Refurbishment

Refurbishment design cannot start from as-built drawings, because as-built drawings are usually wrong. Buildings get altered without record, services get rerouted, openings get formed and filled, and the drawing set that survives is the design intent from before all of that happened. The survey is not an administrative preliminary to the job. It is the mechanism by which an unknown is converted into something that can be designed for, programmed and priced, and it is the single highest-return investment on a refurbishment project because every pound of unresolved uncertainty gets carried by somebody at a much worse exchange rate later.

The survey family divides by what each one is trying to establish. A measured survey produces a dimensional record of what physically exists, usually by laser scan or point cloud, total station or photogrammetry. A condition survey records defects, residual life and repair priority element by element, and feeds maintenance planning, dilapidations and acquisition due diligence. A structural appraisal establishes the actual capacity, load path and integrity of the structure, which is a different question entirely and needs a different competence. Intrusive opening-up removes finishes locally to expose structure, connections, cavity construction and hidden services. Material testing puts numbers on the condition of the material itself.

On the structural side, the current UK best-practice document is the Institution of Structural Engineers publication "Verification of existing building structures", published on 13 November 2025 and running to 164 pages. It sets out a staged verification process running from desk study through inspection, investigation, assessment and reporting, with explicit guidance on planning and interpreting intrusive investigation and on proportionate decision-making. Its predecessor, "Appraisal of existing structures, 3rd edition" of October 2010, is still listed as current in the NBS publication index, and we could not verify that IStructE has formally withdrawn it. Treat the 2025 document as the current best-practice guidance rather than as a confirmed replacement.

The Department for Education guidance "Structural assessments and investigations", published on 11 December 2025 and drawing on the IStructE document, sets out the stages most engineers will recognise: scoping and fee agreement, desk study, visual inspection and assessment, sampling and testing, interpretation and recommendations, then design, coordination and specification. It also makes a point that is worth carrying across the whole subject, whatever the asset. Intrusive testing should only proceed where the engineer can state what knowledge gap exists and how the result will change the answer. Where it will not change the answer, a conservative lower-bound assumption is often the proportionate alternative, and it costs nothing to open up.

On residential work the reference framework is the RICS Home Survey Standard, 1st edition, published on 2 November 2020 and effective from 1 March 2021. It defines three levels: Level 1, a basic condition report with no intrusive or opening-up work; Level 2, a survey with or without valuation, giving a more detailed inspection and advice; and Level 3, the most thorough inspection and report, appropriate to older, larger, altered or unusual properties - which is most refurbishment candidates. RICS is developing a 2nd edition. Its public consultation closed in October 2025 and the April 2026 update confirmed that the expert group is still working through feedback and that an updated publication timeline will be shared as the draft develops. RICS instructs members to continue using the 1st edition until the 2nd edition takes full effect. No publication date for the 2nd edition exists, and anyone quoting one is guessing.

Asbestos is the control that sits in front of everything else, and it is a regulatory duty rather than a matter of good practice. The Control of Asbestos Regulations 2012 impose, at regulation 4, the duty to manage asbestos in non-domestic premises. The survey guidance is HSE publication HSG264, "Asbestos: The survey guide", 2nd edition, 2012. That document is current: the HSE publication page was last updated on 7 August 2025, carries no withdrawal or supersession notice, remains a free download, and is the document HSE itself names when directing duty holders to professional advice. There is no replacement document, and its age is not evidence that there is.

HSG264 defines two survey types and the difference between them decides whether work can start. A management survey locates asbestos-containing materials so they can be managed in place during normal occupation. A refurbishment and demolition survey is fully intrusive and destructive, covers the area to be worked on, and is required before any refurbishment or demolition begins. The area is normally vacated for it and certified fit for reoccupation afterwards. A management survey is not a substitute, and treating one as though it were is a recurring cause of work stopping on site with an unplanned licensed removal in the middle of the programme.

Competence in this area is unevenly regulated and it is worth knowing exactly where the line falls. There is no legal requirement for an asbestos surveyor to hold accreditation, though HSE strongly recommends surveyors accredited to BS EN ISO/IEC 17020, a position restated on its guidance page dated 28 July 2025. Laboratory analysis of samples is different: it must be carried out by an organisation accredited to ISO/IEC 17025, with UKAS the sole accreditation body in Great Britain. UKAS publication RG 8, "Accreditation of Bodies Surveying for Asbestos", confirms that ISO/IEC 17020 accreditation covers bulk sampling for refurbishment and demolition surveys. HSE ran a consultation, "Legislative and guidance proposals for the Control of Asbestos Regulations 2012", open from 10 November 2025 and closed on 9 January 2026, proposing among other things that the CDM client rather than the licensed removal contractor should appoint the analyst for four-stage clearance, and rejecting mandatory surveyor accreditation on small-business cost grounds in favour of better guidance and training. We could not verify that HSE has published its response or made any resulting amendment. Check the position before relying on it.

Asbestos is the most consequential deleterious material but it is not the only one, and a pre-refurbishment investigation that stops there is incomplete. Reinforced autoclaved aerated concrete is now a refurbishment trigger in its own right. The current technical guidance is the IStructE publication "Reinforced Autoclaved Aerated Concrete (RAAC) Investigation and Assessment - Further Guidance", published on 17 April 2023 by the RAAC Study Group, with the page last modified on 13 January 2025. It supplements rather than replaces the earlier IStructE report and the 2019 SCOSS alert, and we found no 2024, 2025 or 2026 revision. The risk drivers on site are panels at or past design life, insufficient or defective end bearing, and water ingress. On scale, the House of Commons Library briefing CBP-9917, last updated on 8 July 2026, records RAAC confirmed in 231 state-funded education settings as at 27 November 2023, and identified at 45 hospital sites as at 17 October 2023, with the DHSC and NHS programme targeting removal from the NHS estate by 2035 and £698m committed across 2021–2025.

Lead paint is the quiet one. The Control of Lead at Work Regulations 2002 apply, supported by HSE Approved Code of Practice and guidance L132, "Control of lead at work", third edition, 2002. The HSE page for L132 carries a notice that HSE has consulted on reviewing its Approved Codes of Practice, and we could not verify whether L132 has been revised, withdrawn or reissued, so confirm its status before citing it. The refurbishment exposure routes are specific and avoidable: dry sanding, burning off and hot work on pre-1992 paint films. Blood-lead monitoring and medical surveillance thresholds sit in the Regulations themselves.

Ground and vegetation complete the picture. Contaminated land is assessed under Land Contamination Risk Management, published by the Environment Agency with NIEA, SEPA and Natural Resources Wales on 8 October 2020 and last updated on 12 June 2025, structured as Stage 1 risk assessment, Stage 2 options appraisal and Stage 3 remediation and verification, with Part 2A of the Environmental Protection Act 1990 as the statutory backstop and BS 10175 as the site investigation practice standard. Japanese knotweed is governed for valuation and survey purposes by the RICS professional standard "Japanese knotweed and residential property", 1st edition, originally published in January 2022 as a guidance note and reissued in October 2022 as a professional standard, effective from 23 March 2022. It replaced the old seven-metre rule with a risk-based decision tree. Section 14 of the Wildlife and Countryside Act 1981 addresses causing it to grow in the wild, and knotweed-contaminated soil is handled as controlled waste.

Material testing is where the numbers come from, and on concrete assets it is a discipline of its own. The usual suite covers cover depth and reinforcement location by covermeter or ground-penetrating radar, carbonation depth, chloride content, half-cell potential mapping, resistivity, petrography, core compressive strength, rebound hammer, timber moisture content and rot, mortar analysis, and metallurgy on wrought and cast iron. Several of the named test standards are cited in the DMRB concrete repair document, including BS EN ISO 12696 for cathodic protection of steel in concrete, BS EN ISO 15257 for cathodic protection personnel competence levels, and ASTM C876 for half-cell potential measurement. For carbonation depth, chloride content and covermeter work the standards commonly cited are BS EN 14630, BS EN 14629 and BS 1881-204, but we could only verify those three through secondary sources rather than a primary BSI record, so this page names them without edition years.

Domestic refurbishment is where most of the industry actually works, and it is the part where the regulatory framework is most often misunderstood - not because it is obscure, but because it is fragmented. A rewire, a boiler or heat pump replacement, a kitchen and bathroom renewal, replastering, window and door replacement, loft insulation, roof covering renewal, internal reconfiguration, floor renewal, damp and timber treatment, and at the upper end a single-storey rear extension or a loft conversion. Each of those can touch a different consent regime, and satisfying one of them does not discharge the others.

The Building Regulations 2010 apply to "building work" as defined, which captures far more of ordinary refurbishment than people expect. It picks up material alterations, material changes of use, the provision or extension of controlled services and fittings including heating, hot water, electrics in dwellings and windows, and work to a thermal element. In practice the domestic touchpoints are Parts A for structure, B for fire, C for moisture, F for ventilation, G, H, J, K, L, M and P for electrical safety in dwellings. Replacing a window is building work. Replacing a boiler is building work. Insulating a roof is building work on a thermal element.

The Part L position needs stating carefully because it is in transition. The currently applicable documents are Approved Document L Volume 1: Dwellings and Volume 2: Buildings other than dwellings, 2021 edition incorporating 2023 amendments. New 2026 editions of Approved Document L Volumes 1 and 2, and Approved Document F Volume 1, were published on 8 April 2026 to implement the Future Homes and Buildings Standard. The amending regulations come into force on 24 March 2027, with regulations 3, 4, 6, 7, 8 and 9 taking effect on 24 September 2027 for higher-risk building work. The 2026 amendments are directed at new buildings and new building work; the guidance does not indicate general application to existing-building refurbishment, though it does address material changes of use and work to existing higher-risk buildings. We did not read the detailed existing-dwelling and consequential-improvement provisions of the 2026 documents in full, so confirm the position for any specific job that straddles the dates.

On planning, householder permitted development sits in Schedule 2, Part 1 of the Town and Country Planning (General Permitted Development) (England) Order 2015. Permitted development rights are disapplied or restricted for flats and maisonettes, for listed buildings, for land in conservation areas, areas of outstanding natural beauty, National Parks and World Heritage Sites - the Article 2(3) land categories - and wherever an Article 4 direction has been made. Two recent amendment orders are worth knowing about, SI 2025/560 and SI 2026/313, the latter in force from 9 April 2026. Neither made a material change to householder permitted development rights. The 2026 order added a new Class CA for reverse vending machines at shops, clarified that rooftop structures on schools, colleges, universities, prisons and hospitals under Class M may not exceed 1.5m above building height, and updated GPDO cross-references from the July 2021 NPPF to the December 2024 NPPF.

The Party Wall etc. Act 1996 applies in England and Wales and catches more domestic refurbishment than most homeowners realise. It bites on three categories of work: work directly to a party structure under section 2, building on the line of junction under section 1, and excavation within 3m or 6m of a neighbouring structure below defined planes under section 6. The statutory route is notice, then consent or dissent, then appointment of surveyors and a party wall award. MHCLG publishes an explanatory booklet on GOV.UK, and notice periods and the award mechanism should be taken from the Act and the booklet directly rather than from summaries. It is a separate regime from planning and building control, running on its own timescales, and it is a frequent source of programme delay precisely because it is left until the design is fixed.

Building control in England runs by two routes for ordinary work: the local authority building control service, or a Registered Building Control Approver, the post-Building Safety Act successor to Approved Inspectors. Applications are made by full plans, by building notice for small domestic work only, or by regularisation where work has already been carried out without approval. Competent Person Schemes allow self-certification of defined work including electrical, gas, window and heating installation, which is how most single-trade domestic work is discharged.

Higher-risk buildings are a different route entirely, and this is the change that has reshaped residential refurbishment in tall blocks. The Building Safety Regulator is now the building control authority for higher-risk buildings in England. A higher-risk building is one of at least 7 storeys or at least 18m with at least 2 residential units, or a hospital or care home. For work to an existing higher-risk building, BSR approval must be obtained before work starts unless the work is exclusively exempt work, competent-person-scheme work or emergency repairs, and proceeding without approval is a criminal offence. Work is classified Category A - external modifications, structural changes, fire safety work, alterations to common areas, and following the 1 April 2026 update expressly including internal or external remediation work - or Category B for everything else. Applications must be evidenced by completed surveys and inspections, with the assessment methodology and findings documented, and unsupported assumptions are explicitly discouraged. The practical effect on refurbishment is direct: it front-loads survey cost and programme onto exactly the buildings where access, decant and occupancy make surveying hardest.

Energy retrofit is the part of refurbishment that has been most heavily systematised in the last decade, and for a specific reason: left to itself, measure-by-measure installation damages buildings. The framework that exists now is a response to that, and it is worth understanding as a system rather than as a set of certificates to be collected.

The controlling standard is PAS 2035:2023, "Retrofitting dwellings for improved energy efficiency - Specification and guidance", which is the current version as at August 2026. It became the required standard for all retrofit projects registered from 31 March 2025, and PAS 2035:2019 assessments could not be uploaded where inspection or handover dates fell after 29 March 2025, with transition and conversion mechanisms published by TrustMark and DESNZ across March to June 2025. Alongside it sits PAS 2030:2023, "Installation of energy efficiency measures in existing dwellings - Specification", which governs the installer side. One caution on PAS 2030: BSI Knowledge shows two records for this version, one dated 30 September 2023 and one 31 August 2024. The 2024 date appears to be a publication-record artefact rather than a separate amendment, and this page does not assert a 2024 edition.

The changes PAS 2035:2023 made to the 2019 version matter on site. The Risk Path procedure has been removed. A medium-term improvement plan is now mandatory for every property, not optional. Distressed replacements of heating appliances are brought into scope, which closes the gap where a failed boiler was replaced outside the retrofit system entirely. Airtightness testing is mandatory only where a target has been set. Significance assessments for historically important buildings gain predefined survey checklists. And the most consequential change of all: design responsibility now rests firmly with a qualified Retrofit Designer rather than the Coordinator, and in-person inspection by the Coordinator is mandatory where measures connect to or penetrate other fabric elements.

The role structure is the operational core of PAS 2035 and each role does a distinct job. The Retrofit Assessor surveys and reports the dwelling. The Retrofit Coordinator runs the project end to end, gathers inspection evidence and holds the compliance thread. The Retrofit Designer designs the measures and now carries design responsibility for them. The Retrofit Evaluator checks performance after the fact. Underneath them sits the PAS 2030-certificated installer who actually does the work. TrustMark is the government-endorsed quality scheme that registers businesses, holds the data warehouse, operates the insurance-backed guarantee and runs the audit and remediation process, including the find-and-fix programme for ECO4 and GBIS with a standard eight-week remediation period.

Where fabric-first retrofit goes wrong, the failure mode is almost always moisture rather than heat. Adding insulation without addressing the vapour and air-pressure regime moves the dew point into the construction. The recognisable outcomes are interstitial condensation within or behind insulation; surface condensation and mould at thermal bridges, particularly at reveals, floor and wall junctions and party wall abutments; trapped moisture in solid walls where an impermeable internal insulation system has been applied to a wall that needs to dry inwards; reduced background ventilation after window replacement without trickle vents; and rainwater ingress through poorly detailed external wall insulation. These are not exotic failures. They are the predictable consequence of treating a dwelling as a set of independent measures rather than as a single moisture and ventilation system.

That is exactly what the PAS 2035 whole-dwelling assessment, the ventilation assessment and the now-mandatory medium-term improvement plan exist to prevent. The technical standard sitting behind the moisture risk assessment is BS 5250:2021, "Management of moisture in buildings - Code of practice", published on 29 July 2021 and currently listed as current by BSI, superseding BS 5250:2011+A1:2016. One point to get right, because it circulates incorrectly: BS 40104:2025, "Retrofit assessment for domestic dwellings - Code of practice", published on 24 September 2025, is a retrofit assessment code of practice. It is not a moisture standard and it does not replace BS 5250.

The consequences of getting this wrong have moved from technical to political. The National Audit Office reported in October 2025 on energy efficiency installations under the Energy Company Obligation, and DESNZ has since extended ECO4 explicitly to allow remediation of non-compliant installations. DESNZ published "Reforming Consumer Protection for Home Upgrade Schemes" on 17 June 2026, with the consultation closing on 10 September 2026. It proposes a new integrated consumer protection service under DESNZ oversight, potentially delegated to a Warm Homes Agency, absorbing many functions currently carried out by TrustMark and MCS; binding agreements with certification bodies; clearer expectations for PAS 2035 Assessors, Coordinators, Designers and Evaluators; scheme access conditional on meeting system-wide quality expectations; and a single complaints and case-management function with an independent ombudsman as the final safeguard. Earliest implementation is from 2028. Anyone building a business around the current TrustMark and MCS arrangement should be reading that consultation rather than this page.

Commercial fit-out has its own vocabulary and it is worth being precise about it, because the same three terms are used to mean slightly different things by landlords, agents and contractors. Shell and core means the structure, envelope and core are complete - lifts, stairs, risers, primary plant, WCs and landlord areas - with the tenant floorplates left bare. Category A is the landlord base finish that makes the space lettable: raised floor, suspended ceiling or an exposed-services equivalent, general lighting, primary HVAC distribution and terminals, fire detection, sprinkler grid, blinds and basic finishes. Category B is the occupier fit-out on top of that: partitioning, meeting and collaboration space, joinery, feature finishes, secondary mechanical and electrical services and controls, AV and IT, catering, furniture and branding.

A fourth term has appeared and is now common in the market: Cat A+, sometimes called plug-and-play, a landlord-delivered furnished product sitting between the two so a tenant can move in without running a fit-out project. It is not a formally defined term in any standard we could verify, so if it appears in a lease or a specification, get it defined in writing rather than assuming a common understanding. The current industry reference for fit-out generally is the BCO Guide to Fit-Out, third edition, published in 2025 and the first update since 2011, covering material selection, procurement, in-use management, smart technology, wellbeing and sustainable design. We could not open the BCO definitions of Cat A and Cat B directly, so the descriptions above are settled market usage rather than quotations from a standard.

The strip-out itself follows a sequence that is dictated by safety and by what needs to survive. A refurbishment and demolition asbestos survey and any resulting removal comes first. Then isolation and making safe of live services, which on a partially occupied building is the item most likely to be underestimated. Then soft strip of furniture, loose fittings and floor coverings. Then mechanical and electrical strip, done carefully both to reclaim value and to preserve landlord infrastructure that the next tenant will need. Then partition and ceiling removal, then raised floor removal, then any structural or slab works, then reinstatement to Cat A and finally Cat B. Duty of care documentation, waste carrier licensing, hazardous waste consignment for asbestos, fluorescent tubes, gas-filled units and WEEE, F-gas recovery from removed cooling plant, and site waste segregation run through the whole of it.

Dilapidations is where the commercial exposure sits at lease end, and it is a distinct professional discipline. The reference document is the RICS professional standard "Dilapidations", England and Wales, 7th edition, September 2016, effective from 1 December 2016 and reissued in December 2023 as an RICS professional standard, having previously been a guidance note. It covers interim schedules served during the term, terminal schedules at or after lease end, quantified demands, responses and Scott schedules, diminution valuations, the pre-action Dilapidations Protocol for terminal claims, break clause compliance, alternative dispute resolution, and the distinct roles a surveyor can take as adviser, expert witness or dispute resolver.

The commercially decisive point is section 18(1) of the Landlord and Tenant Act 1927, which caps damages for breach of a repairing covenant at the diminution in the value of the reversion. Where a landlord intends to strip out and redevelop anyway, the cost of the works is therefore not automatically recoverable, and a terminal schedule priced at cost can substantially overstate the real exposure. Reinstatement obligations arising from licences to alter are a separate contractual limb and are frequently the larger number: a tenant who took a Cat A floor and built Cat B on it may be required to return it to Cat A, including reinstating ceilings, floors, lighting grids and HVAC terminals that were removed years earlier.

Reuse and the circular economy have moved from aspiration to procedural requirement, at least in some jurisdictions, and there are four routes worth naming. The EU route is enforcement-adjacent: the EU Construction and Demolition Waste Management Protocol, revised and republished in August 2024, incorporates guidelines for pre-demolition and pre-renovation audits, building on the European Commission "Guidelines for the waste audits before demolition and renovation works of buildings" of 2018. The standards route is BS EN 17680:2023, "Sustainability of construction works - Evaluation of the potential for sustainable refurbishment of buildings", which gives a European method for assessing whether and how to refurbish rather than replace, with further work in development under CEN/TC 350/SC 1 on circular economy in construction and its UK mirror committee B/558/1 on building and material passports, circularity assessment, product reuse and pre-demolition audits. Those SC 1 outputs were still in development at the time of writing, so no EN numbers are quoted for them here.

The UK route is currently planning-led and London-centric. The GLA London Plan Guidance on Circular Economy Statements requires a Circular Economy Statement, comprising a written report and a template spreadsheet, for all planning applications referable to the Mayor under London Plan Policy SI 7, with pre-demolition and pre-redevelopment audits named as the tool for establishing what can be reclaimed. Boroughs are encouraged to extend the requirement below the referable threshold, local planning authorities are advised to secure a post-construction report by condition or obligation, and the GLA publishes annual monitoring. The certification route is BREEAM: BREEAM Refurbishment and Fit Out Version 7 launched on 15 July 2026, the first major overhaul since 2015, replacing UK RFO 2014 and International RFO 2015 and merging the UK and international schemes. It splits into two standards, BREEAM Refurbishment V7 for fabric, structure and core systems, and BREEAM Fit Out V7 for interiors, finishes and localised services, with emphasis on whole-life carbon, circular materials strategies, performance-based energy and climate resilience. Indoor air quality credits move to Hea 04, with an Indoor Air Quality Plan a minimum requirement at Very Good and above. There is a three-month registration window under the older versions, and projects registered under RFO 2014 or 2015 then have five years to submit for certification.

Structural alteration on an existing building is a different activity from structural design on a new one, and the difference is that the properties of the thing you are working on have to be established rather than specified. You do not choose the strength of the existing masonry, the depth of the existing footing or the span of the existing floor. You find them, and then you design around what you find. Everything in this section follows from that.

The governing standard for temporary works has changed recently and it is worth getting right, because a lot of documentation still refers to the withdrawn version. BS 5975:2019 has been split and replaced by two standards: BS 5975-1:2024, "Temporary works. Management procedures for the control of temporary works. Code of practice", and BS 5975-2:2024, "Temporary works. Falsework: design and implementation. Code of practice". BSI Knowledge records BS 5975-1:2024 with a publication date of 31 December 2024 and status current, and BSI announced both standards in a press release dated 25 February 2025. Part 1 carries what were sections 1 and 2 of BS 5975:2019, the procedural control; Part 2 carries what was section 3, falsework design. The procedural roles that the industry knows as the Designated Individual, Temporary Works Coordinator and Temporary Works Supervisor are the long-standing BS 5975 framework, but we could not verify from a primary source that BS 5975-1:2024 retains those exact titles, so check the document rather than the training slide, and do not quote clause numbers without reading them.

The companion practical guide is CIRIA C740, "Structural stability of buildings during refurbishment", published in 2017 with errata issued in July 2020. It deals specifically with props, needling, propping sequence and the underlying proposition that every plan to dig near or under existing foundations, remove a wall, or open up a roof must be scrutinised by a competent person before work starts. Temporary works design is a design activity with its own designer, its own design brief, its own check category and its own permit-to-load and permit-to-strike regime, and under CDM 2015 the temporary works designer is a designer with designer duties.

That last point has a commercial edge worth stating plainly. Under a design and build or construct-only arrangement the contractor typically carries temporary works design. But the information held by the permanent works designer about the existing structure - loads, capacities, assumptions - is an input the contractor cannot generate on its own. That is why refurbishment disputes so often turn not on the temporary works design itself but on who was supposed to establish the capacity of what already exists.

The sequence for forming an opening in a load-bearing wall is well established and the failures are equally well established. Survey the load path above, including any point loads, purlins, floor spans and party wall interaction. Design the permanent beam and its bearings and padstones. Design the temporary works: needles through the wall at calculated centres, propped both sides onto adequate bearing, with attention to the floor below and to the load path all the way down to the ground. Obtain building control approval and, where relevant, the party wall award. Execute in the designed sequence. Grout or dry-pack the beam bearing, and let it gain strength, before de-propping. The classic failures are props landing on inadequate substrate or on unsupported suspended floors, needles at excessive centres, de-propping before the dry-pack has gained strength, and unrecognised loads - most often a wall that was also restraining a flank wall.

Underpinning is chosen by ground conditions, load, depth and access rather than by preference, and the methods are genuinely different in what they can achieve. Mass concrete pin-and-pier is the cheapest and most common domestically, but it is depth-limited, labour-intensive and depends entirely on strict hit-and-miss bay sequencing. Beam and base suits variable ground and heavier loads and reduces the number of excavations under the wall. Piled underpinning is what you use when competent bearing strata are deep, when groundwater precludes open pits, or when tree-related clay shrinkage requires founding below the influence zone. Jet grouting treats the ground rather than replacing the foundation, and is used where excavation is impractical. The overarching UK code is BS 8004:2015+A1:2020, "Code of practice for foundations", which supersedes the withdrawn BS 8004:2015 and cross-references the European execution standards.

Movement monitoring is the control that tells you whether any of it is working, and the management structure is the same regardless of technique. Precise levelling to fixed datums, crack-width monitoring with tell-tales and demec studs, automated total station monitoring of prisms, tilt and inclinometer beams on adjacent structures, and for deep excavation adjacent to sensitive assets, vibrating-wire piezometers and inclinometers in the ground. What matters is not the instrument but the framework around it: predicted movement, trigger levels set at green, amber and red, and a pre-agreed response for each level, agreed with the asset owner - and for railway or highway assets with the infrastructure manager - before work starts rather than after the first amber reading.

Facade retention means keeping the external envelope, usually a masonry street elevation, while demolishing and rebuilding everything behind it. It is done for a small number of reasons that recur: planning or heritage policy resists loss of the streetscape; the facade is listed or sits in a conservation area; the townscape contribution is the thing the local planning authority is actually protecting; or the existing structure is functionally obsolete on floor-to-ceiling height, floorplate depth or servicing while the elevation is not.

CIRIA C740 characterises it bluntly, and the characterisation is worth carrying into any programme discussion: facade retention is described there as an extreme example of partial demolition, with little if any internal construction retained, requiring substantial temporary works. That is the correct mental model. It is not a conservation exercise with some propping attached. It is a demolition with a very large, very long-duration temporary works project wrapped around the one element that has to survive.

The reference guidance is two CIRIA documents, and their age needs stating up front so readers can judge it. CIRIA C579, "Retention of masonry facades - best practice guide", by Lazarus, Bussell and Ross, published in 2003, is the design-side document, covering planning and strategy, health and safety, hazard and risk assessment, and a list of contributory factors in failures of retained facades that remains the primary published source on how these things actually go wrong. CIRIA C589, "Retention of masonry facades - best practice site handbook", also 2003, is the site-side companion. Both are 2003 and we found no replacement.

The retention systems in use divide by where the steelwork sits, and the choice is usually driven as much by the footway and the highway authority as by structural preference. External freestanding scaffold-based retention uses tube-and-fitting or system scaffold towers outside the facade, tied through, with raking or flying shores. It takes the pavement and needs a highway licence. Internal retention puts a steel frame inside the building line, which frees the footway but occupies the construction zone and complicates the permanent works sequence. The combined internal and external arrangement, often called a sandwich, clamps the wall between internal and external steelwork with through-ties at floor levels, and is the usual answer for tall or slender facades. Vertical spine and mast systems, flying shores across a street to an opposite building where a third-party agreement can be obtained, and temporary works integrated with the permanent frame - where the new structure is erected early and the facade tied to it progressively before the temporary system is struck - complete the range.

The risks are specific and they compound. The demolished floors were providing diaphragm action and restraint, and their removal leaves a high-slenderness wall freestanding against wind loads that exceed anything it experienced in its service life. Differential movement between the retained facade, sitting on old shallow foundations, and the new frame, sitting on piles, is the single most common source of post-completion cracking. The existing masonry has low tensile capacity and may be deteriorated exactly where the ties need to go. Underpinning or new basement excavation below the retained wall adds a second movement mechanism. Water ingress into an unroofed structure degrades the wall while the works proceed. And the temporary works stand for a long time - months to years - which introduces load reversal, tie relaxation and corrosion over the retention period, none of which are issues on a six-week propping job.

Facade retention also sits directly in the heritage argument. It frequently requires listed building consent and will be tested against NPPF chapter 16 on whether the harm is substantial or less than substantial and whether it is justified. Conservation officers increasingly resist facadism as delivering the appearance of conservation without its substance, on the reasonable ground that significance often resides in plan form, internal structure and joinery rather than in the elevation alone. The whole-life carbon case now cuts both ways too: retention saves the embodied carbon in the facade, but the temporary works steel and the extended programme carry their own, and that trade-off is increasingly being quantified rather than asserted.

Listed building work is the one area of refurbishment where getting the consent wrong is a criminal matter rather than a commercial one, and that alone justifies treating it differently from the rest of the programme. In England the legislation is the Planning (Listed Buildings and Conservation Areas) Act 1990, as amended - notably by the Enterprise and Regulatory Reform Act 2013, which introduced listed building heritage partnership agreements, local and national class consents, and certificates of lawfulness of proposed works.

Listed building consent is required to alter, extend or demolish a listed building in a way that affects its character or appearance as a building of special architectural or historic interest. It is granted by the local planning authority, there is no application fee, and carrying out unauthorised works is a criminal offence for which individuals can be prosecuted, with authorities able to require reversal of what has been done. The grades are I, II* and II, and consent is required for all of them. The grade affects the weight given in decision-making, not the need for consent - a point that is misunderstood on site more often than any other in this area.

Historic England does not grant consent. It provides advisory planning services, guidance and standards, enforcement support and technical advice, and is a statutory consultee on defined categories of application. Scheduled monuments are a separate and stricter regime again: written consent must always be obtained before any work on a scheduled monument, and where an asset is both scheduled and listed, scheduled monument consent takes precedence.

The policy test sits in NPPF chapter 16. The current NPPF is the December 2024 version, with the GOV.UK page showing a 12 May 2026 update. Decision-makers must identify and assess the particular significance of any affected heritage asset, including its setting, proportionate to the importance of the asset. Great weight should be given to the conservation of the asset, and the more important the asset, the greater that weight should be. Substantial harm or total loss requires clear and convincing justification, and the threshold rises with the asset: for Grade II listed buildings such harm should be exceptional; for scheduled monuments, Grade I and II* listed buildings, protected wreck sites, registered battlefields, Grade I and II* registered parks and gardens and World Heritage Sites it should be wholly exceptional. Less than substantial harm must be weighed against the public benefits of the proposal. Those are the words the case will be argued in, and a design that cannot be described in them is a design that has not addressed the test.

The four UK nations diverge and the divergence is procedural, not cosmetic. In Scotland, Historic Environment Scotland administers consent for scheduled monuments and is the national body for the historic environment, while listed building consent is administered by planning authorities with HES as consultee. In Wales, scheduled monument consent is applied for from Cadw rather than the local authority, and where a monument is both scheduled and listed, scheduled monument consent is required rather than listed building consent. The Welsh framework has changed materially: the Historic Environment (Wales) Act 2023 came into force on 4 November 2024, supported by regulations laid before the Senedd in September 2024 including the Applications for Scheduled Monument Consent (Wales) Regulations 2024. It is a consolidating code of law for the Welsh historic environment, and one provision worth knowing is section 89(2), which enables an application for listed building consent to be made after work to a listed building has been carried out. We could not verify whether the 2023 Act is fully in force in all respects. In Northern Ireland the Department for Communities Historic Environment Division administers scheduled historic monuments and scheduled monument consent, while listed building consent sits with councils under the Planning Act (Northern Ireland) 2011.

The conservation standard itself is BS 7913:2013, "Guide to the conservation of historic buildings". It is under revision at BSI under standards development project 9025-12038, with a proposed scope covering sustainability, climate change and energy efficiency and clarifying its application to unprotected buildings as well as statutorily protected ones. There is no target publication date and no published successor, so there is no such thing as a BS 7913:2025 and anyone citing one is citing something that does not exist. For energy retrofit of historic buildings specifically, the operative route is PAS 2035:2023, which requires a significance assessment for historically important buildings, now supported by predefined survey checklists.

At the civils end, refurbishment stops being called refurbishment and becomes repair, renewal, rehabilitation, strengthening or asset management. The vocabulary changes but the defining fact does not: something is kept, and it therefore has to be inspected, assessed, protected and worked around. What is different is the scale of the consequences, the presence of a regulated asset owner, and the fact that the asset is usually still in service while the work goes on.

Concrete repair has its own standard family and it repays understanding as a system rather than as a set of product specifications. The BS EN 1504 series, "Products and systems for the protection and repair of concrete structures - Definitions, requirements, quality control and evaluation of conformity", runs to ten parts: Part 1 definitions; Part 2 surface protection systems; Part 3 structural and non-structural repair; Part 4 structural bonding; Part 5 concrete injection; Part 6 anchoring of reinforcing steel bar; Part 7 reinforcement corrosion protection; Part 8 quality control and evaluation of conformity; Part 9 general principles for use of products and systems; and Part 10 site application of products and systems and quality control of the works. Two edition years are verified: BS EN 1504-9:2008, published 31 July 2009 and current, and BS EN 1504-10:2017, published 24 October 2017 and superseding the 2003 edition. The Concrete Society gives Part 1 as January 2006, Part 2 as October 2004 and Part 3 as February 2006. We could not verify current edition years for parts 4, 5, 6, 7 and 8 from a primary source, so no years are given for those parts here.

Part 9 is the one that matters conceptually, because it sets the framework of principles and methods from which a repair strategy is selected. Rather than defaulting to patch repair and hoping, the strategy is chosen against a defined principle: protection against ingress, moisture control, concrete restoration, structural strengthening, increasing physical resistance, resistance to chemicals, preserving or restoring passivity, increasing resistivity, cathodic control, cathodic protection, and control of anodic areas. The first group of these addresses defects in the concrete itself; the second group addresses corrosion of the reinforcement, which is a different problem with different solutions. Choosing a repair method without first identifying which principle you are working to is how the incipient anode effect gets designed into a job, where a patch repair reinstates passivity locally and drives corrosion at the perimeter of the patch instead. Outside the European standards sphere the reference is ISO 16311, "Maintenance and repair of concrete structures", all four parts published in 2024: Part 1 general principles, Part 2 assessment of existing concrete structures, Part 3 design of repairs and prevention, and Part 4 execution of repairs. Part 2 has no direct BS EN 1504 equivalent, which makes ISO 16311 worth knowing even on UK work.

For highway structures in the UK the governing documents are in the Design Manual for Roads and Bridges. Inspection is covered by CS 450, "Inspection of highway structures", at version 0.1.0 dated 29 April 2021, superseding BD 63/17. It defines the inspection regime that structures teams work to: general inspection, visual and without special access, at 24-month intervals; principal inspection, a close examination within touching distance of all accessible parts, at 72-month intervals with a risk-based extension to 12 years for certain structures; safety inspection; special inspection, which requires Technical Approval Authority agreement; and inspection for assessment, which is governed by CS 454. Repair of deteriorated concrete is covered by CS 462, "Repair and management of deteriorated concrete highway structures", Revision 0 of March 2020, superseding BA 35/90 and BA 52/94, which references BS EN 1504 parts 3–9, BS EN ISO 12696, BS EN ISO 15257 with cathodic protection designers required to hold Level 4 in reinforced concrete, and ASTM C876. One flag on CS 462: NBS lists it as current, but the copy served from Standards for Highways carried a superseded marking, so verify its status before relying on it. Cathodic protection itself is covered by CD 370, Revision 2, replacing BA 83/02.

A general caution about DMRB references. These documents are versioned continuously - version 2.5.0 and the like - and are revised without renumbering, so a document number alone is not a citation. Any DMRB version number should be checked against standardsforhighways.co.uk at the time you use it. This page names the documents and gives version numbers only where they were verified.

The typical bridge refurbishment work packages follow from all of that. Concrete repair and patch repair to BS EN 1504 principles. Impressed-current cathodic protection or sacrificial anodes under CD 370 and BS EN ISO 12696. Bearing replacement, where the jacking scheme is itself a significant temporary works design with a high check category, because you are lifting a live structure to a designed load and a designed lift. Expansion joint replacement under CD 357, usually the highest-nuisance package because it requires deck-level possession. Deck waterproofing renewal under CD 358, which is the single most effective durability intervention available, because the majority of chloride-driven deterioration arrives through failed waterproofing and leaking joints rather than from the surface. Strengthening by externally bonded or near-surface-mounted fibre-reinforced polymer, by additional or replacement post-tensioning, by plate bonding, or by reconfiguring the articulation. And management of post-tensioned structures under CS 465, where grout voiding and tendon corrosion are the concerns that drive the inspection regime.

Tunnel refurbishment is dominated by one constraint above all others: possession. The health and safety standard is BS 6164:2019, "Health and safety in tunnelling in the construction industry - Code of practice", which came into effect on 31 October 2019, superseding BS 6164:2011, and expressly includes safety considerations relating to maintenance, renovation and repair of shafts and tunnels, which is the refurbishment-relevant point. The typical scope is lining repair - grouting behind segments and masonry, resin injection, sprayed concrete secondary lining, segment replacement, and repointing and stitching in Victorian brick and masonry assets - invert renewal and drainage, waterproofing and leakage management, mechanical and electrical renewal covering ventilation, lighting, fire detection and suppression, SCADA, communications and power, and portal, shaft and cross-passage works. Nearly all of it is executed in short absolute possessions with fixed handback times, and that reverses the normal productivity logic. The design is optimised for installation rate and handback certainty rather than for material cost, which is why tunnel refurbishment specifications look expensive per unit and are not.

Water assets bring a regulatory structure of their own. For reservoirs in England the Reservoirs Act 1975 applies, with GOV.UK guidance "Reservoirs: owner and operator requirements" last updated on 23 February 2024. Large raised reservoirs holding 25,000 m³ or more above natural ground level must be registered with the Environment Agency, and the EA designates as high-risk those where an uncontrolled release could put life at risk. Three separate qualified civil engineer roles run the safety regime: the construction engineer for design, construction or alteration; the supervising engineer providing continuous supervision, an annual statement and the flood plan; and the inspecting engineer, who inspects at least every ten years and specifies measures in the interests of safety with deadlines, then certifies their completion. GOV.UK states these three roles should be carried out by different people. Owners must implement the measures specified by the inspecting engineer within the deadlines, and failure risks prosecution. The EA runs an ongoing Reservoir Safety Reform Programme, and we could not verify the status of any resulting 2025 or 2026 legislative reform.

Treatment works and service reservoirs are a different problem again, and the dominant characteristic is that the process cannot stop. There is no shutdown of supply or discharge available, so temporary process, bypass and standby arrangements are the enabling works and frequently the largest single package. On top of that sit drinking water quality compliance for any material in contact with potable water, confined space entry with gas and biological hazards, coatings and cathodic protection in concrete and steel tanks, and roof and column replacement in service reservoirs where the roof is the thing that has failed and the structure below it has not. Highways renewal completes the picture and behaves like tunnels: pavement renewal by inlay, overlay, deep reconstruction or concrete pavement maintenance, drainage renewal, vehicle restraint system replacement to modern containment classes, and lighting and technology renewal, all delivered inside narrow contractually enforced possession windows and lane rental regimes that push the cost structure toward pre-assembly, rapid-cure materials and night working.

The differences are not a matter of degree. They are structural, and they show up in sequence, in risk, in logistics and in the way the work gets priced. Anyone moving from new build to refurbishment for the first time tends to underestimate all four in the same way.

Start with sequence. On new build the sequence is chosen: you decide how to phase the frame, when to make the building watertight, how to overlap the trades. On refurbishment the sequence is largely imposed by what exists. You cannot demolish a wall before propping it. You cannot prop it before you know what it carries. You cannot open up before the refurbishment and demolition asbestos survey. And there are things you cannot survey until you open up. The result is a programme made of hard, irreducible dependency chains with very little float-generating parallelism, which means a single late discovery propagates through the whole chain rather than being absorbed by rescheduling something else.

Then unknowns, which is the defining economic feature. Concealed conditions - the actual structural capacity, hidden asbestos-containing materials, unrecorded alterations, live services with no as-built record, differing ground, deteriorated fabric behind sound-looking finishes - are irreducible at tender stage, because full opening-up requires possession the contractor does not yet have. The consequences follow logically. Bidders either price a contingency they cannot substantiate and lose the job, or price it out and rely on claims, in which case the client pays anyway and pays with an argument attached, or the parties agree a mechanism that shares it. That is why survey investment before tender has a far higher marginal return on refurbishment than on new build, and why the DfE test - what knowledge gap does this close, and how will the result change the answer - is the right discipline for spending the survey budget.

Occupied and live buildings add a whole category of constraint with no new-build analogue. Noise, vibration and dust limits and working-hour restrictions. Segregation and a fire strategy for a building with compromised compartmentation and possibly disabled detection. Maintaining means of escape at all times, not most of the time. Permit systems for hot work, isolations and confined spaces. Separate access and welfare. Security and safeguarding on schools, hospitals, custodial and retail sites. And business-continuity obligations that can override construction logic entirely, because the occupier trading on Saturday is worth more than your Saturday shift. Hospital and laboratory refurbishment adds infection control and pressure-regime management on top.

Where full possession is impossible, work is broken into phases with decant - moving occupants into swing space, temporary accommodation or a completed phase. Decant multiplies interfaces, because each phase acquires its own enabling works, temporary services, temporary fire strategy and handover. It extends the programme. And it introduces a class of cost that simply does not exist on new build: the cost of the temporary state itself, which buys nothing permanent. In housing refurbishment, decant carries statutory and tenancy obligations as well as cost.

Access and logistics compound all of it. Constrained urban sites with no laydown area, no tower crane position, existing lifts and stairs as the only vertical transport, floor-loading limits on the existing structure that govern what plant and materials can go where, pavement licences, and the basic fact that everything has to be delivered and removed through openings sized for a building that was never designed to be rebuilt. Waste egress is as much of a constraint as material ingress, and on a strip-out it is usually the harder of the two.

Temporary works dominance is the technical signature of the whole discipline. On new build, temporary works serve the construction of the permanent works. On refurbishment, the temporary works frequently are the engineering: propping schemes, needling, facade retention systems, shoring, edge protection on a structure that has no edge protection design, temporary roofs and scaffolds, and temporary services. The proportion of design effort and design risk sitting in BS 5975-1 territory is far higher than on new build, and critically, that design depends on properties of the existing structure that have to be verified rather than specified.

All of which is why the cost and programme risk behave differently in kind, not just in size. New-build cost risk is broadly symmetric and quantity-driven: you know what you are building, the question is how much of it and at what rate. Refurbishment cost risk is asymmetric and discovery-driven. The downside tail is long - asbestos found, structure weaker than assumed, RAAC identified, contamination encountered - and the upside is capped, because nobody discovers that the structure is stronger than they hoped in a way that saves real money. Programme risk behaves the same way, and because refurbishment sequences are serial rather than parallel, delay is rarely absorbed. Add long-duration temporary works, possession windows and occupier constraints, and the distribution of outcomes is materially different from a new build at the same contract value.

That has a visible effect on the contract forms observed in use, and this is a description of the market rather than a recommendation. Because scope cannot be fully defined at tender, the market is observed to move away from fixed lump sum and toward mechanisms that price work after it is known. RICS practice information "Appropriate contract selection", 2nd edition, April 2024, records that the JCT Prime Cost Building Contract is for projects requiring an early start on site, often for alterations or urgent repair work such as fire damage, where the exact nature and extent of the work is not known until the project is underway, with the contractor paid prime cost plus a contract fee; RICS notes that this carries higher cost risk for the employer, since the outcome depends on how efficiently the contractor works. The same document records JCT Intermediate and Minor Works being used for refurbishment procured traditionally on a fixed-price lump sum where scope is defined, and the Measured Term Contract suiting employers with a regular flow of maintenance, minor works and improvement projects.

On the engineering side the NEC4 Engineering and Construction Contract main Options map onto the same gradient: Option B, a priced contract with bill of quantities, is a remeasurement mechanism though the contractor still carries the risk of the rates; Options C and D are target contracts sharing pain and gain; Option E is cost reimbursable, with the lowest contractor risk and the highest client risk. The observed pattern across UK civils refurbishment - bridges, tunnels, water assets - is a lean toward Options B, C and E rather than the lump-sum Option A, precisely because quantities of repair are not knowable until the structure is exposed. Concrete repair in particular is commonly let on measured rates against provisional quantities, with the actual area of breakout established by hammer-tap and half-cell survey after the access has been erected. None of that is advice on what to use. For what each of these forms actually is, who publishes it and where each is commonly used, see our reference page on /construction-contracts/.

BuildPedia would rather tell you where the evidence runs out than round it off, and on this subject the evidence runs out in a lot of places. Standards in the refurbishment space are revised quietly, withdrawn without announcement, and republished under new numbering. The following were open at the time of writing, and each of them is a point where you should check the primary source rather than this page.