What SFG20 Compliance Actually Costs: Five Hard FM and TFM Perspectives on the Same Broken Standard
SFG20 does not fail the same way for everyone. The cost estimator who loses a contract, the NHS estate director whose dashboard says green while the boilers are failing, the maintenance manager firefighting reactive demand, the Hard FM contractor bleeding margin in year two, and the real estate investor underwriting a portfolio on incomplete data — each one carries a different version of the same structural problem. This article goes through each perspective in full.
The Hard FM Cost Estimator
Pricing from a standard that was never designed to be a cost model
The cost estimator's relationship with SFG20 is the most commercially consequential in the entire Hard FM supply chain, and the least discussed in public. Every Hard FM bid is built on a labour model. Every labour model references SFG20 task codes and durations. And every experienced estimator knows, privately, that those durations are wrong for their specific estate — and builds workarounds that are never visible to the client.
The damage this causes is not confined to the estimator or the contractor. It propagates through the entire contract. A bid priced on SFG20 generic timings that do not reflect site-specific access conditions is not a bid on the right contract. It is a bid on a theoretical estate that does not exist. The contractor who wins on that bid is committed to delivering a real estate on a theoretical price.
The Specific Numbers
A typical Hard FM engineer costs an operator between £38,000 and £52,000 per annum fully loaded, depending on trade, region, and contract type. At 1,350 productive hours per year — the realistic figure on a complex estate, once travel, permits, access prep, documentation, and reactive demand are factored in — the fully loaded cost per productive hour runs to between £28 and £39. SFG20 task durations are built for a theoretical 1,800-hour working year with no access friction. A labour model applying SFG20 durations to a 1,800-hour assumption is undercosting by 25 to 33 percent before a single site-specific variable is introduced.
On a contract with 10 Hard FM engineers, that undercosting is worth between £95,000 and £195,000 per year in misallocated labour budget. Over a five-year contract, it is between £475,000 and £975,000. This is not a rounding error. It is a structural gap that the winning contractor either absorbs, recovers through variations, or delivers around by reducing service quality.
| SFG20 Default Model | Rigorous Site-Specific Model |
|---|---|
| Task hours from generic Facilities-iQ timings, unadjusted | Task hours from SFG20 adjusted by access complexity multiplier (1.0–3.0×), permit-to-work time allowance, and asset age and condition factor |
| Labour hours: 1,800 per engineer per year (theoretical working year) | Labour hours: 1,300–1,400 per engineer per year (productive wrench time, estate-specific) |
| Reactive demand: excluded or absorbed into contingency | Reactive demand: explicitly modelled as 15–25% of engineer capacity |
| Asset register: taken as supplied with no gap allowance | Asset register: gap-adjusted by 10–20% with named contingency line item |
| Risk: absorbed into contingency percentage (typically 5–10%) | Risk: named, quantified, separately priced and disclosed in submission |
| Result: competitive price that may lose money from day one | Result: accurate price that reflects the real cost of the estate being maintained |
Baachu's estimating practice has built and reviewed Hard FM labour models across more than 300 bids covering NHS trusts, local authority estates, PFI portfolios, commercial office portfolios, data centres, and industrial sites. The gap between what SFG20 default pricing produces and what the estate actually costs to maintain is present on virtually every contract. The size of the gap varies. The existence of it does not.
The estimator who prices the risk correctly loses the bid. The estimator who ignores it wins the contract and loses money. SFG20's generic timings make both outcomes more likely than they should be.
The NHS and Public Sector Estate Director
HTM obligations, CQC readiness, and the personal liability SFG20 does not protect against
An NHS Trust Estates Director is not managing a building. They are managing a regulatory position across a live clinical environment where the consequence of a maintenance failure is not a cold office or a broken lift. It is a ward closure, a CQC inspection, an HSE investigation, or in the worst case, patient harm. The stakes are different. The evidence requirements are different. And the gap between what an SFG20-aligned PPM schedule claims and what is actually happening on the estate is a personal professional and legal risk that lands on one person: the Responsible Person named in the trust's building safety documentation.
The specific regulatory obligations an NHS Estates Director carries go well beyond what SFG20 covers. HTM 04-01 governs water safety and L8 legionella management. HTM 06-02 and HTM 06-03 cover electrical safety and electrical distribution systems. HTM 01-05 covers decontamination and sterile services. Each of these Health Technical Memoranda sets maintenance requirements that are more specific, more demanding, and more legally exposed than the SFG20 equivalent task. A trust that manages its statutory compliance position through SFG20 and a CAFM dashboard is managing to the wrong standard. HTMs are the floor for NHS estates. SFG20 is not.
What a CQC Inspection Actually Asks For, and Where SFG20 Falls Short
When the Care Quality Commission inspects a trust, or when the HSE arrives following an incident, the question is not 'do you have an SFG20 subscription?' The question is: can you demonstrate, with evidence, that the right maintenance was done on the right asset by a competent person at the required interval, and that any deviations were risk-assessed and documented? That evidence chain has four components. SFG20 contributes to one of them.
| Evidence the regulator wants | What SFG20 provides | What SFG20 does not provide |
|---|---|---|
| Correct task, correct frequency, correct statutory standard | Task codes and frequencies aligned to legislation (partial) | HTM-specific task requirements, which often exceed SFG20 scope |
| Competent person performed the task | Nothing — SFG20 does not verify or record engineer competency | Competency records, AP/CP designations, WAMITAB, HTM training records |
| Evidence of task execution, not just task closure | CAFM closure record | Physical evidence, photographic record, engineer signature on task sheet, service logbook entry |
| Risk assessment for any deviation or deferral | Nothing — SFG20 does not require deviation documentation | Written risk assessment, signed by authorised person, retained in the PPM record |
The NHS Estates Director whose contractor delivers SFG20-aligned PPM at 97 percent completion rate is not necessarily in a defensible position. If the 3 percent outstanding tasks include any statutory Red-coded items, any HTM-critical maintenance, or any assets in clinical areas where failure creates patient risk, the completion rate is irrelevant. What matters is the specific gap, the evidence of how it was managed, and the documented risk assessment that shows the Responsible Person made an informed decision about each deferral.
There is also the P22 backlog maintenance problem specific to NHS estates. NHS England's Property and Infrastructure database estimates the NHS estate backlog maintenance liability at over £10 billion. Most trusts carry a backlog position. SFG20 schedules generate tasks based on what should be done. They do not account for the backlog of what was not done in previous years, on ageing plant that has been deferred because of capital constraints, ward closures, or infection control restrictions during COVID. An NHS Estates Director managing a trust with a significant backlog position is managing triage: what must be done immediately, what can be deferred with documented risk, and what needs to be escalated as a capital bid. SFG20 provides no framework for that triage. That is the gap that costs trusts money and exposes directors to personal liability.
The CQC does not ask whether you have an SFG20 subscription. It asks whether you can prove that the right maintenance was done, by the right person, with the right evidence, and that any gap was risk-assessed. Most NHS estates cannot answer all four questions. SFG20 only helps with the first.
The Asset Maintenance Manager
The engineer who knows the estate is right. The CAFM system that ignores him is costing money.
The asset maintenance manager sits between the compliance requirement and the operational reality. They know the estate. They know which assets are approaching end of life, which plant rooms flood when it rains, which FCUs have been running hot since the last service, and which statutory tasks are genuinely critical versus which ones are administrative compliance exercises. And they know that the SFG20 schedule, as generated by their CAFM system, does not reflect most of that knowledge.
SFG20 is a frequency-based maintenance model. It tells you to service asset X every 12 months. It does not tell you that asset X is a 15-year-old AHU that has been running continuously since the last bearing change three years ago, that its BMS data shows it is running 8 degrees above normal operating temperature, and that the realistic maintenance requirement is quarterly inspection with a bearing replacement scheduled within 90 days. The SFG20 schedule will generate an annual service task. The asset maintenance manager knows this is wrong. The CAFM system does not.
The Cost of Over-Maintenance and Under-Maintenance: Both Are Real
The SFG20 debate is usually framed around under-maintenance: the risk that frequency-based schedules miss deteriorating assets. The equally real and less discussed problem is over-maintenance: the cost of servicing assets that do not need servicing at the frequency the standard specifies, because they are new, because they are condition-monitored, or because the manufacturer's own specification is less demanding than the SFG20 generic schedule.
| Scenario | SFG20 schedule says | What intelligent maintenance says | Annual cost difference |
|---|---|---|---|
| 2-year-old BMS-integrated AHU with continuous performance monitoring | Annual full service + quarterly filter checks | Condition-triggered inspection when BMS flags deviation from baseline | Save: £800–£1,200 per unit |
| 15-year-old chiller approaching end of design life, no BMS integration | Annual service per SFG20 frequency | Quarterly inspection, vibration analysis, oil sampling, planned replacement programme | Spend more now: avoid £45,000–£120,000 emergency replacement |
| New variable speed drive on a pump system, manufacturer: 2-year service interval | SFG20 specifies annual service | Manufacturer interval honoured, performance data logged | Save: £400–£600 per unit per year |
| Critical standby generator on an NHS site, tested monthly | Monthly test + annual full service | Monthly test, 6-monthly inspection, annual full service with load bank test | Correct: this is genuinely critical and SFG20 is right |
The asset maintenance manager who is operating entirely within the SFG20 frequency model is over-spending on some assets and under-spending on others simultaneously. The net financial impact on a large estate is significant. More importantly, the risk profile is wrong: the assets that need more attention are getting the same schedule as the assets that need less, because the standard does not differentiate by condition, age, criticality, or manufacturer specification.
The financial consequence of running a pure SFG20 frequency model on a large estate is measurable. On a 500-asset estate where 30 percent of plant is new or BMS-integrated and 20 percent is at or approaching end of design life, the over-maintenance waste on the new plant and the under-maintenance risk on the ageing plant are running simultaneously and invisibly. The maintenance manager who flags this is right. The CAFM system generating the SFG20 schedule does not know the difference. The gap between what the system says and what the estate needs is sitting in the maintenance manager's head — uncosted, undocumented, and unrecognised until something fails.
When a chiller that needed quarterly vibration analysis was getting an annual service because SFG20 said so, and it fails in August during a heat wave, the reactive repair cost is £18,000 to £45,000 plus the consequential cost of the failure. The PPM records will show 100 percent completion on that asset right up to the failure date. The maintenance manager knew it was running wrong. Nobody gave them a mechanism to act on that knowledge within the SFG20 framework.
The maintenance manager's knowledge of the estate is the most valuable asset in the Hard FM operation. SFG20 provides no mechanism to capture it, apply it, or price it. That is not a gap in the manager. It is a gap in the standard.
The Hard FM Contractor
Winning the bid, losing the contract, and finding out why in year two
The Hard FM contractor's experience of SFG20 dependency is the most commercially damaging in the supply chain, because it is the least recoverable. An estate director can commission an asset register audit. An estimator can build a better model on the next bid. A maintenance manager can propose a criticality assessment. A contractor who has won a five-year Hard FM contract at a price built on SFG20 generic timings and an incomplete asset register is committed. The contract is signed. The mobilisation has started. The engineers are on site. The discovery of the pricing gap is a matter of when, not whether.
The typical trajectory of a mispriced Hard FM contract follows a recognisable pattern. Months one to six: mobilisation, site familiarisation, asset register validation — this is when the gap between the register provided at tender and the estate that actually exists becomes visible. Months six to twelve: reactive demand exceeds the comprehensive threshold model. Labour is being pulled off PPM to cover reactive, PPM completion rates begin to slip, deferred tasks accumulate. Months twelve to eighteen: the commercial manager runs the first full year reconciliation — labour costs are above budget, reactive recovery costs have not been fully recharged, the comprehensive fund is under-water. Month eighteen to twenty-four: the first contract review. The client sees PPM completion rates below target. The contractor is losing money and failing to deliver. Both parties are right. The problem was created at bid stage.
The Three Mechanisms of Margin Erosion on a Mispriced Hard FM Contract
Bid assumption: 1,800 productive hours per engineer per year.
Estate reality: 1,300–1,400 hours on a complex multi-site or high-permit estate.
On a contract with 15 engineers, the gap is between 6,000 and 7,500 hours per year. At a blended fully loaded cost of £32/hour, that is £192,000 to £240,000 per year in labour that was not priced.
Over five years: £960,000 to £1,200,000. This gap exists before any site-specific access or permit variables are applied.
Bid assumption: client-provided register is broadly accurate.
Estate reality: 15–25% of assets are missing, mis-categorised, or wrong.
On a contract where the provided register shows 800 maintainable assets and the actual estate has 960–1,000, the PPM schedule is 20% under-resourced from day one. The contractor either delivers an incomplete PPM programme or absorbs the cost of additional engineer time not priced in the bid.
Asset register gap cost on a mid-size estate: £30,000–£80,000/yr in unpriced labour.
Bid assumption: reactive demand contained within comprehensive threshold.
Estate reality: reactive demand is 20–30% above the threshold model on many estates.
When reactive pulls engineers off PPM, one of two things happens: PPM slips (deferred tasks accumulate, client performance deductions follow) or additional resource is deployed (unbudgeted cost absorbed by the contractor).
On a contract with 15 engineers and reactive demand running 25% above model, the annual cost of the gap is between £85,000 and £140,000 in unpriced resource.
This figure assumes the deferred PPM does not result in asset failure. When it does, the reactive repair cost is additional and usually rechargeable only at the standard reactive rate, not at the true cost of the failure.
The Hard FM contractor who identifies these three mechanisms at bid stage and prices them explicitly is operating to a different standard than the one who absorbs them into a contingency percentage and hopes. On a competitive bid, explicit pricing of these risks may cost the contract. Not pricing them will cost the margin. The choice is structural, not accidental, and it is created by a standard that was never designed to be used as a cost model.
Baachu's bid support practice has worked on both sides of this problem: building the rigorous model that prices the risk correctly on new bids, and working with contractors in contract who need to understand where the model went wrong and what can be recovered. The recovery options are real but limited — scope review, TUPE headcount analysis, task bundling and route density optimisation, variation claims on scope not included in the register. None of them recover a contract mispriced by 15 to 20 percent. They reduce the loss. They do not eliminate it.
The Real Estate Investor and Asset Owner
Underwriting a portfolio on compliance statistics that measure the wrong thing
A real estate investor acquiring a commercial office portfolio, an industrial logistics estate, or a mixed-use urban asset will typically receive, as part of vendor due diligence, a Hard FM compliance report. That report will show an SFG20-aligned maintenance regime, a CAFM extract with PPM completion statistics, and a schedule of current maintenance contracts. The acquisition team will note the SFG20 compliance, satisfy themselves that a maintenance framework is in place, and move on to the financial model.
Within eighteen months of acquisition, it is not uncommon for that same team to discover that the maintenance liability of the estate is materially larger than the compliance documentation suggested. The scenario is specific enough to be worth walking through in detail, because the mechanism by which SFG20 compliance documentation creates a false picture for investors is structural, not accidental.
A Deal Scenario: The £2.3 Million Gap Nobody Saw Coming
Investor acquires a portfolio of four grade B commercial office buildings across the South East. Total acquisition cost: £47 million. Vendor provides SFG20-aligned PPM compliance report showing 96 percent completion rate for the preceding 12 months. Asset register shows 1,847 maintainable assets across the four buildings. The Hard FM maintenance contract is with a mid-tier FM operator, two years into a five-year term. All looks clean.
Twelve months post-acquisition, the investor's asset manager commissions an independent Hard FM condition survey as part of a refinancing exercise. The survey finds the following:
This scenario is not an outlier. It is a routine consequence of accepting SFG20 compliance documentation as a proxy for actual maintenance liability. The mechanisms are always the same: asset register not independently verified, backlog deferred and removed from outstanding counts, plant age and capital replacement liability not in any SFG20-derived report, statutory records fragmented across specialist subcontractors.
The investment team that commissions proper Hard FM due diligence — independent asset register survey, physical condition inspection, backlog quantification, capital replacement modelling, statutory records integration — knows what they are buying. The team that accepts the SFG20 compliance report finds out after completion.
The SFG20 compliance report tells the investor that the maintenance framework exists and that tasks are being closed. It does not tell them what the building will cost over the next ten years. Those are different documents. Most investors receive only one of them.
Frequently Asked Questions
On a contract with 15 engineers, the three main mechanisms — the labour hour gap, the asset register gap, and the reactive demand interaction — can together account for between £300,000 and £460,000 per year in unpriced cost. Over a five-year contract, the cumulative exposure runs from £1.5 million to over £2 million. The size depends on estate complexity, asset register quality, and the accuracy of the reactive demand model at bid stage.
For an NHS trust, SFG20 compliance is a necessary but insufficient condition for regulatory defensibility. The binding standards are the Health Technical Memoranda — HTM 04-01, HTM 06-02, HTM 06-03, HTM 01-05 and others — which set more specific and more demanding requirements than SFG20 in clinical environments. SFG20 compliance demonstrates that a maintenance framework is in place. HTM compliance demonstrates that the specific statutory obligations for healthcare estates are being met. Regulators and inspectors focus on the latter.
Five questions that go beyond the PPM completion report: Has the asset register been independently verified, or is it carried over from the previous operator? What is the backlog maintenance position, and how has it been treated in the completion statistics? What is the age profile of primary plant and when does each major system require capital replacement? Are all statutory compliance records — particularly L8, electrical, lifts, and gas — held centrally or fragmented across specialist subcontractors? What is the total cost of SFG20 compliance infrastructure including subscriptions, CAFM integration, and internal resource?
The Building Safety Act 2022 places personal liability on the Responsible Person for ensuring that fire and structural safety obligations are met and evidenced. This liability is not discharged by pointing to an SFG20 subscription or a high CAFM completion rate. It is discharged by demonstrating, with evidence, that the right maintenance was performed on the right assets by competent persons, that deviations were risk-assessed, and that the golden thread of building safety information is maintained and current. SFG20 provides a task framework. The BSA 2022 requires an evidence framework. The two are not the same.
- Art. 1 Is SFG20 Outdated? The Hard FM Baseline That Built an Industry and Why It Is Now Costing You Money
- Art. 2 SFG20 Labour Hours: Why Your PPM Pricing Is Wrong Before the Contract Starts
- Art. 3 Who Owns SFG20? BESA, Facilities-iQ, and the Hard FM Commercial Risk
- Art. 4 What SFG20 Compliance Actually Costs: Five Hard FM and TFM Perspectives on the Same Broken Standard (this article)
- Art. 5 Why Is the Hard FM Asset Register Always Wrong? TUPE, retendering, SFG20 mapping failures, and 30 years of outsourcing.
- Art. 6 Is There a Credible Alternative to SFG20? ISO 55001, condition-based maintenance, and the hybrid model.
- Art. 7 What Does Technology Do to the Case for SFG20? IoT, BMS integration, digital twins, and AI maintenance analytics.
- Art. 8 How to Build a Defensible Hard FM Maintenance Framework Without SFG20 as the Anchor.
- Art. 9 SFG20 State of FM Report 2026: What the Data Actually Shows About SFG20 Compliance, Asset Registers and Hard FM Dependency.
Which perspective applies to your situation? One email routes you to the right conversation. → hello@baachu.com
Next: Article 5 · The Asset Register Lie: Why Hard FM Compliance Data Is Unreliable and How SFG20 Dependency Makes It Worse
Read Article 5 →