The renewal meeting runs forty minutes. Three contractors are discussed. One is kept because the account manager is responsive and everyone likes him. One is dropped because of a bad job in March that people still remember. One is renewed by default because nobody has anything specific to say about them. No numbers are presented, because no numbers exist — and a contract worth several hundred thousand a year is decided on recency, relationship and the loudest recent complaint. That is contractor evaluation in most facility organisations, and it is not a discipline problem; it is a data problem. OxMaint turns every dispatch into scorecard evidence so the renewal conversation starts from a number instead of a feeling.
The measurement gap in outsourced facility maintenance
60–70%
Of maintenance work in the average FM operation is delivered by contracted service providers rather than direct staff
Under 30%
Of those operations have a documented SLA against which contractor performance is actually measured
67%
Of commercial facility managers report limited or no formal vendor performance tracking beyond invoice approval
$8K–22K
Estimated annual hidden cost per vendor in undocumented rework, unclaimed credits and missed service intervals
Monitoring Tells You What Happened. Evaluation Decides What Happens Next.
Most facility teams that believe they have a contractor problem actually have an evidence problem. Work is being done, invoices are being approved, and somewhere in that flow the information needed to judge a supplier is being destroyed rather than captured. By the time renewal arrives, four predictable distortions have taken the place of data.
Recency
The last ninety days outweigh the previous three quarters. A strong performer with one visible failure loses to a mediocre performer with a quiet quarter.
Relationship
Responsiveness on the phone is mistaken for responsiveness in the field. The vendor who communicates well about delays outranks the one who simply does not have them.
Invoice-only view
Finance can produce spend per vendor instantly and quality per vendor not at all, so the only comparable number in the room is price.
Silence
No complaints is read as good performance. It usually means the failures are being absorbed internally by a coordinator nobody is counting.
None of these are failures of judgement. They are what any reasonable person does when asked to compare suppliers without comparable information, and they are remarkably consistent across organisations of every size. The fix is not better meetings or more disciplined note-taking — it is capturing the four or five numbers that make the comparison possible, as a by-product of work that is happening anyway.
The Cost-Versus-Quality Quadrant
Once response, first-time fix, rework and true cost are all being captured, every contractor lands in one of four positions — and each position has a different correct action. The dangerous quadrant is not the expensive one. It is the cheap one that looks like a saving on the ledger while consuming your team's time and your assets' life.
High quality · Higher cost
Negotiate and protect
Genuinely good work at a premium. Do not put this contract out to bid on price alone — quantify the rework and downtime they prevent, then negotiate scope and rate against evidence.
High quality · Lower cost
Grow and lock in
Your best commercial position. Expand scope, extend term for rate certainty, and make sure the scorecard is shared so the performance that earned it is visible to them too.
Low quality · Higher cost
Exit deliberately
The clearest decision on the board, and usually the one delayed longest because no evidence file exists. Build the record for two quarters, then transition with documented cause.
Low quality · Lower cost
The false bargain
Wins bids, looks efficient in the spend report, and quietly transfers cost into your organisation as callbacks, coordination hours and shortened asset life. The most expensive vendor you have.
Cost position — lower on the right
The quadrant only works if the vertical axis is measured rather than assumed, which is exactly the part most organisations skip. First-time fix rate is the single most useful proxy: industry benchmarks put facility maintenance between 85% and 92%, and a contractor sitting below 80% is generating repeat visits that your own team has to coordinate. Book a session and we will map your current vendors onto this grid using your own work order history.
Contractor Evaluation — OxMaint
You Cannot Negotiate With a Vendor Using an Opinion.
OxMaint timestamps creation, acknowledgement, arrival and resolution on every dispatch, calculates SLA compliance per contractor automatically, and generates the scorecard that turns a renewal conversation into a negotiation.
What the Invoice Never Shows You
A contractor's true cost is their invoice plus everything their performance pushes onto your organisation. Almost none of that second half appears in a spend report, which is why the cheapest bidder so often wins and so rarely saves money.
Total contractor cost — visible and invisible
Contracted invoice value
Visible — the only figure most reviews use
Callbacks and repeat visits
Invisible — absorbed as coordination time
Internal coordination hours
Invisible — chasing, rescheduling, verifying
Unclaimed SLA credits
Invisible — contract clauses never invoked
Shortened asset life
Invisible — surfaces years later as capex
Missed emergency response commitments commonly carry service credits in the region of $150 to $500 per dispatch. The reason those credits go unclaimed is almost never leniency — it is the absence of a timestamped arrival record that would survive a dispute with the contractor.
A Scorecard That Changes Behaviour, Not Just Reporting
Weighting matters more than the metric list. A scorecard where every category counts equally tells a contractor nothing about your priorities, and a scorecard with fifteen metrics gets ignored by everyone. The starting model below is widely used for general facilities contracts — adjust the weights before the first reporting period, with procurement, finance and the contract owner all signed up to them.
Category
Weight
What Gets Measured
Benchmark or Trigger
Quality of work
30%
First-time fix rate, rework percentage, 30-day callback rate, completion quality checks
FTFR 85–92%; below 80% triggers review
Delivery and responsiveness
25%
Response time against tier, on-time completion rate, schedule adherence on planned work
Below 90% emergency compliance triggers a plan
Cost compliance
20%
Quote-to-invoice variance, rate card adherence, cost per work order against trade peers
Variance beyond agreed tolerance requires sign-off
Safety and risk
15%
Licence and insurance currency, permit compliance, incidents, method statement adherence
Expired credential blocks assignment outright
Documentation and service
10%
Report submission timeliness, completeness of records, communication on delays
Late documentation is a compliance gap, not a nuisance
Weight for the risk profile rather than copying the table. A life-safety systems contractor should carry far more weight on safety, compliance and documentation, because a missed interval there is a regulatory exposure rather than an inconvenience. A low-risk commodity service can weight invoice accuracy and price stability more heavily. What must not change is that the scores come from work order data automatically, not from someone's impression typed into a form the week before the review.
What a Composite Score Looks Like in Practice
Abstract weighting only becomes persuasive when you see it resolve a real decision. Consider a mechanical contractor across twelve months of tracked work — the kind of vendor that generates no complaints, submits tidy invoices, and would almost certainly have been renewed on instinct.
Vendor A — mechanical services, 240 work orders over twelve months
At first review
On-time completion
82.5% — 198 of 240 closed within window
First-time fix
71% — well below the 85% benchmark
Average response
52 minutes — genuinely strong
Rework rate
11% of jobs required a return visit
Composite
74.2 — under the 75 threshold
Two quarters later
Action taken
Scorecard shared, 60-day improvement target set on rework and first-time fix
Commercial lever
A portion of planned spend held back pending the next review
Composite
81.7 — comfortably above threshold
Nothing about this outcome required replacing the contractor. It required showing them a number they had never been shown, attaching a consequence to it, and measuring again on a fixed date. The response time that looked impressive in isolation was masking a first-time fix rate that was quietly costing the facility team a return visit on roughly one job in nine.
This is the practical case for evaluation over monitoring. Monitoring would have reported 52-minute response times and a healthy-looking dashboard. Evaluation surfaced the 71% first-time fix rate sitting behind it, converted that into a target with a date, and recovered the performance without a tender exercise, a transition period or a single difficult conversation about ending the relationship.
The Review Cadence That Makes It Stick
A scorecard is a rhythm before it is a document. The value comes from the fact that the same measurements arrive on the same dates, so a change in performance is legible as a movement rather than an anecdote.
Scorecards fail for one of two reasons: they are produced manually and therefore stop being produced, or they are produced and never shown to the contractor. The rhythm below solves both.
Monthly
Capture, automatically
Every dispatch timestamps creation, acknowledgement, arrival and resolution. Nobody assembles anything — the score accumulates as a by-product of running the work.
Quarterly
Review, with the contractor in the room
Share the actual scorecard. Contractors consistently respond better to a number they can see and argue with than to vague feedback they cannot act on.
On breach
Improvement plan with a deadline
A composite below threshold triggers a written plan with specific targets and a defined window, rather than a warning that everyone forgets by the next quarter.
Annually
Renew, consolidate or exit
Twelve months of comparable data across every vendor in the trade. Consolidation decisions become defensible, and exits carry documented cause rather than a difficult phone call.
Tracking first-time fix per vendor and feeding it into the renewal scorecard is associated with a substantial fall in 30-day callback rates — around 40% in reported facility portfolios. The mechanism is not clever analytics. Quality improves because it is finally being counted, and the contractor knows it.
What OxMaint Handles for You
None of this requires a perfect data set to begin. The first quarter of tracked dispatches is usually enough to separate the top and bottom of a vendor list, and the picture sharpens with every month that passes.
SLA clock on every work order
Priority tier sets the response and resolution window; the system measures against it automatically and records breach evidence that holds up in a commercial dispute. Emergency and routine tiers run their own clocks, so a vendor strong on urgent work cannot hide a weak record on planned maintenance.
Automatic scorecard generation
Response, first-time fix, rework, cost variance, safety and documentation compiled from work order data — no manual collection, no spreadsheet maintained by one person.
Credential blocking
Licence, certification and insurance expiry held on the vendor record, with assignment blocked until current documents are uploaded — so an uncertified technician cannot be dispatched.
Site, region and portfolio rollup
A national vendor scored per site, per region and as a portfolio composite, so a strong average does not hide two locations where service has collapsed.
Cost per work order by trade
Comparable cost data across vendors in the same trade, which is what converts a rate negotiation from an assertion into a benchmarked position — and shows where a low headline rate is being recovered through job volume.
Evidence retained per vendor
Completed work, photos, sign-offs, certificates and breach history retained against the vendor record — available for audits, disputes and the next tender exercise.
Frequently Asked Questions
How many metrics should a contractor scorecard actually contain?
Five weighted categories is the practical ceiling for something contractors will engage with. Quality, responsiveness, cost compliance, safety and documentation cover the decision. More metrics produce a longer report that changes nobody's behaviour.
What response times should we write into the SLA?
Emergency dispatch commonly sits at two to four hours and routine work at 24 to 72 hours, tightened for healthcare and data centre environments. The tier structure matters less than enforcing it —
the timestamp is what makes the clause real.
Won't contractors resist being scored?
Good ones generally welcome it, because it finally distinguishes them from cheaper competitors on something other than rate. Resistance usually comes from vendors whose performance has never been visible, which is information in itself.
We have forty vendors and a spreadsheet. Where do we start?
Start with the trades carrying the most spend and the most risk, and score on three metrics before adding more. A big-bang rollout across every vendor and category typically stalls before the first full quarter of data exists.
Can this recover SLA credits we are currently not claiming?
Only where the contract already contains the clause — but that is usually the case, and the missing piece is evidence rather than entitlement.
A demo covers how breach records are captured and presented.
Facility Contractor Evaluation Software
Renew on Evidence. Exit on Evidence. Negotiate From Evidence.
5
weighted scorecard categories
Every
dispatch on an SLA clock
Put every contractor in your portfolio on the same measured footing — response, first-time fix, rework, cost and compliance — and walk into the next renewal with twelve months of comparable data instead of an impression.
