Contract Guards, Proprietary Officers, or Technology
You are the security manager for a 1,200-employee food-processing plant running three shifts. The contract guard force costs $1.1M a year for two officers around the clock plus a day-shift supervisor. The vendor's officer turnover is 78%, and in the past year there were two documented failures: a gate log with a six-hour gap and a missed 2 a.m. patrol. The plant manager wants proprietary officers at $1.6M including benefits. A systems integrator quotes $900K capital for license-plate readers, video analytics and employee turnstiles, which would let you cut to one 24/7 post at $650K a year. The CFO says any capital project needs an 18-month payback and asks why the current contract cannot simply be tightened. A customer food-safety and security audit is scheduled in four months.
1. What must be in the comparison before you recommend anything?
Read the worked answer
Compare total cost of ownership over at least three years, not year one. The contract option is $1.1M plus the internal cost of managing the vendor, plus a quality risk you can price: 78% turnover means the officer at your gate has weeks of experience, and the two documented failures are the visible samples of a larger population.
Proprietary at $1.6M buys continuity but also brings recruitment burden, benefits inflation, pension or union exposure, overtime for call-offs, and much less flexibility to flex down. Quantify the coverage risk - who fills a post at 0300 when an officer calls out?
Technology: $900K capital plus maintenance at roughly 10-15% annually, plus $650K opex, gives about $1.0M in year one and $750-800K thereafter. Against the $1.1M baseline that is a payback nearer 30-36 months than 18. Say so plainly.
Then the non-financial column: audit findings, deterrence value, use-of-force and vicarious liability exposure, evidence quality, and whether one post can plausibly do gate control, patrol and response. It cannot do all three well, and the model should show that gap rather than bury it.
2. How do you answer the CFO's question - why not simply tighten the contract?
Read the worked answer
Agree that it is the cheapest first move and pursue it, but explain its ceiling.
Tightening means revised post orders, SLAs with liquidated damages, turnover reporting, wage transparency, audit rights, monthly KPI review, a named account manager and a cure period with termination for cause. All of that is free and should be done in the next 30 days.
The honest caveat: 78% turnover is a wage economics problem. If the bill rate supports only a near-minimum wage, the vendor cannot recruit or retain better officers no matter what the SLA says - you will document the same failures with better paperwork. Ask for the wage build-up, and if it is thin, propose a modest rate increase tied to a retention metric so the SLA has something to enforce.
The audit in four months changes the sequence. You need visible, documented improvement now - revised post orders, a functioning gate log, patrol verification - not a capital project that goes live after the auditor leaves. Present it as sequence, not as a contest between options.
3. What is your recommendation and what conditions attach to it?
Read the worked answer
Phased recommendation, presented with honest payback:
- Immediate (0-60 days): contract amendment with KPIs, liquidated damages, wage transparency and a retention-linked rate adjustment; post orders rewritten; patrol verification technology at low cost.
- Pilot (months 2-9): deploy turnstiles and ALPR at the main gate only, and measure whether the second post's workload is genuinely absorbed. Hold the third post during the pilot - do not cut headcount on the strength of a vendor's model.
- Decide at month 12 on full rollout, with the audit results and pilot metrics in hand.
Conditions: no reduction in post coverage until the pilot proves response times and patrol completion hold; a transition plan with a 60-day contingency if the integrator slips; and a documented statement to the CFO that 18-month payback is not achievable on these numbers, with the risk-adjusted figure instead. Promising 18 months to win approval buys a credibility problem in year two.