, which earns more than a square-footage benchmark ever will
A specific intervention tied to a specific outcome, fewer assaults after you added a post, less shrink after you moved people around, which proves the last dollar worked so they believe you on the next one
Response time won't build or expand your program. It's an excellent way to know whether you're inside the standard, and that's exactly where it stops, because the people who hold the budget already expect your response times to be within standard. You don't get funded for clearing a bar they assumed you cleared a long time ago.
Kelley's right that response matters and that too many programs never look at it. I'd only add that the button gets help called fast, and how much that speed is worth depends on whether help is actually there to give. Getting it there is a budget problem before it's a technology problem, and closing that gap is most of what I build the command-center standards around, and the heart of what I'll be talking through at GSX this fall.
Touring
Close the loop
A tenant stops an officer in the lobby to report that somebody keeps parking in the fire lane. The officer nods, maybe keys it into the radio, and the tenant never hears another word about it. The burnt-out light, the door that sticks, all of it gets reported once and then disappears, and the tenant learns that talking to security goes nowhere. So the next thing they notice, they keep to themselves.
Closing the loop fixes that. Every request an officer takes gets logged, and somebody gets back to the person who raised it once it is handled. It costs a logbook and a habit. What matters is how you close it, because a reply that says nothing is barely better than silence.
Two ways work. The first is a written response that says more than thanks. A tenant who gets,"Thanks for letting us know" assumes the report went in a drawer. Give them the detail instead. Tell them you talked to the driver in the fire lane, moved him along, and asked the garage to watch for the plate, and now they know a person read it and did something. The detail is the proof, and the proof is what earns you the next report.
The second, and usually the stronger one, is to handle it in person. The officer who took the complaint, or whoever walks that tour next, finds the tenant on the next round and closes it out face to face. The tenant gets a real answer, the officer becomes a familiar face instead of a uniform, and the next time something looks wrong that tenant comes looking for them by name.
The payoff is the same either way. A tenant who gets a real answer keeps reporting, and a building full of people who report is the widest and cheapest detection layer you will ever field. No camera covers the ground that a few hundred residents do once they trust that security acts on what they say.
Two numbers tell you it is working. Callback completion, meaning how many logged requests actually got a reply, and report volume over time, which climbs as the building learns that security listens.
Signals
Christmas in July… for cargo thieves
Cargo theft climbs around the holidays, and CargoNet flagged the July 4 weekend as one of the worst windows of the year. Two things make a long weekend attractive to the people who steal freight for a living.
The first is staffing. Warehouses and yards run short over a holiday, supervisors are off, and the calls you would make to verify a pickup go to voicemail, so a driver who should have been questioned gets waved through and a load that goes missing has two or three days to travel before anyone is back at a desk to notice.
The second is that much of modern cargo theft is fraud rather than a break-in. Thieves set up fake trucking companies, or steal the identity of a real one, book a legitimate load online, and show up with paperwork that looks correct. The freight gets handed over willingly and driven away, and nobody realizes it was stolen until the real carrier never arrives to pick it up. A guard watching for someone climbing a fence is never going to see that happen.
For an operator the takeaway is simple. A holiday is exactly when you should not be running thin. Keep your coverage up when the calendar is tempting you to cut it, confirm every pickup with the broker directly instead of trusting the documents in the driver's hand, and hold any load that does not match a confirmed appointment. Put your most experienced people on the gate over a long weekend, not your newest.
None of this takes new technology. The truck that drives off with clean paperwork gets stopped by a phone call to the broker, not a better camera, and that call is the easiest thing to skip on the one weekend it matters most. The next holiday is already on the calendar, and the crews planning to take your freight are counting on you to treat it like an ordinary day.
Passdown
dictated by Ricky Portezzo, Senior Security Supervisor in Center City, Philadelphia
Friday night shift. Slow one. Somebody left the good flashlight in the other truck again so I had the little one that barely works.
So Mikey's new thing is we call people back now when they complain about something. Log it, then actually get back to them. I told him Mikey I got five sites to cover, now I'm the complaint department too?
Couple weeks back the cashier at the corner store, been there longer than me practically, she tells Donna the light over the back door is out and she don't like locking up in the dark back there. Normally that goes nowhere, but we got the new rule so Donna writes it up and I call the property people. Then I go by and tell the cashier its getting fixed next week. You would think I handed her a lottery ticket. Nobody called her back about nothin' for twenty years.
Then last night she calls me first. "Ricky, there's a guy been in three nights straight, comes in, buys a pack of gum he don't want, and just stands by the register looking at the drawer a good ten minutes before he leaves and he’s loitering outside." So I roll up and park right out front and mister Three Nights takes one look and keeps walking, never even walks in.
So fine. The calling back thing. It works. Do not tell Mikey I said that.
Employees can be security's greatest resource, but only if they believe their reports go somewhere.
SCC Spectrum Security Almanacs. www.gsoc911.com/products