The 15% Guard Force Cut: Defending a Security Budget with Evidence
You are the newly hired security director for a 1,200-employee, three-building suburban campus. The contract guard force is two 24/7 posts billed at $58/hour - roughly $1.0 million annually - and the incumbent has proposed a 9% increase at renewal. Finance has directed you to cut contract security spending by 15% next fiscal year. Last year's incident log shows 12 vehicle break-ins, 3 storage-shed burglaries, and two after-hours trespassing arrests - all in the west parking lot, which has one unmonitored gate. An integrator quotes $340,000 for gate readers, 22 cameras, and video analytics, plus $45,000/year maintenance; the design would let you eliminate one patrol post. Badge data shows 140 second-shift employees use the west gate nightly.
1. What analysis must you complete before agreeing to - or resisting - the 15% reduction? What data would you demand, and why?
Read the worked answer
Start by converting the budget into a unit basis: two 24/7 posts = 17,520 billed hours/year; at $58/hour that is $1.02M, plus the proposed 9% ($1.09M). Now you can price any option instead of arguing about a lump sum.
Next, test whether the spend matches the risk. Map every incident by location, time, and loss value. Twelve vehicle break-ins plus three burglaries may total only a few thousand dollars in direct loss - but the guards also provide deterrence, response, escorts, and after-hours coverage that no incident log captures. Do not let a low loss total alone justify removing posts.
Demand: (1) the contract's scope, KPIs, penalty clauses, and notice period; (2) insurance and lease requirements that may mandate staffing or patrols; (3) alarm and access-control data for the west gate; (4) call-for-service and response-time records; (5) the integrator's throughput, false-alarm, and maintenance-escalation assumptions.
The analysis should show the risk is concentrated at one unmonitored gate - a controllable variable - not uniformly across the campus. That reframes the conversation from 'cut guards' to 'reduce exposure at the west gate.'
2. Finance insists on the 15% cut. What options do you present, and how do you structure the recommendation?
Read the worked answer
Present at least three costed options with risk consequences, not one number.
Option A - Technology plus one post: install the gate and camera package, retain one 24/7 post, replace the eliminated post with scheduled mobile patrol. Three-year cost: $340,000 + 3 × $45,000 = $475,000, against roughly $509,000 saved by removing one post - roughly break-even in year three, with better coverage of the actual loss location. Savings exceed 15% in year one only if the capital is amortized.
Option B - Re-scope posts: convert one 24/7 post to 16 hours/5 days where badge data shows low demand, and negotiate rate, overtime, and holiday billing with the incumbent. Smaller savings, no capital risk.
Option C - Defer: accept a 5–7% reduction now, pilot technology for 90 days, and commit to the larger cut only after the pilot data supports it.
Structure the memo as: risk statement, option, cost (including total cost of ownership), residual risk, and decision requested. Note the contract notice period and any service-level penalties before promising a date. A phased cut you can defend beats a cut that creates an incident you cannot explain.
3. What metrics would you track to prove the new model is working - and what would tell you to reverse it?
Read the worked answer
Track measures tied to the risk you claimed to reduce, not activity for its own sake:
- Incidents per 1,000 badge-holders, segmented by west gate vs. other areas
- Response time to alarms and calls, from dispatch to arrival
- Alarm and analytics false-alarm rate (a high rate degrades guard trust and response)
- Tailgating/piggybacking detections at the gate
- Guard utilization: percentage of shift spent on patrols vs. static posts
- Cost per protected asset or per square foot, for the CFO
- Audit compliance: patrol checkpoints completed, logs filed
Predefine thresholds before launch. For example: if west-lot incidents do not fall within two quarters, or false alarms exceed an agreed rate, the pilot fails and the post is restored. Publishing reversal criteria in advance protects your credibility and makes the pilot a genuine test rather than a cost-cutting exercise with security's name on it.