Whether the Equality and Human Rights Commission’s updated single-sex Code of Practice has formally “commenced” is, for governance purposes, a red herring. It changes neither the underlying law nor what an organisation is already required to do. The Supreme Court’s 2025 judgment in For Women Scotland Ltd v The Scottish Ministers and the Equality Act 2010 itself already stand; the Code is implementation guidance on an existing duty, not the event that creates it. Nor should anyone expect the wider legal landscape to settle down around the updated Code — litigation, further tribunal decisions and future authorities will keep shaping how the law is applied for years. That is why “wait for clarity” is not a viable governance strategy: it is a decision to stay exposed for however long the wait turns out to last. The organisations in the strongest position are not the ones holding out for certainty, but the ones making principled, evidence-based, proportionate decisions now — and documenting them so those decisions remain defensible however the law develops next.
The commencement date is not the starting gun
The EHRC has updated its statutory Code of Practice on single-sex services to reflect the Supreme Court’s ruling, revising the earlier 2011 Code. It is tempting to treat a new Code as a grace period — a reason to hold off on decisions until the picture feels settled. That reading misunderstands what the Code is. A statutory code of practice does not create the underlying legal obligation; it explains how to comply with an obligation that already exists in the Equality Act 2010, as that Act has now been authoritatively interpreted by the Supreme Court. The obligation to act proportionately in single-sex provision, to keep exclusion case-specific rather than blanket, and to have due regard under the public sector equality duty does not depend on any code — it flows from the statute and the case law themselves. Whichever version of the Code you consult, the statute and the case law behind it are already binding. The precise status of any code is a detail worth tracking; it is not a reason to defer a decision that the law already requires.
What already requires action, now
The Supreme Court’s ruling did the substantive work well before any Code caught up with it. On the toolkit’s reading of the judgment: “sex” in the Equality Act 2010 means biological sex, not sex as modified by a Gender Recognition Certificate; a GRC does not change a person’s sex for the Act’s purposes; the single-sex exceptions in Schedule 3 operate by reference to biological sex but still require a case-by-case proportionality justification — never an automatic or blanket exclusion, and never a requirement to exclude at all, since the exceptions are permissive powers a provider may use where justified, not duties to restrict; the public sector equality duty, so far as it concerns sex, relates to biological sex, but public authorities must still have due regard to gender reassignment separately, because gender reassignment protection under section 7 is entirely unaffected and stands as its own protected characteristic. None of that depends on the Code at all. It has applied since the judgment was handed down, and it already shapes what a defensible equality impact assessment, a defensible facilities policy, and a defensible single-sex-space decision look like today. Organisations that are quietly waiting for the Code before updating their equality impact assessments, their policies or their facilities decisions are waiting for confirmation of a duty that is already running.
The cost of waiting: reactive decisions are the expensive path
How costly reactive, last-minute compliance can be is illustrated — not proven — by the EHRC’s own modelling for the updated Code. Its final stage impact assessment, published on GOV.UK on 14 May 2026, puts the central (“middle”) estimate of the ten-year cost of implementing the associated changes at £703.1 million across the private, public and voluntary sectors, with the largest single component (around 72.6% of the total) attributable to physical changes to facilities such as toilets and changing rooms. That headline sits inside a genuinely wide range: the same assessment’s own lower- and upper-bound annualised estimates run from £5.5 million to £159.1 million a year, a spread the EHRC itself attributes to “the lack of robust data in this area.” Treat £703.1 million as a central, sector-wide planning estimate rather than a fixed bill, and note that it measures the aggregate cost of implementing new statutory guidance across the whole economy — not the cost to any single organisation of getting its own decisions wrong.
Press coverage of the same assessment — reported by The National, a Scottish pro-independence title, drawing on the EHRC’s figures — has additionally cited modelled counts of roughly 12,895 toilets, 5,526 changing rooms and 18,422 signs affected nationally. Those unit counts derive from the EHRC’s own stated assumptions (building-regulation floor-space standards, a small consultation sample, and an industry estimate that around 4% of premises have only unisex facilities); they are modelled projections, not a measured count of actual buildings, and should be read with the same caution as the cost total and checked against the primary assessment before being repeated as settled fact.
What this national cost picture illustrates for an individual organisation is a governance point, not a legal one: reactive, undocumented decision-making is already the norm rather than the exception. 23.1% of UK organisations report having already faced external pressure to reduce their trans-inclusion commitments; 50.7% have no formal policy on trans or nonbinary inclusion at all; and 26.9% have no set schedule for reviewing whatever policy they do have (Beyond Compliance survey, n=136 UK organisations, 2025–26). None of that is a legal cost estimate — it is a readiness gap. Waiting for the wider legal picture to resolve does nothing to close it; it simply leaves the gap open for longer, at the point when external pressure and scrutiny are both already rising.
The defensible-decisions imperative
The alternative to waiting is not rushing to a fixed position either. The EHRC’s statutory Code, in both its earlier and updated forms, shares the same underlying architecture: mixed provision is the starting point, single-sex or separate-sex provision is a justified exception, and the burden of showing that an exception is proportionate sits with the provider, not with the person affected by it — decided case by case, never as a blanket rule. That architecture is the services regime: single-sex facilities provided to staff sit instead under the Equality Act’s employment provisions and separate workplace and health-and-safety regulations, not this Code — a line worth keeping clear wherever the same premises are both a workplace and a public-facing service. Getting the direction of that burden right matters more than getting to an answer quickly. A defensible decision, whatever it concludes, rests on a documented rationale — an equality impact assessment that considers sex and gender reassignment as separate characteristics, not one collapsed into the other — reached proportionately on the specific facts in front of the organisation, with the reasoning kept on file rather than only the outcome. That is what makes a decision stand up to scrutiny: not that it anticipated every future legal development correctly, but that it was principled, evidenced and proportionate when it was made.
This is also why the thesis does not expire when the Code is updated. A documented, proportionate decision reached under today’s law is defensible today, and it stays defensible — or gets revisited on the record — as the law moves. An undecided, undocumented position is defensible under no version of the law, past, present or future.
Review on schedule, not under pressure
The final piece of the imperative is ongoing, not one-off: set a review date, name an accountable owner, and revisit the decision as authorities develop — on a schedule, not only when a challenge or a new judgment forces the issue. A decision with a review date attached is a governance position; one without it is a static answer to a question the law keeps asking again. The updated Code is one more data point for organisations to weigh against their existing evidence trail, not a reset button. Organisations that already hold a documented, evidenced, proportionate position will revisit it with a paper trail behind them. Organisations that were waiting for the Code will discover that it asks nothing more of them than the law already did — and that the waiting period has left nothing on record to show for it. Governance built on documented process outlasts any single legal event, whichever direction that event turns out to move the law.