Facade Retention & Restoration

Facade retention means keeping an external wall - usually a masonry street elevation - standing and supported while the structure behind it is partly or wholly demolished and replaced. Restoration, on this page, means the repair of the facade fabric itself: cleaning, repointing, stone repair and replacement, and the remaking of architectural terracotta, faience and cast stone. The two halves meet constantly in practice. A retention scheme fixes ties into masonry whose condition is a restoration question; a restoration contract on a retained wall is programmed around scaffolding and temporary works that exist for engineering reasons. BuildPedia's refurbishment reference page covers facade retention in one section as part of the wider refurbishment sequence; this page is the deeper treatment of the specialist trade, and it does not repeat the refurbishment page's four-nation heritage consent table or its retrofit and strip-out material.

One boundary is worth drawing early. The sector's scope includes recladding and external wall remediation - the fire-safety-driven replacement of modern cladding systems - but that work belongs to the cladding and curtain walling discipline and its own regulatory stream, and it is not expanded here. This page is about loadbearing masonry facades and the traditional materials they are faced with.

The engineering half of the trade has a small and stable published core. CIRIA C579, "Retention of masonry facades - best practice guide", by Lazarus, Bussell and Ross, was published in 2003 from CIRIA research project 626 and remains listed as current; its published scope runs from planning and strategy, health and safety and hazard and risk assessment through investigation and appraisal of facades, temporary facade retention structures and work on site to cleaning, repair and case studies - which is close to a table of contents for this page. Its companion, C589, "Retention of masonry facades - best practice site handbook", also 2003, repackages the guide for site use: checklists, references and induction material aimed at temporary works coordinators, site agents, engineers and safety advisers. Both are paid documents. The free document is CIRIA C740, "Structural stability of buildings during refurbishment", downloadable in full, which treats facade retention within the wider problem of keeping partly dismantled buildings standing and characterises it as an extreme example of partial demolition, with little if any internal construction retained, requiring substantial temporary works. Alongside these sit BS 6187:2011, the current code of practice for full and partial demolition, published 31 December 2011, and BS 5975, the temporary works code, republished on 31 December 2024 as two parts - Part 1 on management procedures and Part 2 on falsework design - replacing BS 5975:2019.

The craft half has its own standards. BS 8221-1:2012 is the code of practice for cleaning of natural stone, brick, terracotta and concrete - its published scope covers the principal cleaning methods in use, the characteristics of surface deposits, and materials from natural stones and clay brick to calcium silicate brick, glazed and unglazed terracotta and faience, and cast concrete products including cast stone. BS 8221-2:2000 is the companion code for surface repair of natural stones, brick and terracotta, with published coverage including repointing and the selection of mortars, repair of masonry, terracotta and faience, surface coatings and consolidants. BS 7913:2013, the guide to the conservation of historic buildings, describes best practice in managing and treating historic buildings whether or not they carry statutory protection, and remains the current edition. The building lime itself is specified under BS EN 459-1:2015, currently marked as under review on the BSI record.

The regulatory picture is layered in a familiar way. Construction work, including demolition and temporary works, sits under CDM 2015, enforced by HSE, which remains Great Britain's workplace safety regulator on site. Demolition has its own statutory notice regime under the Building Act 1984, described in the demolition section below. Planning and heritage consents - the reason most facades are retained at all - are covered in their own sections. On building regulation more widely, the current institutional position needs stating precisely because it changed recently: the Building Safety Regulator has been a standalone body since 27 January 2026, when the Building Safety Regulator (Establishment of New Body and Transfer of Functions etc.) Regulations 2026 (SI 2026/20) came into force, transferring the building safety functions previously exercised through the Health and Safety Executive. The BSR is sponsored by the Ministry of Housing, Communities and Local Government, and GOV.UK announced the change as a step towards a single construction regulator. It is no longer part of HSE, and descriptions that place it there are out of date - while HSE, separately, continues to publish and enforce the construction safety guidance this page cites.

On site, the two halves meet through investigation, and C579's published scope puts investigation and appraisal of facades ahead of its temporary works chapters for a reason. The observed sequence follows it: a measured survey of the elevation; a condition survey of the fabric recording decay, past repairs, cracking, bulges and lean; a structural appraisal of what the wall actually is - thicknesses, bonding, lintel types, bonding timbers, previous alterations - commonly with opening up, because drawings of Victorian buildings describe intentions, not outcomes; and a desk study of the building's history, since every alteration in it changed the load paths. The delivery side then splits along the same line as the knowledge: temporary works specialists and demolition contractors on one side, conservation contractors, masons, terracotta makers and accredited professionals on the other, commonly sharing one scaffold and one programme. The interfaces are managed rather than discovered - ties that pass through zones a conservator must later repair, cleaning water that runs over frame fixings, monitoring targets glued to stones that are due for replacement - and a scheme in which the engineering and the conservation teams first meet on site is a scheme that has already paid for that meeting twice.

Finally, what this page is not. It is not a design guide and it does not answer "which retention system is best" or "which cleaning method is right for this wall" - the published guidance does not answer those questions either, because they turn on the investigated condition of a specific building, its ground, its neighbours and its consents. Where this page names a document, the document is the authority; where it names a regulator, the regulator's current published position governs.

Facades are retained for a short list of reasons that recur on project after project, and most of them are written into the planning system rather than chosen by the developer. The first is the conservation area. Conservation areas have existed since the Civic Amenities Act 1967 and there are now nearly 10,000 in England; within them, demolition or substantial demolition of a building usually requires planning permission from the council, and carrying out demolition in a conservation area without permission is a criminal offence. Where an unlisted building makes a positive contribution to a conservation area's significance, national policy treats its loss as harm in its own right. A street elevation that the planning authority regards as part of the area's character is, in practical terms, very hard to remove and comparatively easy to keep.

The second reason is listing. Works to alter, extend or demolish a listed building in a way that affects its character as a building of special architectural or historic interest require listed building consent from the local planning authority; there is no application fee, unauthorised works are a criminal offence for which individuals can be prosecuted, and the authority can require unauthorised work to be reversed. Consent is needed for grade II as much as for grade I - the grade affects the weight a decision-maker gives the building, not the need for consent. Where redevelopment of a listed building is contemplated at all, retention of the principal elevations is commonly the shape a consentable scheme takes.

The third is national policy. Chapter 16 of the National Planning Policy Framework requires great weight to be given to the conservation of a designated heritage asset - the more important the asset, the greater the weight - and requires clear and convincing justification for any harm. In the version current at the time of writing, substantial harm to a grade II listed building should be exceptional, and substantial harm to assets of the highest significance - scheduled monuments, grade I and II* buildings and their equivalents - should be wholly exceptional, with consent refused unless tightly drawn tests are met. Less than substantial harm is weighed against public benefits. Decisions where listed buildings, their settings and conservation areas are in play must also address the statutory duties in the Planning (Listed Buildings and Conservation Areas) Act 1990 - the government's planning practice guidance points in particular to sections 16, 66 and 72. Retention of a facade is frequently the mechanism by which a scheme's harm is argued down from substantial to less than substantial: the streetscape survives, and the argument moves to what is behind it.

Behind the policy sits the commercial logic. Buildings are commonly redeveloped behind their facades because the elevation is the part that works and the structure is the part that does not: floor-to-floor heights too low for modern services, floorplates too shallow or too cellular for modern occupiers, cores and risers in the wrong places, structures that cannot take the loads or the openings a new use needs. Retention lets a site deliver a modern building while keeping the townscape contribution - and, not incidentally, the consent - that the old elevation carries. The same logic appears at the modest end of the market: published case work from Liverpool's buildings-at-risk programme records terraces so decayed that safe internal access was impossible, where the facade was the only element that could be saved, retained under urgent works powers while the owner's redevelopment position was resolved.

The practice has critics, and the criticism has a name. Facadism - keeping the skin and losing the building - is resisted by conservation officers and amenity societies on the ground that a building's significance often lives in its plan form, interiors and structure as much as its street face, so that retention delivers the appearance of conservation without its substance. The SPAB's published approach is built on exactly this point: the fabric of a building is treated as the primary source of its meaning, which sits uneasily with schemes that keep one wall of it. The refurbishment page's facade retention section carries the whole-life carbon version of the same argument - retention saves the embodied carbon in the wall but spends carbon on long-duration temporary steel - and both arguments now appear regularly in planning committee reports. None of that changes the observed outcome: where policy protects a frontage and the building behind it has failed, retention is the tool the market reaches for.

Setting and the street add a further layer that catches schemes off guard. The planning guidance defines significance as deriving not only from an asset's physical presence but also from its setting, and expects cumulative change to be considered - which is why a facade's context in a terrace, crescent or square matters to the decision as much as the wall itself, and why set pieces are read as wholes. The same logic reappears later in the work: the Scottish cleaning guidance records that piecemeal cleaning of grouped buildings has produced some of the worst visual outcomes, because a streetscape argument does not stop at the party wall. And there is a final route into retention that has nothing to do with development at all: failure. Local authorities hold powers over listed buildings falling into disrepair - urgent works notices under section 54 of the 1990 Act with cost recovery under section 55, repairs notices under section 48, and the dangerous structures procedures of the Building Act - and the documented Liverpool case work shows those powers producing facade retention directly, with the authority stabilising the frontages of derelict listed terraces under urgent works notices because the only alternative was losing them. Historic England's Heritage at Risk register is the national census of the buildings this happens to, and buildings-at-risk programmes are where some of the most technically demanding retention work originates - on walls in the worst condition, with no paying development behind them.

Retention systems are classified by where the supporting structure sits relative to the wall, and the choice is driven as much by the footway, the highway authority and the demolition method as by structural preference. The refurbishment page's facade retention section summarises the families; the deeper picture is about what each family costs in land, licences and interfaces.

External systems put the whole temporary structure outside the facade: braced towers or cantilever frames, in proprietary steel or structural scaffold, stabilised against overturning by kentledge - dead weight, typically concrete blocks - or by anchors and holding-down details. They leave the site interior clear for demolition and construction, which is their great virtue, and they occupy the footway and sometimes the carriageway, which is their great cost. Everything about the public side then becomes a design constraint. Published case work from Liverpool shows both ends of the range: at Seel Street a cantilever braced frame was deliberately sized so that its projection into the highway avoided a full road closure, with the kentledge positioned away from the building line to avoid surcharging cellars and coal chutes below the pavement; at Shaw Street a structural scaffold retention was founded on concrete pads cast in the lightwell and on the footway, with props inside the arched cellar carrying the load down under the portico. A supplier's published checklist adds two recurring points: services under the footway can dictate what foundations the retention can use, and where the footway must stay open the frame needs a designed pedestrian portal through it.

Internal systems reverse the trade. The steel goes inside the building line - frames, towers and walings within the footprint - freeing the highway but occupying exactly the space the demolition and the new structure need, which turns the temporary works into a sequencing problem: the frame must come out in stages as the permanent structure arrives, and every stage change is a designed condition. The combined arrangement, often called a sandwich, clamps the wall between external and internal members with through-ties, and is the usual answer for tall or slender facades where one-sided support cannot develop enough stiffness. In the Liverpool case work the walings were bolted together through existing structural openings - windows, in plain terms - to clamp the wall without drilling weathered face masonry, with timber packers taking up the line of the elevation.

Two further families complete the set. Flying shores and horizontal trusses span between the retained facade and something else that can take the load - a party wall, a neighbouring building under a third-party agreement, or another retained elevation across the plot; a supplier's guidance notes that the most economical schemes commonly use the existing building's own walls as restraint where they can. And integrated schemes bring the permanent works forward: the new frame is erected early, close behind the wall, and the facade is tied to it progressively so the temporary system can be struck sooner. That saves hire duration - the dominant cost on schemes that run for years - at the price of welding the demolition, temporary works and permanent works programmes into a single dependency chain.

Equipment divides between proprietary modular systems and engineered structural scaffold, with fabricated steelwork for the heavy or awkward cases. The proprietary route - modular soldiers, shoring towers and raking props from the temporary works hire market - is fast to erect and adapt and is priced by the week, which reads differently on a six-month job than a three-year one; the Liverpool evaluation records a scheme where outright purchase of a structural scaffold was the economical answer precisely because nobody could say when it would come down. Whatever the kit, the wall side of the interface is the same problem: the frame has to make contact with old masonry without damaging it, through spreader timbers and packers, at levels chosen to match where the wall actually has strength.

Under every external scheme sits a foundation question that is easy to underrate. Kentledge works by weight, and it needs ground that can carry weight exactly where city pavements are least reliable: cellars, vaults and coal chutes commonly run out under the footway, and the Liverpool case work records kentledge positioned deliberately away from the building line to avoid surcharging basement retaining walls, cellar lights and chutes - while at the second scheme the load went the other way, onto props inside an arched cellar and onto concrete pads cast in the lightwell and footway. Buried services occupy the same ground: a supplier's published checklist notes that utilities under the footway can dictate what foundations a retention system can use at all. The building itself sets the other half of the equation - its weight, height, slenderness and condition decide how much frame, at what stiffness, bearing where - and the observed pattern is that trial pits, service tracing and cellar surveys before the scheme is drawn cost a fraction of redesigning a founded, licensed, half-erected frame after.

The highway dimension runs through all of it. Scaffolds and similar structures on a highway or its footway need a licence from the highway authority, which can impose conditions; councils commonly issue scaffold and pavement licences alongside the demolition consultation, and HSE's demolition guidance expects road closures to be discussed with the highway authority before tender where collapse consequences could reach the road. Impact from vehicles is a documented concern for retention structures adjacent to traffic - a CROSS safety report records both the gap in general scaffold guidance and the fact that C579 addresses impact loads on temporary facade retention systems directly, with physical barriers spaced clear of the structure as the observed protection measure.

A retained facade is a temporary works project of unusual length and consequence, and the roles around it follow the UK's general temporary works framework. The procedures document is BS 5975, which was republished on 31 December 2024 as two parts, replacing BS 5975:2019: Part 1 carries the management procedures for the control of temporary works - the procedural controls, training and role definitions applied so that temporary works are conceived, designed, specified, constructed, used and dismantled in a controlled manner - and Part 2 carries the design of falsework, with content that the BSI record notes can be applied to the design of other temporary works. The published audience list for Part 1 names temporary works designers, permanent works designers, temporary works coordinators, contractors, clients, suppliers and manufacturers, which is a fair census of who touches a retention scheme.

The two roles that matter daily are the temporary works designer and the temporary works coordinator. HSE's temporary works guidance describes the arrangement in plain terms: there are no specific legal requirements about how temporary works are managed, but an effective way of managing them is to appoint a temporary works coordinator to ensure suitable designs are prepared, checked and correctly erected on site - formally appointed, with adequate authority to carry out the role, including stopping the work if it is not satisfactory. On complex projects the same guidance describes this person as commonly a chartered civil or structural engineer with relevant experience, based on site. The designer produces the scheme; the coordinator controls its life on site - erection, use, alteration and removal - and the interfaces with everything else the site is doing. The law underneath is not role-specific: any temporary structure must be designed, installed and maintained to withstand any foreseeable loads which may be imposed on it, and used only for the purposes for which it was designed, installed and maintained, with inspection by a competent person on a regular basis.

Checking is where facade retention sits apart from routine temporary works. HSE's guidance on temporary works notes that in all but the simplest situations a design with calculations is needed and that in more complex works the design will also need to be checked by an independent party; its structural stability guidance goes further for this trade specifically, stating that where a building needs restraint, facade retention - whether fabricated steel or scaffold construction - must be designed by a competent person, erected, inspected, maintained and modified in line with the designer's instructions, and will likely require independent third-party checks of the temporary works. BS 5975's graded design-check practice is the machinery the industry uses to deliver that; the grades themselves are in the standard, which is a paid document, so this page does not restate them.

CDM 2015 wraps around all of it. Demolition and structural alteration must be planned and carried out by competent people to avoid unplanned collapse; clients must pass on what they know about the structure - stability, structural form, significant design assumptions, suggested work methods and sequences - and principal designers and principal contractors plan and manage the pre-construction and construction phases respectively. CIRIA C740, written for exactly this territory, shows the failure modes the framework exists to prevent: partial demolitions that turn into unplanned demolition, apparently non-structural strip-out that removes hidden stability, props placed without design and loaded eccentrically. Its facade retention case study describes the observed working pattern on a well-run scheme: full investigation and survey of the existing structure first, a retention scheme designed by the temporary works designer, installation by a competent contractor, and extensive crack and movement monitoring throughout.

Wind is the load case that defines the design, and it is treated seriously in the current documents. BS 5975-2:2024's published change list includes wind: maximum wind forces are now calculated on a likely erection duration of one year - formerly two - with a ten-year return period, and the part adds information on the effect of adjacent tall structures, funnelling of terrain and buildings, and susceptibility to vortex shedding, with its loads conforming to BS EN 1991-1-4, the Eurocode for wind actions. Two things follow for retention schemes. First, a facade frame that will stand for three years is outside the standard short-duration assumption, which is a design decision to be made visibly, not an accident. Second, C740's technical guidance carries a blunt cross-check for structures in the temporary condition: lateral strength in any direction adequate for a disturbing force equal to 2.5% of the weight supported, carried in addition to the normal code checks - a deliberately simple test for a condition with few redundancies. Duration then works on everything else: load reversal as demolition and rebuild change what the frame carries, relaxation of ties, corrosion protection specified for years rather than weeks, and periodic inspection and re-tensioning written into the scheme rather than left to memory.

What the roles exchange is information, and the deliverables observed on well-run schemes are specific. The temporary works designer works from the investigation - the measured and condition surveys, the wall's real thicknesses and lean, the ground, cellar and services picture - and returns more than a frame drawing: erection and striking sequences, permitted conditions at each stage, tie positions and loads, foundation loads, the inspection and maintenance regime, and the monitoring arrangement with its limits. The coordinator then holds the scheme's life against that document: checks before each stage is loaded, control of alterations - on a structure that stands for years, every small change is a designed change - and the formal points at which demolition, construction or striking is permitted to proceed. C589 exists precisely because this discipline lives or dies at site level: its published purpose is to put the guide's checklists, references and induction material into the hands of the coordinator, agent and engineer as part of the site documentation, in a format written to be used outside the office.

The wider ecosystem is small and useful. The Temporary Works Forum - a membership body promoting temporary works best practice, and the body HSE's own page points to - publishes and discusses across the field, and sponsors the Temporary Works Toolkit, a series of articles in The Structural Engineer written primarily to help permanent works designers deal with temporary works; parts 1 to 19 are free to read and cover the ground from CDM responsibilities and integration of permanent and temporary works to propping, needling and demolition. For a permanent works engineer meeting a retention scheme for the first time, that series and C740 are the accessible way in; C579 remains the trade's own book.

The defining condition of facade retention is the empty shell. A masonry street elevation was never designed to stand alone: its floors provided restraint at every level and acted as diaphragms carrying wind back to cross-walls and party walls, its returns propped its ends, and its own weight sat in compression paths that assumed all of that was there. Demolition removes the lot. What remains is a tall, slender wall of old masonry with low tensile capacity, possibly out of plumb, possibly decayed exactly where it matters, carrying wind pressure and suction over its whole height with no floors left to share the load. C740's guidance for the temporary condition lists what lateral capacity has to cover: wind, eccentric loads that cause sideways lurching, out-of-vertical walls needing support, accidental impact where there are known risks, and unreliable foundations - and its case studies show what the shell looks like when nobody has done that arithmetic, including an 18m frontage left standing with no returns after every cross-wall and floor had been stripped out.

The connections between frame and wall are where the scheme succeeds or fails, because old masonry is a poor anchorage and a worse one near its weathered face. Observed practice works with the wall rather than against it: horizontal walings on both faces bolted to each other through existing structural openings, clamping the masonry instead of relying on drilled fixings; vertical timber packers taking up the irregular line of the elevation; support lines set at floor levels, where bonding timbers, lintels and the wall's own construction give it something to push against; and ties detailed for removal and re-fixing, because demolition and construction will both need to pass through the zones the ties occupy. The published supplier guidance adds the quiet requirement that the frame must bear on the facade without damaging it - a spreader-and-packer detail question that decides whether the retention scheme is itself a restoration problem. Where the wall carries parapets, chimneys, cornices or loose face material, those elements are commonly secured separately - strapping, dowelling and localised repair before the shell condition arrives - because the frame holds the wall, not the things sitting on it.

Weather protection is part of the same condition. A retained facade spends months or years as the outside face of nothing: rain enters the exposed wallhead and the floor-joist pockets, saturates the core of the wall, washes lime out of bedding joints and feeds frost damage through winters in which the building no longer heats itself. Temporary capping to wallheads, protection to pockets and openings, and drainage that takes water somewhere deliberate are standard items in the site-side guidance; the Liverpool case work shows the end state where decades of that exposure went unmanaged - upper masonry too far gone to retain, taken down against a photographic record with the lintels and handmade bricks salvaged for reuse.

Monitoring is how the scheme knows it is working, and the published practice is specific about its shape even where the numbers are project-specific. C740 describes the arrangement as setting up a movement monitoring regime and agreeing the response if the limits set are exceeded - the limits and the response are agreed before work starts, not negotiated during an alarm - and elsewhere describes conservative acceptance values for movement and cracking with work stopping and support being reviewed when they are reached. Its facade retention case study records extensive crack and movement monitoring as a feature of a competently run scheme. In observed practice the toolkit is survey targets on the wall read by total station, precise levelling points, crack gauges across existing defects, and increasingly automated instruments reporting continuously - the temporary works hire market now sells structural monitoring alongside the steel - with readings taken against baseline surveys of a wall whose existing cracks and leans were recorded before anything moved. The refurbishment page's warning about what the monitoring is for bears repeating here: the single most common source of post-completion cracking on retention schemes is differential movement between the retained wall and the new structure, which means monitoring matters most at exactly the stages - excavation, foundation loading, tie transfer, de-propping - when the temptation is to stop reading it.

The baseline matters as much as the readings. Observed practice records the wall before anything moves: a photographic record, a schedule of existing cracks and defects, lean measured storey by storey - partly because a monitoring regime without a datum measures nothing, and partly because the same record serves the party wall schedules of condition next door and, on consented buildings, the recording that conditions commonly require. Reading frequency then follows risk, rising through the stages that put energy into the ground or take restraint out of the wall, and the review loop is part of the design: the coordinator and designer see the readings, and the response to a breached limit - stop, prop, investigate - was agreed before the first reading was taken, which is the difference between a trigger and a surprise. One further physical effect belongs to long-duration steel: temperature. Published toolkit commentary on retention schemes records the trade-off that stiffening a scheme by increasing member sizes to reduce movement under load also increases its thermal movements, while stiffening through geometry does not - a reminder that a frame clamped to a masonry wall through three summers moves with the weather even when nothing else is moving at all, and that the monitoring record will show that cycle whether or not anyone expected it.

Impact and the public complete the shell condition. Retention structures stand next to traffic for years, and vehicle impact on a frame that is holding a wall over a footway has consequences a routine scaffold strike does not. The CROSS report on scaffolds and traffic protection records the reporter's practice - physical barriers such as temporary concrete units spaced at least 500mm clear of the structure so deflection can happen without contact - together with the expert panel's notes that C579 addresses impact loads on temporary facade retention systems and that the Highways Act licence regime is where the highway authority imposes conditions. On the site side, the pedestrian portal through an external frame, the lighting and hoarding around kentledge, and the exclusion of the public from drop zones are ordinary but permanent features of the years the structure stands.

Demolition behind a retained facade is ordinary demolition wearing a straitjacket, and the ordinary rules come first. Section 80 of the Building Act 1984 requires notice to the local authority before most demolition begins - the statute carves out small buildings and internal works, and HSE's guidance describes the notice as due at least six weeks before starting - with copies to the occupiers of adjacent buildings and to the gas and electricity undertakers. The authority can respond with a notice under section 81 imposing requirements, and councils commonly deal with scaffold and pavement licences, working time restrictions and traffic in the same conversation; HSE's guidance pushes clients to have that conversation before the contract is awarded, and to discuss any road closure with the highway authority before tender where a collapse could reach the road. In law demolition is construction work, so CDM 2015 applies in full - including the regulations on stability of structures and on demolition and dismantling, which require the work to be planned and carried out so as to prevent danger, with the arrangements recorded in writing before the work begins. The technical code is BS 6187:2011, whose published scope covers full and partial demolition, maintaining structural stability including temporary structural support, exclusion zones, and responsibilities through all phases. The National Federation of Demolition Contractors - the trade body, running its own training group and accreditation since 1941 - is where the specialist contractors sit.

Before anything structural moves, two surveys govern. The client must give the contractor a refurbishment or demolition asbestos survey by a competent surveyor - the intrusive kind, not a landlord's management survey - and a structural survey and assessment must establish the form, condition and load paths of what is to come down, because the method depends on knowledge of the structure: its age, alterations, stability-sensitive elements such as arches and overhangs, and where its stiffness actually lives. C740's guidance on soft strip is a catalogue of why this matters: walling assumed non-structural that turns out to carry a floor, panelling and boarding that were quietly providing lateral stability, deterioration hidden behind linings, and strip-out debris stored on upper floors until something fails. Behind a retained facade the same soft-strip mistakes have a sharper edge, because the wall being relied on for nothing is now the wall being kept.

The method itself is commonly top-down: progressively reducing the building floor by floor with small plant, which HSE's guidance describes as allowing control of both the stability of the structure and the build-up and spread of debris, and as often the preferred method for a town-centre site with little space - exactly the sites where facades are retained. The sequence is designed so that elements not critical to stability go first and the work proceeds towards the stiff elements, with the demolition plan documenting exclusion zones, debris zones and safe working areas that change as the building comes down. Two scaffold rules from HSE's stability guidance bear directly on retention sites: a standard tied scaffold will not restrain a building during demolition - it borrows its sideways stability from the building, not the reverse - and a sheeted scaffold acts as a sail, so any sheeted access scaffold relying on ties must be reduced in height with the building rather than left standing proud of it. The facade retention structure is the one part of the site designed to restrain anything, and the demolition method is written around never asking anything else to.

Two load paths deserve their own paragraph. Debris is the structure's own weight coming back as an imposed load: C740 lists the build-up of demolition arisings among the standard triggers for temporary support, alongside machinery operating on floors - and behind a retained facade both act on old floors whose capacity is an assessment, not a fact, which is why floor-by-floor methods specify what plant may stand where and how much material may accumulate before it is cleared. The second path is the environment the work exports: dust, noise and vibration reach the retained fabric, the monitoring instruments, the neighbours and the party wall schedules all at once. Vibration and dust monitoring now sit in the same hire catalogues as the propping systems, and method statements for demolition behind retained walls commonly zone the site - what may be broken mechanically, what must be cut or taken down by hand near the retained elevation, where the crushing and processing happen - so that the wall being kept is not shaken by the removal of everything it was attached to.

Then the direction of work reverses, and the new structure arrives with the problem that outlasts the project: movement. The retained wall commonly stands on its original shallow footings; the new frame commonly stands on piles or new pads taken to better ground. As the frame is loaded it settles by amounts the old wall will not match, and where a new basement is dug behind or beneath the retained elevation, excavation and underpinning add a second, independent movement mechanism. The refurbishment page records the blunt observed fact that differential movement between retained facade and new frame is the most common source of post-completion cracking on these schemes, and C740's client guidance makes differential vertical movement between new and existing structure a standard consideration on any project that marries the two. Where the excavation comes within 3m or 6m of a neighbouring structure below the defined planes, the Party Wall etc. Act 1996 notice machinery is engaged as well, with its own surveyors, awards and condition surveys.

The tie-in is the last designed act of the retention. The permanent connections between wall and frame - restraint ties at floor levels, bearings where new floors enter old masonry, movement joints where calculated movements have to be absorbed rather than resisted - are installed progressively, and the temporary system is struck in a designed de-propping sequence in which each stage hands restraint to permanent work that has been verified as ready to take it. Striking is not the end of the engineering: the monitoring regime commonly runs through and beyond it, because the load transfer is when calculated behaviour meets the wall's opinion. The facades that crack are rarely the ones that moved during demolition; they are the ones whose new supports settled after everyone went home.

Cleaning is the restoration operation most likely to be regretted, because its damage is irreversible and its benefits are partly cosmetic. The starting point in published guidance is not which method but whether. Historic Environment Scotland's advice on sandstone is unusually direct for a public body: aged masonry develops a patina - formed over decades by erosion, soiling, mineral change and biological growth - that is valued in itself and helps protect the stone, and the safest way to avoid cleaning damage is not to clean in the first place. The same guidance reports the cost of getting it wrong at national scale: Scottish research found cleaned buildings decaying six to ten times faster than uncleaned neighbours, with the worst cases needing 25% to 50% of the stone in the cleaned walls replaced. Cleaning can still be justified - where soiling is itself causing decay, where safety requires it, where a repair contract needs a clean substrate - but the trade's own literature treats it as an intervention to be argued for, not a default. There is also a streetscape dimension recorded in the guidance: piecemeal cleaning of grouped buildings - one terrace house scrubbed bright between sooty neighbours - has produced some of the worst visual results, which is one reason cleaning on listed buildings and in conservation areas commonly needs consent, and why in Scotland stone cleaning is explicitly treated as an alteration.

The reference standard is BS 8221-1:2012, the code of practice for cleaning of natural stone, brick, terracotta and concrete. Its published scope is the removal or partial removal of deposits from the original surface, covering the principal cleaning methods in use and the characteristics of surface deposits, across natural stones, clay and calcium silicate brick, glazed and unglazed terracotta and faience, and cast concrete products including cast stone. That scope list is itself a warning: the right method is a function of the substrate, and a facade commonly carries several substrates in one elevation.

The method families in observed use each carry a characteristic risk. Water-based methods - washing, nebulous sprays, steam - mobilise soiling gently but saturate the wall, with consequences for embedded ironwork, salts and drying time; the low-pressure superheated-water systems widely used on sensitive work are known across the trade by the DOFF name, one of the systems manufactured by Stonehealth, which has supplied masonry cleaning equipment for over thirty years to buildings up to and including Westminster Abbey, and which runs an approved-contractor directory for its systems. Abrasive methods - dry and wet grit blasting, rotating discs, and their gentler descendants including low-pressure vortex systems worked with fine granulate - remove soiling and surface together unless closely controlled; HES records that the highly damaging abrasive techniques of past decades have almost disappeared from reputable practice, while manual brushing with bristle - not wire - and water remains the one physical method described as relatively safe. Chemical methods - acidic and alkaline cleaners reacting with the soiling or the surface - rinse into porous stone as well as off it, and the documented consequences include irreversible bleaching and staining, salt blooms, pitting of susceptible minerals and accelerated greening. Newer techniques - latex poultices, bacterial cleaning, lasers - are treated in the Scottish guidance with the caution that the last generation of new techniques earned.

What separates competent cleaning from damage in the published practice is trials and records. The Scottish guidance describes a three-stage testing programme: inspection of the building's defects, surface variation, stone types and soiling patterns; test cleaning in areas that are both unobtrusive and representative of the full range of surfaces; and analysis and reporting of the results with photographs, kept for reference in case anything later goes wrong. The trial panel then becomes the standard the contract is measured against - method, dwell times, pressures, distances and operatives all fixed by what the panel proved. On consented buildings the method statement and trial results are commonly what the conservation officer approves; changing method mid-contract because the soiling turned out harder than expected is a consent question as well as a technical one.

Cleaning a retained facade adds the scheme's own constraints. The wall is at its most vulnerable exactly when access to it is best: saturating a masonry wall that is standing as a free shell, in winter, with its wallhead weather protection still temporary, is a different proposition from cleaning an occupied building, and run-off finds the ties, walings and fixings of the retention scheme before it finds a drain. Programming cleaning and repair around the retention stages - and around the months in which lime and water-based work can cure - is part of why these projects are planned in seasons rather than weeks.

Repointing is the most common restoration operation on any masonry facade, and the most commonly done wrong. The principle underneath it is that in traditional construction the mortar is sacrificial: lime mortar is more permeable and weaker than the brick or stone around it, so frost and salt action concentrate in the joints, which erode first and protect the masonry units - and are then renewed. Repointing in a strong, dense cement mortar inverts that arrangement. The joints become the least permeable part of the wall, moisture movement and salt action transfer into the faces of the bricks and stones, trapped moisture stops the wall drying, and the masonry decays faster while the pointing stays perfect. Both the SPAB and Historic Environment Scotland publish this mechanism in plain terms, and it is the reason lime is the default on pre-1919 solid-walled buildings in current conservation practice - not nostalgia, but moisture physics.

The published craft rules are specific. Repointing is premature until the mortar has weathered back to a depth about equal to the joint width, or is genuinely loose - soft mortar alone is not a defect, and unnecessary repointing damages arrises and destroys evidence. Preparation is decisive: joints raked out to around twice their width, by hand where discs would widen them, cleaned and dampened. New mortar matches the old - assessed from sheltered areas such as under eaves - in constituents, colour, texture and strength, and is designed weaker than the masonry units; well-graded sharp aggregate with the largest particles no more than a third of the joint width is the standard sand rule. Joint finishes are commonly flush or slightly recessed to shed water without proud edges; the deeply recessed pointing fashionable in the late twentieth century is now read as a dating feature rather than a model. Failures cluster around weather: new lime needs protection from frost, rapid drying and rain, which makes winter repointing a risk and protection and aftercare part of the specification. Sample panels and mortar trials before the main contract are standard practice, and on consented buildings commonly a condition.

The lime itself comes in families with different behaviour, specified under BS EN 459-1:2015 - the building lime standard, currently marked under review at BSI. Air limes (non-hydraulic lime, used as putty or dry hydrate) harden slowly by carbonation and stay soft and permeable; natural hydraulic limes (NHLs), burned from limestones with clay impurities, set partly by hydration and are sold in graded strengths; and hot-mixed mortars - quicklime slaked together with the aggregate on site, used warm or stored - are the method by which most historic mortars were actually made. The hot-mix revival is one of the notable movements in current practice: analysis of historic mortars commonly points to hot mixing, and the Scottish guidance lists the observed working advantages - stickiness, good fill from the slaking expansion, early stiffening, apparent frost tolerance and economy - alongside the safety realities of handling a caustic material that slakes at high temperature, with full protective equipment and specific first-aid provision on site. Gauged mixes - quicklime with a proportion of NHL - and traditional pozzolans such as brick dust are the routes to modest hydraulic set where exposure demands it. What the guidance repeatedly warns against is the builders'-merchant shortcut: bagged hydrated lime sold as a plasticiser for cement mortars is not a substitute for NHL or quicklime, and adding cement to lime mortars is now discouraged in the published advice.

Specification in practice starts from the wall, not the catalogue. Existing mortar is assessed - visually, and by analysis where the project justifies it - for binder, aggregate and mix; the exposure of the elevation, the season of the work and the condition of the masonry then shape the choice within the lime families. The same body of guidance is candid that copying the original mix is not always right, since the original may itself have failed, and that original cement mortars on later buildings can be as worthy of matching as lime on earlier ones - the test is compatibility with the wall as built, established case by case.

The knowledge infrastructure around lime is unusually good for a craft subject. Historic England's Practical Building Conservation volume on mortars, renders and plasters is the standard reference book; Historic Environment Scotland publishes free short guides and Inform leaflets on lime mortars, hot mixing and repointing that put the essentials in a site cabin for nothing; the SPAB runs a technical advice line and courses; and the Building Limes Forum - a charity with over 500 members across producers, suppliers, specifiers and users - is where the practitioners compare notes. For a facade project, the practical consequence of all this material is that mortar decisions are checkable: a specification that cannot say what the existing mortar is, what the new mix is and why, and how the work will be protected, is visibly short of the published standard of care.

Stone decay announces itself in recognisable forms - powdering and scaling surfaces, honeycombing, blistering, salt crusts, fractures crossing stones, blurred mouldings, stones eroded back so far that the mortar stands proud - and the first published principle of repair is restraint. Sandstone in particular can erode a long way before it threatens structure, and the guidance frames replacement around function: each wall face is expected to be structurally sound and to keep water out of the building, and the best return on repair money is at the stones that were designed to throw water off the face - copings, cornices, string courses, sills - and no longer do. An aged surface is not a defect; the marks of time are part of what listing and conservation area designation protect, and some replacement projects recorded in the guidance were carried out at great expense where brushing down loose material and repointing in lime would have done.

Where replacement is justified, the craft operation is indenting: cutting out the decayed stone to a sound, regular depth and setting in a new stone to match. The published practice is detailed and consistent. The replacement matches the original in size, bedding, porosity, mineral content and colour so far as the quarries allow; surface tooling - the angle and depth of chisel work, margins and grooves - is copied, but the new stone is not artificially distressed, because it will weather down to its neighbours within a few years; joints follow the original setting-out, staggered as the ashlar was, and are kept as fine as the original work; the new stone beds in lime mortar with voids packed solid; and cutting out is done without chipping the arrises of the stones that stay. Sourcing is the hard part. The number of active quarries is a fraction of what built the Victorian city, matches are commonly imperfect, and an unsuitable stone - harder, denser, differently bedded - can accelerate decay in the wall around it. The industry has partial answers: Stone Federation Great Britain, the trade body, maintains technical resources, a members' directory and a British stone sourcing tool covering over 80 indigenous stones, and its training arm and listed colleges feed the banker masons and fixers the work depends on. The one repair the guidance warns against consistently is the so-called plastic repair done badly: impermeable synthetic mortar patches spread across decayed faces trap water behind them, part company with the stone, and eventually fall off a facade into the street. Mortar repairs have a legitimate place - small, well-keyed, lime-based and weaker than the stone - and consolidants appear within the scope of BS 8221-2 for weathered surfaces, but the cheap skim that makes a wall look repaired is the signature of work that will be done again.

Terracotta and faience are a different trade inside the trade. These are fired clay units - faience being the glazed form - commonly hollow, filled and built in as blocks, carrying moulded ornament that brick and stone cannot economically match; they built the department stores, theatres, pubs and faience-fronted commercial streets of the late nineteenth and early twentieth centuries. Their repair logic differs from stone because the material's strength is its fired skin and glaze: once that surface is breached - by impact, salts, cramp corrosion or hard cement pointing - decay proceeds from the inside, and cutting a damaged unit back to sound material, as a mason would, only removes the best part of it. Replacement is therefore unit-by-unit remanufacture: survey and record each block, model or mould from the original profiles, allow for firing shrinkage, match the clay body and glaze against the weathered originals, fire, and fix back into the bonding pattern. That capacity barely survives in Britain. Darwen Terracotta in Lancashire - the largest UK specialist manufacturer, working in both terracotta and faience with over forty years of accumulated expertise and around 25,000 units a year - is the scale of the national resource, and its published projects show what the work looks like at its heaviest: over 8,400 decorated loadbearing blocks manufactured to match the grade I listed Royal Albert Hall for its new south porch, and colour-matched restoration work on glazed and unglazed fronts from York Art Gallery to churches and commercial buildings across three continents. Lead times for modelling, mould-making and firing put terracotta on the critical path of any programme that needs it, which is a planning fact as much as a craft one.

Cast stone - concrete cast to imitate dressed stone, common from the late nineteenth century onward - sits at the edge of the same family and inside the scope of BS 8221-1's cleaning code. Its repair follows concrete logic (matching mixes, reinforcement corrosion, patch adhesion) rather than quarry logic, and misidentifying cast stone as natural stone, or vice versa, is a standard survey error with expensive consequences: cleaning methods, repair mortars and replacement routes all differ. On a retained facade the identification questions all arrive at once - stone, brick, terracotta, faience and cast stone can share one elevation - which is why the restoration side of these projects starts, like the engineering side, with investigation: opening up, sampling, and knowing what the wall is before deciding what to do to it.

The consent framework around facade work is the same one that drives retention in the first place, applied at the level of method statements. Listed building consent is required for works that affect the character of a listed building as a building of special architectural or historic interest - which reaches cleaning methods, repointing mortars, stone replacement and the fixing positions of a retention scheme's ties, not just demolition. It is granted by the local planning authority, there is no application fee, and the government's planning practice guidance states the sanction plainly: undertaking such works without consent is a criminal offence under section 9 of the 1990 Act, and authorities can require reversal of unauthorised work. The same guidance records the newer instruments - listed building heritage partnership agreements, local and national class consents, certificates of lawfulness of proposed works - that can take repetitive consent applications out of long programmes. In conservation areas, demolition control and Article 4 directions do the equivalent work for unlisted buildings. Historic England does not grant consent: it advises, is a statutory consultee on defined categories of application, and publishes the guidance the negotiations are conducted in. Scotland, Wales and Northern Ireland run parallel but distinct regimes - the refurbishment reference page carries the four-nation table, and this page does not repeat it.

What the consent process feels like on a facade contract is a negotiation over method, conducted stone by stone. Conservation officers commonly condition consents on trial panels for cleaning and repointing, sample approvals for replacement stone and terracotta, mortar analyses, recording before and during work, and named specialist subcontractors or accredited professionals. The accreditation infrastructure exists for exactly this: the Conservation Register run by Icon lists accredited conservator-restorers, and for building-scale work the registers it points to - the RIBA Conservation Register for architects, the AABC register for architects accredited in building conservation, and the IHBC's HESPR listing for heritage service providers - are the standard answers to a condition that asks who will direct the work. None of this is bureaucratic decoration; it is how an authority with one conservation officer manages risk on a wall it cannot inspect daily.

Behind the consents sits the harder constraint: the people. Facade restoration depends on banker masons, fixer masons, carvers, terracotta modellers and mould-makers, lime plasterers and bricklayers fluent in lime - trades that train slowly and have thinned. The published evidence is consistent. Historic Environment Scotland states outright that the number of fully qualified stonemasons available for good quality work in Scotland is currently limited, alongside the shortage of matching stone; Heritage Crafts, the UNESCO-accredited charity for traditional craft skills, has ranked UK crafts by survival prospects in its Red List of Endangered Crafts since 2017, with the 2025 edition adding a further 20 crafts to the endangered categories and classifying every craft from viable to critically endangered and extinct. The response side is real but slow: Historic England runs training, webinars and work-based routes including heritage apprenticeships, and publishes material on delivering heritage construction skills and training through projects - the mechanism by which a large facade contract can be conditioned or procured to train as it builds; Stone Federation's colleges and the industry's competitions feed the masonry end; the lime bodies and the Engine Shed do the same for mortars. On the terracotta side the workforce question is starkest: the specialist capacity is measured in individual firms, and the craft of architectural ceramics survives commercially because restoration contracts keep kilns firing.

For a project, the practical reading of all this is lead time and dependency. Consents take months and condition the method; accredited specialists and craft subcontractors are booked ahead; replacement stone comes from a short list of quarries; terracotta is made to order by very few hands. The retention engineering commonly determines whether a scheme is possible; the consents and the craft supply determine when.

The alternative to retaining a facade is taking the building down and building again - sometimes with a facsimile front, sometimes without - and the differences between the two routes are structural, commercial and procedural at once. Understanding them is the closest this page comes to explaining why retention schemes cost what they cost.

Structurally, demolition and rebuild is one project; retention is three structures negotiated against each other. The retained wall is an old, imperfect, partly unknown structure whose condition is discovered progressively; the temporary works are a designed structure that must accommodate that discovery; the new frame is a modern structure whose foundations, stiffness and movements are unlike either. Every interface between the three is designed, monitored and eventually handed over. C740's characterisation carries the weight here: facade retention is an extreme example of partial demolition, with little if any internal construction retained, and requires substantial temporary works. A clean rebuild has none of those interfaces - and none of the townscape, none of the fabric, and commonly none of the consent.

Commercially, the differences concentrate in time and in items that a new-build estimate does not contain. The temporary works stand for months to years, hired by the week or bought outright, with inspection, maintenance, re-tensioning and monitoring running for the duration; highway licences, footway portals, impact protection and hoarding run with them. Demolition is slower and more careful behind a retained wall than in an open site - top-down, floor by floor, with debris controlled and plant loads checked against old floors. The restoration package - cleaning trials, repointing, stone and terracotta replacement - has its own specialist supply chain with lead times measured in months. Against all that, the retained scheme carries the value that justified it: the consent that a clean-sweep application might never have won, and an elevation whose replacement cost - if it could be replicated at all - would include the same scarce crafts at new-build quantities. The whole-life carbon comparison appears increasingly in planning submissions on both sides of the argument, retained fabric against temporary steel and extended programme, and the observed trend recorded on the refurbishment page is that the trade-off is now quantified rather than asserted.

Procedurally, rebuild simplifies consents and complicates them at the same time. Demolition of the whole building sharpens the heritage tests - substantial harm or total loss of a designated asset needs clear and convincing justification against the highest bars in national policy, and loss of a positive contributor harms a conservation area - while a rebuilt replica raises its own policy questions about authenticity that conservation officers approach warily. Retention, by contrast, multiplies process: listed building consent conditions on method, party wall awards, highway licences, building act notices, temporary works design checks and monitoring regimes, all running concurrently. There are also cases with no choice in them. The documented Liverpool schemes show the end of the spectrum where decay had gone too far for internal access, and the options were facade retention under urgent works powers or loss of the frontage; even there, part of one facade proved unsaveable and was taken down against a full photographic record, with its gutters, stone lintels and handmade bricks salvaged for reuse - the fallback position that separates a managed loss from a plain one.

What the two routes share is the discipline that runs through this whole page: investigation before commitment. The rebuild decision and the retention decision are both made properly only on a surveyed, opened-up, understood building - its structure, its materials, its movement history, its ground - and the recurring theme in the failure literature is not exotic schemes going wrong but ordinary assumptions doing so: the wall that was thicker on the drawing than in the flesh, the "non-structural" partition that was holding the front of the building, the settlement that was assumed to have finished a century ago. Retention is the route that turns those discoveries into engineering; demolition is the route that turns them into rubble. Both are legitimate. Which one a given building gets is decided, in the end, by the planning balance and the state of the wall - in that order.

BuildPedia would rather tell you where the evidence runs out than round it off. The engineering side of this trade is documented mainly in paid CIRIA guides and British Standards, and the craft side in a mix of free public guidance and paywalled standards, so parts of this page rest on publication records, free companion documents and published case work rather than the primary texts. The following were the open points at the time of writing, and each is a place to check the primary source rather than this page.