Evidence checked October 31, 2026. Numerical examples and construction scenarios are hypothetical. Engineering acceptance and construction decisions require qualified professionals. This article does not prescribe a structural mix or determine compliance for a project.
1. The environmental document meets the construction site
Colorado's Department of Transportation identifies a July 2025 revision to its environmental-product-declaration requirements and provides a construction bulletin explaining the change. New York's current embodied-carbon guidance discusses performance-based concrete specifications rather than fixed recipes. These public records show that lower-carbon construction is becoming a question of procurement and project evidence, not only laboratory invention. Colorado DOT EPD requirements, New York embodied-carbon guidance.
The construction site introduces a demanding test. A concrete mixture must be placed, finished and cured under real conditions, then provide the required service. A favourable carbon number cannot compensate for unsuitable performance. At the same time, an unnecessarily restrictive recipe can prevent suppliers from offering a mixture that meets the engineering requirements with lower emissions.
The central challenge is to compare equivalent functions. A cubic metre of one concrete is not automatically interchangeable with a cubic metre of another. Strength, exposure conditions, early-age requirements and intended service life can differ. A comparison that ignores those differences may select the lowest number while failing to select the appropriate product.
An environmental product declaration, or EPD, helps describe impacts under a stated method. It is not a certificate that the material is suitable for every structure, and it does not prove that the declared mix was the one delivered to the job. Engineering qualification, environmental accounting and construction quality control remain separate evidence streams.
This article follows those streams from the design brief through mix qualification, EPD comparison, delivery and curing. It explains why cement substitution is only one part of the story, why strength at a stated age matters, why a product-stage carbon reduction may differ from a whole-project reduction and why field records complete the claim.
The goal is a useful standard for public purchasers and readers. Lower-carbon concrete should be judged neither by a marketing adjective nor by a reflexive assumption that unfamiliar binders are unsuitable. It should be judged by evidence that the proposed mixture can provide the required service and that the claimed emissions comparison uses compatible boundaries and actual quantities.
2. Cement and concrete are different accounting objects
Concrete is a composite material. Its cementitious binder is one component among aggregates, water and other constituents. A change in the binder can substantially influence the environmental account, but a claim about cement should not automatically be presented as the same percentage reduction for the finished concrete or the entire structure.
The National Concrete Pavement Technology Center describes supplementary cementitious materials such as fly ash, slag cement, silica fume and calcined clays as ways to reduce the amount of portland cement needed in a mixture. The technical opportunity is real, but the appropriate material and proportion depend on the intended concrete performance. Cementitious-materials resource.
Consider a hypothetical concrete whose reported production impact is 300 kilograms of carbon-dioxide equivalent per cubic metre. Suppose a binder change reduces the binder-related contribution by 60 kilograms while all other assessed contributions remain unchanged. The concrete's total falls to 240, a 20 percent reduction. If the binder component itself fell by a larger percentage, that larger percentage should not be applied to the whole concrete without checking the arithmetic.
The same distinction applies at project level. Concrete may be only one part of a building's material inventory. A reduction in its product-stage impact does not equal the same reduction in the entire building's embodied impact. The project account needs the quantities and impacts of all relevant components under its chosen scope.
Terminology should therefore identify the object: clinker, cement, binder, concrete mixture, structural element or whole project. Each can be a legitimate reporting level. Confusion arises when an achievement at one level is promoted into a broader claim without carrying the other materials and processes through the calculation.
A supplier's declaration should also be matched to the product being purchased. A cement EPD does not replace a ready-mixed-concrete EPD if the procurement requires the latter. It can be an input to the concrete's lifecycle calculation, but batching, other constituents and applicable transport assumptions may still matter.
These distinctions help identify real opportunities. Reducing the impact of the binder can be important. So can reducing unnecessary concrete volume, improving structural design or preventing replacement through good durability. The best choice depends on the project's function, not on which stage offers the most dramatic percentage in isolation.
3. Specify the service before specifying the recipe
A performance specification describes what the concrete must do. A prescriptive specification can also constrain how it is made. Both approaches can contain justified requirements, but an inherited recipe may restrict lower-carbon options even when alternative mixtures could meet the actual engineering need.
New York's guidance describes performance-based specifications in terms of properties such as compressive strength, durability and workability rather than a fixed mix recipe. That is a useful distinction, not permission to omit essential requirements or delegate structural responsibility to a carbon calculator. New York performance-specification explanation.
A hypothetical pavement project might require a particular strength, resistance to relevant exposure and a reopening time. A specification that adds an unnecessarily high minimum cement content could limit the supplier's ability to optimize the mixture. The design team should examine whether that prescription is needed for a documented performance reason or is simply carried forward from an older project.
Conversely, removing every material constraint can be inappropriate where the performance tests do not capture an important risk or where field experience is limited. A performance approach still needs a robust acceptance framework. The question is whether the requirement is connected to the intended service and supported by evidence.
The specification should identify the exposure environment. A sheltered interior element and a bridge deck exposed to deicing salts face different conditions. A mixture's suitability cannot be established solely by matching one compressive-strength value. The required tests and construction controls should reflect the actual environment.
The construction sequence belongs in the brief as well. A structure may need sufficient early strength before formwork removal, loading or another activity. If the schedule can accommodate later strength development, that flexibility may expand available mixtures. If it cannot, the early-age requirement should be stated explicitly rather than discovered after a low-carbon mix is selected.
The most productive procurement discussion therefore begins before tender. Designers, suppliers and constructors can identify the non-negotiable performance requirements and the areas where alternatives are possible. That early clarity reduces the chance that an environmental objective is added late as an extra constraint with no time for qualification or adjustment.
4. Strength is measured at an age and under conditions
A compressive-strength result is associated with a specimen, preparation method, curing condition and test age. Saying that two mixtures have the same strength without those details can conceal a meaningful difference in how they develop performance. Early-age and later-age requirements may serve different construction and service purposes.
A hypothetical mixture A reaches the specified strength at 28 days, while mixture B reaches it at 56 days. If the project requires the strength at 28 days, the later result does not establish compliance. If the design and construction plan legitimately allow 56 days, rejecting B solely because it develops strength more slowly could unnecessarily eliminate an option. The decision belongs in the approved engineering requirements.
The National Academies' report on supplementary cementitious materials discusses a range of materials, reactions and performance characteristics. It provides evidence that these materials cannot be treated as one uniform substitute with a fixed effect. The report's detailed findings should be applied within their material and test contexts. National Academies SCM report.
Laboratory strength is also not a complete description of in-place concrete. Specimens may experience conditions different from the structure. Field quality control and curing records help establish whether the installed material had a reasonable opportunity to develop the intended properties. A successful laboratory mix qualification is necessary evidence for some claims, but it does not supervise the pour.
A useful submittal therefore identifies the test ages relevant to both construction and final acceptance. It records variability across batches rather than only the best specimen. The design team can then assess whether the mixture has sufficient margin and consistency for the project rather than relying on a single favourable result.
Carbon comparisons should preserve the same performance basis. A low-impact mixture meeting a lower strength requirement is not automatically a better substitute for a higher-strength mixture. Nor should a supplier be rewarded for producing much more strength than necessary if doing so increases material use without a service benefit. The comparison needs the required class and actual design function.
This is an engineering optimization problem, not a contest to maximize one property. The appropriate mixture balances the relevant strength development, durability, workability and construction constraints. Its environmental account becomes meaningful only after those requirements have been defined.
5. Curing connects the laboratory result to the structure
Concrete performance develops through processes that depend on moisture, temperature and time. Curing practices influence that development. A lower-carbon mixture cannot be evaluated only as a recipe in a report; the project must also provide the conditions required for the installed material to achieve its intended properties.
Caltrans' construction guidance directs attention to curing materials and equipment before concrete work begins and to prequalification where specified. This is a practical example of construction control accompanying material selection. It should be read as Caltrans guidance within its context, not a universal procedure for every project. Caltrans concrete construction guidance.
A hypothetical project may approve a mixture based on carefully cured laboratory specimens but place it during weather that increases drying risk. If the site lacks the planned curing resources or stops protection early, the field material may not reproduce the laboratory evidence. The environmental claim remains attached to a product whose intended service has not been adequately protected.
Curing also interacts with schedule. A programme that changes binder composition without adjusting the construction plan may create pressure to load or expose the concrete before it is ready. The solution is not automatically to abandon the mixture. It is to align the material's verified behaviour with the sequence of work and the necessary quality checks.
Internal curing is another technical approach with its own evidence. FHWA describes it as a way to mitigate shrinkage cracking in relevant concrete applications, including low water-to-cementitious-material mixtures. It is not a blanket substitute for all external curing or a guarantee of durability in every environment. FHWA internal-curing programme.
The public purchaser should therefore ask what curing assumptions underlie the qualification and how the site will verify them. Records can include the approved method, timing, weather conditions and deviations. The article need not provide operational instructions to recognize that those records are part of the performance evidence.
A successful low-carbon specification protects this link. It does not allow the environmental objective to be used as an excuse for weak construction control, and it does not allow avoidable field mistakes to become a blanket indictment of the material. The installed result must be evaluated against both the approved mixture and the conditions required to make it work.
6. Durability is exposure-specific
Durability concerns the ability to maintain required service under the conditions encountered. It is not a single intrinsic score that can be inferred from compressive strength alone. Different exposure mechanisms can matter in different structures, and a mixture that performs well against one may need additional evidence for another.
A hypothetical coastal structure may require attention to the movement of chlorides toward reinforcement. A cold-region pavement may need performance under freezing and deicing exposure. A mass placement may face thermal and cracking concerns. These examples illustrate why the design environment determines the evaluation rather than an all-purpose claim that a binder is more durable.
The National Academies SCM report reviews material-specific effects on concrete properties and durability. Its breadth is useful precisely because the answer varies with composition, processing and application. A general statement that supplementary materials improve every property would be as misleading as claiming that they necessarily weaken concrete. SCM technical report.
Testing should match the relevant mechanism and account for conditioning. A laboratory procedure can accelerate or isolate an exposure, but the relationship to field service needs interpretation. The report should identify the test, specimen age and limitations rather than converting a passing result into an unlimited service-life guarantee.
Cracking also complicates simple comparisons. A mixture with favourable transport properties in an intact specimen can perform differently if construction or restraint produces cracks. Structural design, detailing and curing therefore remain part of the durability system. Material optimization cannot be separated completely from how the structure is built and used.
A lifecycle carbon comparison may assume a service life or maintenance schedule. Those assumptions should be consistent with the durability evidence. A low initial impact can be offset if the structure requires substantially earlier replacement, but that possibility should be evaluated rather than asserted without data. Conversely, a justified longer service life can be a real benefit without being invented to improve the carbon calculation.
The practical standard is to specify the exposure, select appropriate tests and preserve field evidence. Durability claims should be narrow enough to be supported and broad enough to address the actual project. That gives lower-carbon mixtures a fair opportunity while protecting the service the structure must provide.
7. Supplementary materials have supply chains too
Replacing part of the conventional binder changes the demand for other materials. Their source, processing, transport and consistency matter to both performance and emissions. A substitution percentage alone does not establish the final environmental benefit or the reliability of supply.
A hypothetical mix uses a supplementary material that is locally available during the qualification trial. If later production draws from a different source with different properties, the original test evidence may not fully describe the delivered mixture. The change needs appropriate review. Material identity is more specific than the generic category printed in the sustainability narrative.
The same issue affects lifecycle accounting. A by-product, processed material or manufactured alternative can carry different burdens depending on the applicable allocation method and production data. The EPD should disclose the method rather than treating every supplementary material as having zero impact by definition.
DOE's 2024 environmental-review document for a proposed calcined-clay research project describes characterization, processing and small-batch evaluation of strength, durability, environmental impact and economics. It is evidence of planned research scope, not proof that a particular commercial product has already met all those objectives. DOE calcined-clay research document.
That distinction is useful when reading innovation news. A promising laboratory material may still need scale-up, source qualification and consistent production. A pilot can establish one stage without proving national supply availability. The article should preserve the achieved scale and the remaining steps instead of using the word breakthrough to skip them.
Regional availability can change the best option. A material with a favourable production profile may require long transport or additional processing to reach a project. Another locally available option may perform differently. A fair comparison uses the actual supply arrangement and does not assume that a global material category has one universal emissions factor.
Procurement should therefore connect the approved mix to its constituent sources and change-control process. If supply changes, the performance and environmental records may need updating. This creates a more reliable low-carbon pathway than specifying an ambitious substitution target while leaving the identity and availability of the substitute unresolved.
8. Read the EPD's declared unit and boundary
An EPD presents environmental information under a product-category framework. Its result must be read with the declared or functional unit, lifecycle stages, data period and product scope. A number expressed per cubic metre is not self-explanatory without knowing which concrete it describes and which processes are included.
A hypothetical EPD reports 250 kilograms of carbon-dioxide equivalent per cubic metre for production stages, while another reports 270 including additional transport. Ranking them directly would compare different boundaries. The analyst should align the scope or report the difference rather than declaring the lower number superior.
Product scope is equally important. An industry-average declaration can provide context but may not represent a particular plant and mixture. A multi-plant declaration can cover several facilities under stated rules. A plant-specific declaration may offer more relevant information for a purchase, but its details still need to match the supplied product.
The NRMCA certified-EPD directory distinguishes declarations covering particular plants and sets of mixtures. Its entries demonstrate why the certification identifier and scope matter. Membership in a directory does not mean every product sold by a company shares the same declaration or emissions value. NRMCA certified EPD directory.
Verification also has a defined role. A third-party review can assess conformity of the declaration and calculation with the relevant rules. It is not a structural approval, a guarantee of future plant performance or proof that the construction site received the declared mix. Those claims require additional records.
A useful comparison table therefore includes more than the global-warming-potential value. It identifies the mix, plant scope, strength and age, unit, stages, rule version, validity period and verification status. The extra columns prevent a tidy ranking from hiding differences that determine whether the products are actually comparable.
9. Product category rules and versions affect comparability
Product category rules specify how declarations within a category are prepared. They can define boundaries, allocation, data requirements and other choices that influence results. Two EPDs with similar formatting may still use different rule versions or methodological assumptions. The reader should inspect those details before treating the values as interchangeable.
Caltrans' EPD resource page lists product-category-rule information and validity periods for relevant materials. That emphasis on versions and dates is a useful reminder that an environmental declaration belongs to a methodological framework that can change. Caltrans EPD resources.
A hypothetical supplier updates an EPD after its background database changes. The new value may differ even if the physical mix remains unchanged. Another supplier may reduce actual cement use while retaining the same method. Both produce a changed number, but only the second necessarily demonstrates a physical production improvement. A trend should separate those causes.
The data period also matters. An EPD can remain valid while describing an earlier operating period under its rules. That does not make it useless, but a claim about a recent plant upgrade should be supported by evidence that the declaration reflects the upgrade or by a clearly identified supplementary calculation. The validity date and measurement period answer different questions.
Allocation choices can be particularly consequential for co-products and secondary materials. A change in how burdens are assigned can alter the reported concrete impact without a corresponding change in total system emissions. The analysis should follow the chosen standard consistently and avoid treating one allocation convention as a direct measurement of causal benefit.
Comparability does not require pretending that all methodological uncertainty can be eliminated. It requires identifying the important differences and deciding whether they affect the decision. If two products are close in reported impact but use materially different assumptions, a confident ranking may not be justified. A wider comparison or additional information may be needed.
Version control protects both purchasers and suppliers. It allows a project to preserve the evidence used at award, identify changes during delivery and explain later revisions. Without that record, an environmental claim can become detached from the document and method that originally supported it.
10. Compare quantities at the project level
A product with lower impact per cubic metre does not automatically produce a lower-impact structure if the design requires more of it. Conversely, a material with a somewhat higher unit impact can sometimes reduce total quantity through a different structural design. The comparison must preserve required performance and include the actual quantities involved.
Consider a hypothetical option A requiring 1,000 cubic metres at 300 kilograms of carbon-dioxide equivalent per cubic metre. Its assessed product-stage total is 300 tonnes. Option B requires 1,100 cubic metres at 250 kilograms per cubic metre, giving 275 tonnes. B's unit impact is about 16.7 percent lower, but its project-stage total is only about 8.3 percent lower because the quantity differs.
The example is arithmetic, not a claim that low-carbon mixtures generally require more material. Its purpose is to show why unit comparisons and project comparisons answer different questions. If both options use the same quantity and meet the same function, the unit reduction carries through directly. If quantities change, the calculation must change too.
The design comparison should include other affected materials. A different structural solution may alter reinforcement, formwork or foundations. Focusing only on concrete can miss a shift elsewhere. The appropriate scope depends on the decision, but the report should say which related changes are included and which remain outside the calculation.
Construction waste and rejected loads can also affect actual quantities. An estimate based on design volume may differ from the amount produced and delivered. The final account should reconcile planned, ordered, delivered and installed quantities where the claim concerns completed work. A design-stage estimate should remain labelled as such until those records exist.
The baseline should be a credible alternative. Comparing the selected mix with an unusually high-impact hypothetical recipe can exaggerate savings. A useful baseline might be the original approved design or a documented locally feasible alternative, with its performance requirements and method stated. The choice should be made transparently rather than selected after the preferred result is known.
Project-level accounting therefore connects environmental documentation with design and construction records. It moves the discussion from a product's advertised intensity to the actual material service purchased. That is the level at which public claims about a completed building, bridge or pavement should ultimately be tested.
11. Transport and installation belong to wider claims
Many concrete EPD comparisons focus on production stages. A wider project assessment may also need transport to the site and installation activities. Those stages should be added consistently rather than assumed negligible or included for only one option. The choice of boundary should follow the claim.
New York's guidance explicitly discusses its treatment of transport and installation stages in embodied-carbon reporting. The important lesson is to inspect the programme's scope rather than assuming that every published number covers the complete construction process. New York embodied-carbon FAQ.
A hypothetical lower-impact mix supplied from farther away may still be preferable overall, but the comparison requires an actual transport calculation. Distance alone is not the full answer; vehicle loading, return trips and the relevant logistics matter. Equally, a nearby plant does not automatically have the lowest production impact. The stages need to be evaluated together.
Installation requirements can differ. Pumping, curing, heating or other activities may affect the project account. If a mixture requires a changed process to meet the schedule, the associated impacts and costs should be considered. A claim about the whole installed element cannot stop at the plant gate merely because that is where the EPD ends.
The assessment should avoid double counting. Transport of constituents to the concrete plant may already be included in the product stages, while transport from the plant to the site may be separate. Adding both again as generic transport can inflate the total. A clear stage map prevents the same activity from appearing twice.
Time and geography can affect emissions factors. A construction-energy assumption from another region or year may be suitable for an early estimate but should be identified as such. Where project records become available, the analyst can replace estimates with actual quantities under a consistent method.
The practical outcome is a layered report: product-stage intensity, delivered quantity and any additional project stages included. Readers can then see what has been established without forcing every EPD to answer questions outside its scope. A narrow but accurate claim is more useful than a broad claim assembled from mismatched pieces.
12. The delivered mix needs a traceable identity
An approved mix design and an EPD are documents. The concrete arriving at the site is a material batch. The environmental and engineering claim becomes credible when records connect the delivered material to the approved product and identify any changes. Without that connection, a project may possess excellent documentation for a mixture it did not actually use.
A hypothetical supplier changes a constituent source during construction because the original supply is unavailable. The change may be acceptable after review, but the project needs to know whether it affects performance qualification or the EPD's applicability. Treating all products under the same informal mix nickname as identical can conceal a meaningful revision.
Delivery tickets, batch records and acceptance documentation serve different purposes in this chain. They can identify quantities, mix codes, timing and other relevant information. The project should define which records establish the environmental claim and how they are reconciled with the engineering submittal. The exact requirements depend on the governing contract.
FHWA's structural-concrete mix-design submittal form, marked Rev3-2026, includes fields for supplementary materials and curing-related information. It illustrates the specificity of a technical submittal, but a form alone does not establish that every field was completed accurately or that the site followed the approved design. FHWA mix-design submittal form.
A useful change-control process distinguishes minor administrative corrections from material changes requiring review. It identifies who can approve a substitution and when a revised environmental calculation is needed. The process should be clear before a shortage or schedule pressure creates an urgent decision.
Final reporting should reconcile exceptions. If some loads used an alternative approved mix, the completed-project carbon account should reflect those quantities rather than applying the preferred mix's value to the entire job. If records are missing, the uncertainty should be stated and treated under a defined conservative method rather than silently ignored.
Traceability is not an argument for publishing proprietary recipes in full. A public summary can identify the approved mix family, applicable declaration and verified quantities while preserving legitimate confidential details. What matters is that an independent reviewer can connect the claim to the material actually supplied.
13. Schedule flexibility can be a design resource
Construction schedules often shape concrete specifications before environmental alternatives are considered. A requirement for rapid strength development may be essential in some work and merely convenient in other work. Examining that distinction early can create room for lower-impact options without compromising safety or project function.
A hypothetical project might be able to pour an element earlier or adjust the sequence so that later strength acceptance does not delay completion. Another project may have a narrow closure window and need rapid reopening. The same mix strategy will not suit both. The environmental assessment should respect the real constraint while asking whether the constraint has been tested.
The cost comparison should include the whole construction sequence. A lower-cost material that extends critical-path work may create additional expense. A mixture that costs more per cubic metre may reduce other costs or meet a difficult exposure requirement. Neither the material price nor the carbon number should be evaluated independently of the service and schedule.
Trial batches and mock-ups can reduce uncertainty before critical work. They can help establish workability, finishing and strength-development behaviour under relevant conditions. Their scale and representativeness should be stated. A successful small trial supports the next step; it does not eliminate the need for production quality control.
Early supplier engagement can also reveal what is available locally and what lead time qualification requires. Adding a low-carbon requirement after procurement is nearly complete can create unnecessary conflict. A well-designed process gives the supplier a defined performance target and enough time to demonstrate an alternative.
Schedule flexibility is therefore a resource to be evaluated, not an excuse to relax engineering requirements. Where it exists, it can improve the feasible set of mixtures. Where it does not, the project should document the limitation and pursue other reductions rather than claiming that one recipe must work everywhere.
14. Check the volume and the comparison margin
Concrete quantities can be expressed in different units across design, ordering and environmental documents. A cubic-yard purchase record cannot be multiplied directly by an EPD expressed per cubic metre. The conversion should be explicit and applied once. A spreadsheet with correct emissions values can still produce a wrong project total through a unit mismatch or a duplicated conversion.
The quantity record should also distinguish delivered volume from accepted and installed volume. A rejected load has a production history even if it does not become part of the structure. The appropriate treatment depends on the project's assessment boundary, but it should not disappear merely because the design model contains only the final structural volume. Returned material and waste need consistent treatment rather than selective exclusion.
Small differences between products deserve proportionate interpretation. If two comparable declarations differ by only a few percent, uncertainty in data, quantity and production variability may matter to the ranking. The purchaser can still use a defined procurement threshold, but the public should not describe a narrow numerical difference as a scientifically exact superiority unless the evidence supports that precision.
A sensitivity check can identify whether the decision is robust. Recalculate the project total under plausible quantity changes and documented methodological alternatives. If the preferred option remains substantially lower while meeting performance requirements, confidence improves. If the ranking reverses easily, the decision may need more information or should be described as a close comparison. This is especially useful when a project combines several mixtures with different quantities rather than buying one uniform product.
15. A credible claim survives the final account
The completed-project claim should connect four things: the required function, the qualified mixture, the applicable environmental declaration and the actual quantity delivered and installed. Curing and acceptance records support the performance side. Method versions and boundaries support the environmental side. The final account should preserve both.
A useful report distinguishes design-stage savings from as-built savings. It states the baseline, the selected option, the comparable lifecycle stages and any changes during construction. It identifies material exclusions and uncertainty rather than presenting a rounded percentage with no denominator. Readers should be able to understand whether the figure concerns concrete alone or a wider structure.
The report should also avoid treating an EPD as an award for being environmentally superior. A declaration can describe a relatively high-impact product accurately. Its value lies in transparent information under a method, not automatic endorsement. Procurement criteria and engineering requirements determine whether the product is appropriate for the project.
Longer-term performance remains relevant. If a project claims a service-life benefit, maintenance and condition information should eventually test that claim. A projected lifespan is not an observed outcome at commissioning. The report can acknowledge the projection while preserving the evidence needed to revisit it later.
Lower-carbon concrete is best understood as a coordinated design and construction task. Binder choices matter, but so do structural quantities, local materials, workability, curing and durability. A successful project aligns these factors rather than asking a single substitution percentage to carry the entire environmental promise.
The public should therefore ask a straightforward question: lower than what, for which required service, and based on which material actually used? A credible answer can point to the baseline, the performance evidence and the delivery record. A weak answer stops at a product label or a laboratory result unrelated to the completed work.
The carbon claim begins before the pour because the design and specification determine what options are possible. It becomes trustworthy after the pour only when the project can show that the chosen option was supplied, protected and accepted as intended. That connection is what turns a lower number on paper into a defensible improvement in construction.