On paper, a lighting project can look perfectly defined. Fixtures are specified, layouts coordinated, details drawn, levels calculated.
Then comes the construction site.
A material chosen during design is substituted. The composition of a pane of glass changes. A downstand beam or retained building services emerge in an existing building. A ceiling isn't built to the geometry that was drawn. A product becomes unavailable, while the handover date does not move.
The site doesn't necessarily call the quality of the design into question. It confronts it with reality.
What experience teaches us is that the economics of a lighting project cannot be reduced to its CAPEX : equipment, installation, design and initial programming. It also includes its OPEX : energy consumption, maintenance, replacement and subsequent interventions. Decisions made between design and construction directly affect both dimensions.
No Lighting Decision Acts Alone
Light is tied to almost everything that shapes how a space is perceived: its proportions, materials, colours, furniture, ceilings, façades, and the details that connect them. A change that seems minor can therefore carry consequences far beyond its own scope.
A few centimetres of deviation in a ceiling's geometry can shift a line of light or unbalance an architectural composition. A change in the thickness or composition of a pane of glass, or a few millimetres of deviation in an installation detail, can transform, or even destroy, a grazing light effect that was carefully studied. A source meant to disappear within a detail can become visible again if execution tolerances shift.
In our practice, examining a lighting-related issue sometimes reveals an anomaly whose impact extends beyond lighting. This does not extend the lighting designer's remit, as design lead for the lighting scope, to overall site coordination, which remains the responsibility of the lead architect. But the work requires a precise reading of the relationships between geometry, material, distance and viewpoint: light can thus reveal a gap between the architectural intent and the result taking shape.
Flagged early enough, this gap can still be studied and corrected. Deferred until after handover to protect the completion date, it becomes rework: dismantling, modifying, reordering, reinstalling and readjusting. In the meantime, the space may have opened with a compromised result already fixed in its first photographs.
The Cost Doesn't Start With Rework
According to feedback from some of our construction partners, errors related to lighting can account for up to 10% of construction costs. The published studies cited here, by contrast, concern construction rework across all trades¹, with estimates that vary according to the projects, scope and measurement methods. The Construction Industry Institute guide, published in July 2011, gives a range of 2 to 20% of contract value². A study by Peter E. D. Love, published online in October 2025 and subsequently in the January 2026 issue of the Journal of Construction Engineering and Management, reports pre-completion rework costs that are lower on average than earlier estimates, but 300% higher than the amounts initially reported³.
On a commercial project, anonymised here, rework arising from architectural execution and installation errors amounted to approximately 15% of the construction budget, borne by the contractor. The client, meanwhile, had to operate for several months in premises with defects, then close for several days to allow repairs. The opportunity to save lay upstream: identifying deviations, deciding how to address them and intervening before they required dismantling and rebuilding.
This figure relates to a single case and is not a generalisable rate. It nevertheless illustrates the economic shift that occurs when deviations are only corrected after execution. Their cost then extends beyond the repair bill alone, whoever pays it: it includes additional design time, further consultations, urgent deliveries, remobilisation of contractors, and sometimes impacts on the schedule and operation of the premises, or even on brand image when photographs or media features published after handover show a result still marred by defects.
Prototyping Before a Detail Becomes Irreversible
In projects where light and material shape visual identity together, the prototype is a design tool and a means of reducing risk. It may address an assembly, an integration detail, or a full section of façade, ceiling or furniture. When the result depends on the precision of a joint, the behaviour of a material or the concealment of a source, full-scale work at 1:1 may be necessary.
We do not use it merely to check that a fixture switches on: we observe the actual distribution of light, its continuity, the visibility of the source, reflections, shadows, glare, fixings and perception from positions of use. Calculations, simulations and renders remain indispensable, but they cannot fully reproduce the behaviour of a real material, manufacturing tolerances or the perception of a source within its built environment.
A prototype retains its value only if it is produced before ordering or series production, and if the tested components match those that will be built. If any of those components changes, the effect must be reassessed; otherwise, the prototype no longer serves to choose, only to observe.
Tender Documentation Isn't the End of Design
The tender documentation (DCE in French) formalises the project at a given point in time, but conditions continue to evolve: details developed further by the architect, finishes still under discussion, execution methods proposed by contractors, and constraints revealed on site.
The issue is not change itself, but the late discovery of its consequences for lighting, after ordering, manufacturing or closing up the works. Continuity in lighting design allows the project to be reassessed as soon as new information appears.
In contracts awarded in separate lots, particularly in public procurement, a lack of coordination between packages creates unnecessary costs. On an office project with several open-plan floors, the blinds and lighting packages each included their own occupancy detection, with no shared strategy for exchanging this information between the two systems. Both sets of devices were therefore installed. Coordination at an earlier stage could have avoided this duplication; once the equipment had been ordered, correcting it required rework. The additional cost came less from the product itself than from a control decision that had not been resolved before ordering.
This continuity can also be weakened when the lighting designer's appointment ends before tender review or follow-up during construction. Design intentions that are still evolving may then be simplified or abandoned without reassessment, causing the project to lose part of the value already invested in the design phase.
Certain points therefore require a deliberate pause: before ordering a family of fixtures, repeating a detail, closing up a ceiling, or beginning series production. A real test can then verify the layout, light distribution, orientation, relationship with materials, and conformity of the submitted product with the studied specification.
The lighting designer's review of these elements, within the scope of their brief, has to happen while a change is still physically and economically possible.
A Cheaper Product Isn't Necessarily a Saving
Substituting fixtures is one of the decisions where apparent savings and real savings diverge most easily.
Two fixtures with similar luminous flux and wattage are not necessarily equivalent. Their photometric distribution, spectral quality, source luminance, glare control, optics, dimming, mechanical precision, finishes and lifespan can produce very different results. In luxury environments, consistency of colour temperature, material rendering and physical appearance between visible fixtures becomes part of the project itself.
A substitution should therefore be assessed on the expected result, its integration, and its full cost, in terms of both CAPEX and OPEX, not just the difference between two purchase prices. The net saving equals the initial saving minus the cost of additional studies, prototypes, adaptations, accessories, maintenance risk, and any eventual rework. If the replacement requires modifying a ceiling or piece of furniture, or leads to dismantling and reordering after handover, the announced saving can disappear very quickly.
Technical equivalence must also be logistical. The contractor proposing the substitution must demonstrate availability of the required quantity, stock reservation, manufacturing and delivery lead times, and coordinated delivery of power supplies, optics, accessories and replacement parts. A solution unavailable when needed is not equivalent; nor does immediate availability justify a loss of effect or integration.
The lighting designer examines the technical and perceptual conformity of the alternative within the scope of their brief, while responsibility for quantities, procurement, stock reservations and lead times remains with the contractor proposing and supplying the equipment.
On-Site Tests Prepare the Final Fine-Tuning
Final fine-tuning can act only on what was designed to remain adjustable. It cannot compensate for poor light distribution, a source that has become visible or a layout incompatible with the architecture. Nor can it recreate the effect of a specified fixture if the installed product does not share its characteristics.
Room for adjustment must therefore be built in from the design stage and preserved through execution: aimable fixtures, interchangeable optics or accessories, spare equipment, properly distributed circuits and a control system that allows levels to be adjusted. This flexibility becomes essential when final materials can only be assessed once installed.
Final fine-tuning establishes a hierarchy between planes, balances luminances, controls reflections and adjusts the relationship between light, material and use.
Final Fine-Tuning Also Requires an Eye on Site
On projects requiring lighting control system programming, the programmer's competence and presence are decisive. They must master the project's conceptual and technical requirements and take part in fine-tuning visits with the lighting designer, client and architect.
Yet this joint presence is not systematic. Follow-up budgets set too early sometimes lead to the assumption that photographs, videos and online exchanges are sufficient. They cannot replace direct observation of the result in the space itself. On site, the lighting designer assesses that result against the project's intent, checks the relationship between light, materials and use, and works with the programmer to define the necessary adjustments. This work becomes all the more important during commissioning as deadlines tighten.
Keeping the Decision Open Until the Right Moment
A well-coordinated project is not one that never changes, but one whose changes remain legible enough to be tested, decided on and absorbed before they turn into rework.
The lighting designer's value also lies in the continuity of their design oversight: following interfaces, requesting the necessary prototypes, examining alternatives and preserving room for adjustment through to commissioning.
Early decisions.
Rework avoided.
Savings preserved.
© MI. Bouhlel, Atelier dada. All rights reserved.
The texts, concepts, and reflections published on this website may not be reproduced, adapted, or distributed without prior authorization.
References
¹ Hwang, B.-G., Thomas, S. R., Haas, C. T. and Caldas, C. H. | Measuring the Impact of Rework on Construction Cost Performance
² Construction Industry Institute | A Guide to Construction Rework Reduction | Guide IR252-2b, 1 July 2011.
³ Peter E. D. Love | Quantifying the Costs of Field Rework in Construction | Journal of Construction Engineering and Management, vol. 152, no. 1, January 2026; published online on 28 October 2025. DOI: 10.1061/JCEMD4.COENG-17026.