Deciding What to Outsource in Building Operations

Outsourcing building services is not one decision. It is a set of decisions about which parts of a service you hand to someone else, and which parts stay with you whatever the contract says. Managers who blur that line usually find they have outsourced the work and kept the blame.

Every service has six components: delivery, supervision, evidence of quality, exception response, compliance records and cost risk. Contracts move delivery first. The rest move only when you deliberately hand them over, and some should not move at all.

Key takeaways

• Outsource delivery, not accountability. The standard, the record and the answer to the owner stay with you.
• Decide service by service, on four tests: capability, variability, criticality and controllability.
• A task you cannot specify with a frequency, an area and a measurable output is not ready to be outsourced.
• Routine periodic work carries well. Judgement work — complaint triage, contractor supervision, emergency coordination — is where outsourcing goes wrong.
• Compare fully loaded in-house cost against contract price plus your own management and audit hours, never wage rate against quote.
• Regulated specialist work is always contracted. Verifying it is still yours.

What is the actual decision you are making?

It is a decision about control, not about staffing. The question is not "employees or a vendor" but which of the six components above you are willing to lose visibility of, and what you will do to see it again.

That reframing changes the analysis. A cleaning contract that costs less and removes your ability to prove condition to an owner has not saved money; it has moved a reporting obligation onto you with no evidence behind it. The same logic applies in reverse: keeping a small in-house team to service a building you visit twice a month can be the most expensive way to buy certainty. Deciding it well is part of property management operations, not a procurement exercise bolted on afterwards.

Two failure modes dominate. The first is outsourcing a service you could never measure, because the blind spot transfers to the vendor and neither side can settle a dispute. The second is insourcing work that needs a staffing depth you cannot keep: one trained person and no cover.

The four tests that decide most cases

Score a service across all four before you write a scope. A service that scores badly on controllability and high on criticality is the one to keep close, even if the price difference looks attractive.

Where should outsourcing building services usually stop?

Delivery of routine, frequency-defined work outsources well: periodic deep cleans, carpet extraction, hard-floor restoration, window cleaning and high dusting.

Judgement work does not. Complaint triage, the decision to escalate a water ingress at 11 pm, supervision of a specialist trade on a roof, and the call on whether a condition is normal or a defect — these depend on context a visiting vendor does not hold. Keep the decision, and outsource only the task the decision produces.

Safety-critical control is the same. Fire and life-safety systems, emergency lighting and any hazard control imposed by an insurer are delivered by specialists and verified by you or your consultant, never inferred from a clean invoice.

What can you contract but never delegate?

The work, not the responsibility. CCOHS states plainly in its health and safety programme elements that while some responsibilities can be delegated, the supervisor remains accountable for seeing that they are carried out. The same structure applies to a manager and a vendor: you can buy the activity, and you still own the outcome.

In practice three things stay on your side of the line, every time:

• The standard. You define what "clean" means, at what frequency, in which areas — an APPA 1–5 level per area is the usual currency.
• The evidence. Photos, dated inspection records and deficiency history must live in a system you can export, not only in the vendor's portal.
• The response obligation. Who answers at 2 am, in what time, with what authority to commit money.

Regulated specialist work deserves a note: asbestos-containing materials, confined spaces, electrical work, refrigeration and pest treatment are commissioned, not managed hands-on. Confirm the scope in writing, verify credentials and records, and report rather than act.

What does a service need before it is ready to outsource?

Five items. If any is missing, the scope is not ready, and this is what separates a tender from a wish.

1. Area definition — named zones or rooms, not "all common areas".
2. Frequency — daily, weekly, monthly, quarterly, or trigger-based (after an event).
3. Standard — the level expected in each area, expressed so two people would score it the same way.
4. Evidence — what the vendor hands back: photos per completed item, a signed schedule, exception reports.
5. Response window — the time to attend and the time to resolve, by priority.

A service missing two or more of those items will be argued about within a quarter. Sort the specification before you sort the vendor list; the sequence in writing a cleaning specification is the one to follow, and no tender evaluation is worth much until the scope is firm.

How do you compare in-house cost with a contract honestly?

Put the same line items on both sides. Wage rate against contract price is not a comparison; it is a rounding error with a story attached.

The honest comparison usually narrows the gap in both directions. In-house looks cheaper than its real cost because supervision and cover are hidden in salaried hours. Contracted looks cheaper than its real cost because audit time and transition risk are not quoted.

What should the contract say about verification?

Four clauses decide whether you can hold the service to a standard, and all four are routinely left out.

• Access and audit rights at reasonable notice, including joint inspections with your checklist.
• Sample and cadence: how many items or areas are inspected each month, and by whom.
• Critical failure definition: which items constitute a failure regardless of the overall score, and the remediation window.
• Records and export: photos and completion data are yours, exportable in a usable format on request and at exit.

Two supporting clauses are worth negotiating: a price adjustment tied to scope change rather than goodwill, and a dispute process that starts from the joint inspection record rather than competing recollections. A commercial cleaning contract template is a starting point, not a substitute for the four clauses above.

Where a portfolio runs several sites, the same clauses should appear in every contract, or comparison becomes guesswork — the argument for a standing vendor panel with one verification standard.

When should work come back in-house?

Three triggers, any one of which justifies reopening the decision:

• Repeat critical failures. The same critical item fails three times in a rolling twelve months, with remediation that does not hold.
• Evidence you cannot obtain. Two consecutive quarters where you cannot produce the completion records you need, or cannot export them at all.
• Oversight cost exceeding the saving. Your team spends more hours verifying the service than it would spend supervising it directly.

Bringing work back is a project, not a switch. Staffing, equipment, induction and the first ninety days of a new operating model all need sequencing, and the incumbent's records should transfer to you in a readable format before the last invoice is paid. A re-tender that repeats the old specification buys the same result.

One distinction: choosing to deliver a service rather than contract it is not the same as choosing one vendor rather than several. Self-delivered versus contracted services covers how the mix is usually assembled in a mid-size portfolio, often as in-house supervision over contracted delivery — the combination that keeps evidence in your hands.

Frequently asked questions

Is outsourcing always cheaper than an in-house team?

No, and the two figures are rarely compared honestly. An in-house team hides supervision, cover and equipment costs in salaried hours; a contract hides audit time and transition risk outside the price. Put both on one sheet with the same line items and the difference is usually smaller than either side claims.

What should never be outsourced?

Accountability for condition, safety and records. You can buy cleaning, maintenance and inspection labour; you cannot buy the obligation to know what state the building is in. Regulated specialist work is contracted by necessity, and verification of it stays with you.

How do we stop an outsourced service drifting after year one?

Keep the verification cadence unchanged when the novelty fades. A monthly sample of scored items, a quarterly joint inspection and a review of repeat findings will catch drift faster than a satisfaction survey. Drift is almost always a change in how the work is supervised, not in what the contract says.

Can we outsource part of a service and keep the rest?

Yes, and it is often the best answer. Keep judgement and supervision in-house, contract the periodic and specialist delivery. Split the scope at the evidence boundary: whoever produces the record you rely on should be the party you can audit directly.

How often should the outsourcing decision be reviewed?

Annually as a minimum, and immediately after a critical failure or a change of building use. A decision made for a building under construction rarely fits the same building two years into occupancy.

Put the decision on paper before the contract

The best outsourcing decisions are dull: a scope with five unambiguous items, a vendor who reports in the format you already use, and a monthly sample that takes an hour. The worst are the ones where nobody wrote down what "done" looks like, so every conversation is a negotiation.

NeatScore is the verification half of that arrangement: area-specific checklists with weighted and critical items, photos per item, offline inspections that sync, deficiency tracking, dashboards, and a dated branded report you can put in front of an owner or an auditor. Try it free for 14 days, or see pricing.