To decide whether to retain or demolish a building, compare the options element by element, not as two whole buildings. Give every element a pathway (retain, reuse, refinish, replace or new), weight it by its share of a new building’s embodied carbon, and add up. On the illustrative office building below, a deep retrofit lands at about 47 and a light retrofit at about 34 on an index where a like-for-like new building is 100, against 110 for demolish-and-rebuild. That is a 57–69 % embodied-carbon saving from retaining. Almost all of it comes from keeping the foundations and the frame, which is also what published projects show.
This is the calculation Elementa is built to run at concept stage. Here it is done by hand, so you can see every step and check it against your own building. For the broader carbon and cost case behind the decision, see refurbishment vs demolition.
The short version
- Method: element list → share of new-build embodied carbon per element → pathway per element → pathway factor → sum. Compare scenarios on the same list.
- Elementa’s indicative pathway factors: Retain 0.05, Reuse 0.15, Refinish 0.40, New 1.00, Replace 1.10. Replace sits above New because it is new manufacture plus removing what was there. The factors are Elementa’s own stated assumptions, not published values.
- Illustrative result: demolish and rebuild 110, deep retrofit ≈ 47, light retrofit ≈ 34. That is a 57–69 % saving, in line with published adaptive-reuse projects at 68–82 % (Architectural Record, 2022).
- The saving is driven by the structure’s share. If foundations and frame are 55 % of new-build carbon, retaining them is most of the answer.
- Demolition wins when the frame fails, is contaminated, or can’t reach the required performance. The element list shows you which case you’re in.
- Scope: embodied carbon (A1–A3) only. Operational energy is a separate calculation and can change the answer for the envelope.
- All figures here are indicative or illustrative, for early comparison, not for certification or compliance.
What’s the method?
Five steps, and the first one does most of the work.
- List the elements. Use systems, not rooms: foundations, frame and slabs, envelope, windows, internal walls, finishes, services. Seven lines is enough at concept stage.
- Give each element its share of a new building’s embodied carbon. Take the shares from a comparable new-build assessment, or from your own concept-stage estimate (see estimating embodied carbon at concept stage). They add up to 100.
- Choose a pathway per element per scenario. Retain it in place, reuse it elsewhere, refinish its surface, replace it with new, or build new where nothing existed.
- Multiply by the pathway factor. Retain 0.05, Reuse 0.15, Refinish 0.40, New 1.00, Replace 1.10 (see circularity pathways).
- Sum per scenario and compare. The absolute number doesn’t matter at this stage. The gap between scenarios does, and so does the element that drives the gap.
Citation capsule: A retain-or-demolish comparison at concept stage lists the building’s elements, weights each by its share of a new building’s embodied carbon, assigns a pathway per element (retain 0.05, reuse 0.15, refinish 0.40, new 1.00, replace 1.10 in Elementa’s indicative factors), and sums per scenario. The comparison is relative, and its value is showing which element drives the difference.
What does it look like on a real building type?
Take a 1970s concrete-framed office building. The frame is sound, the facade is at end of life, and services are obsolete. The shares below are illustrative, chosen to be plausible for a concrete-framed building, so the arithmetic can be followed. Put your own in their place.
| Element | Share of new-build embodied carbon | A: Demolish and rebuild | B: Deep retrofit | C: Light retrofit |
|---|---|---|---|---|
| Foundations | 15 | Replace 1.10 → 16.5 | Retain 0.05 → 0.75 | Retain 0.05 → 0.75 |
| Frame and slabs | 40 | Replace 1.10 → 44.0 | Retain 0.05 → 2.0 | Retain 0.05 → 2.0 |
| Facade (opaque) | 15 | Replace 1.10 → 16.5 | Replace 1.10 → 16.5 | Refinish 0.40 → 6.0 |
| Windows | 5 | Replace 1.10 → 5.5 | Replace 1.10 → 5.5 | Replace 1.10 → 5.5 |
| Internal walls | 8 | Replace 1.10 → 8.8 | Refinish 0.40 → 3.2 | Reuse 0.15 → 1.2 |
| Finishes | 7 | Replace 1.10 → 7.7 | Replace 1.10 → 7.7 | Replace 1.10 → 7.7 |
| Services | 10 | Replace 1.10 → 11.0 | Replace 1.10 → 11.0 | Replace 1.10 → 11.0 |
| Total (index) | 100 | 110 | ≈ 47 | ≈ 34 |
| vs demolish and rebuild | — | −57 % | −69 % |
Three things come out of the table immediately.
The structure is the decision. Foundations and frame are 55 of the 100 points. Retaining them costs 2.75 against 60.5 for replacing them. That 58-point gap is most of the difference between A and B. Everything else in the building is second-order.
The facade is the biggest retrofit choice. Between B and C, the only large move is refinishing the facade (6.0) instead of replacing it (16.5). That is where the retrofit scenarios differ, and it’s also where operational energy comes back in (see below).
Services and finishes cost the same in every scenario. They get replaced in all three, so they don’t help you choose. Leave them out of a first comparison if you’re short of time. The same per-element logic applies when you weigh a single element on its own; see reusing building elements vs new.
The result sits inside the published range. Architectural Record’s 2022 continuing-education feature reports a 68 % saving at Gensler’s The Link in Denver (2,165 against 6,750 t CO₂e) and 82 % at ZGF’s Cal State LA building, both from retaining the existing structure (Architectural Record, 1 February 2022). German studies summarised by dena’s Gebäudeforum put renovation at 50–66 % lower A1–A3 emissions than new build (gebaeudeforum.de).
Citation capsule: On an illustrative 1970s concrete-framed office building, with structure at 55 % of new-build embodied carbon, Elementa’s indicative pathway factors give demolish-and-rebuild 110, deep retrofit about 47 and light retrofit about 34 on an index of 100 for a new building. That is a 57–69 % saving, driven almost entirely by retaining foundations and frame. This is consistent with published adaptive-reuse projects reporting 68–82 % (Architectural Record, 2022).
Why is demolish-and-rebuild 110, not 100?
Because a rebuild replaces every element, and replacement is new manufacture plus stripping out and disposing of what was there. Elementa scores Replace at 1.10. The direction is certain. The 10 % is Elementa’s own stated assumption, not a published end-of-life figure, and the product labels it that way. Set it to 1.00 if you prefer: demolish-and-rebuild becomes 100, the retrofits become about 43 and 32, and the savings stay at about 57 % and 68 %, because the retrofit scenarios replace elements too. The conclusion doesn’t depend on the 10 %.
When does demolition win?
The element list also tells you when retention is the wrong answer, because you can see which line breaks.
- The frame fails. If the structure can’t carry the new loads, or is degraded beyond economic repair, the 40-point line becomes Replace and the gap largely disappears. Reinforcing a failing frame can cost more carbon than replacing it.
- Contamination. Where hazardous materials run through the structure, safe retention may be impractical. That’s a safety constraint, not a carbon one.
- Performance that can’t be reached. If floor-to-floor heights, the structural grid or the fabric can’t meet the required use, a retrofit that works around them may need so much replacement that scenario B drifts towards A.
Even then, partial retention is worth checking. Keeping only the foundations still saves about 16 index points.
What does the embodied comparison leave out?
Operational energy. Scenario C refinishes the facade instead of replacing it. That saves embodied carbon, but it may leave the building with a worse U-value for decades. For the envelope in particular, the embodied saving has to be set against the operational one; reclaimed windows walks through that trade-off. Elementa’s comparison covers embodied carbon only, and it says so.
Modules beyond A1–A3. The pathway factors apply to production-stage carbon. The EPBD calculation for new buildings covers the full life cycle to EN 15978 (see life-cycle GWP modules). A concept-stage comparison is for choosing a direction, not for disclosure.
Cost. This worked example covers carbon only. Elementa shows indicative cost alongside carbon, because the retain decision is rarely made on carbon alone: for each assembly you choose an intervention, and it is priced from a versioned 2026 Austrian reference of expert estimates by LX Design Studio, adjusted by region, with a stated uncertainty of mostly ±15–25 %. The figures are for early comparison, not tendering, and a quote you enter replaces the reference rate.
Are there other tools for this?
Yes, two public ones are worth knowing.
- CARE Tool (Architecture 2030 and partners, funded by grants and donations) compares the total carbon of renovating against replacing, embodied and operational, at pre- or early-design stage. Its own page lists easier international use as a future addition (caretool.org).
- Gebäude-Klimarechner (g-kr.online) is free, compares renovation strategies against replacement construction in CO₂e plus a resource footprint, and launched in 2026 as a beta for residential buildings (bau.bi, 4 February 2026).
Both give a whole-building answer. What the element-by-element method adds is the why: which line carries the saving, and which one flips it. For a wider view of tools, see choosing embodied carbon software.
Frequently asked questions
How is the retain-or-demolish index calculated?
List the building’s elements, give each its share of a new building’s embodied carbon so the shares add up to 100, choose a pathway per element per scenario, multiply each share by the pathway factor (Elementa’s indicative Retain 0.05, Reuse 0.15, Refinish 0.40, New 1.00, Replace 1.10) and sum per scenario. A like-for-like new building scores 100. The result is the gap between scenarios and the element that drives it, not the absolute number.
What share of a building’s embodied carbon is in the structure?
Typically the largest share. The example uses 55 % for foundations and frame, as an illustration. Use a comparable new-build assessment, or a concept-stage estimate of your own building, for your number. Because the retain-or-demolish result depends on this share more than on anything else, it’s the one figure worth getting right first.
Is there a free retain-or-demolish calculator?
The Gebäude-Klimarechner (g-kr.online) is free and compares renovation strategies with replacement construction; it launched as a beta for residential buildings in 2026. The CARE Tool, from Architecture 2030 and partners, compares renovating with replacing, embodied and operational, at early design. Both give a whole-building answer; the element-by-element method shows which element drives it.
Why does Elementa score Replace higher than New?
Because replacing an element means manufacturing a new one and also stripping out and disposing of the old one. Elementa uses 1.10. The direction is certain, and the 10 % is Elementa’s labelled assumption, since Elementa’s material data covers the product stage (A1–A3) only and models no end-of-life modules.
Can this comparison be used for EPBD disclosure or certification?
No. It’s an indicative, production-stage comparison for choosing a direction at concept stage. Disclosure and certification need a full EN 15978 life-cycle assessment by a qualified assessor.
The takeaway
Retain-or-demolish is a structure question dressed up as a building question. List the elements, weight them, give each a pathway, and the answer is usually visible in two lines of the table: foundations and frame. If those can stay, retention saves most of the embodied carbon. If they can’t, the same table shows you why, and how much partial retention still saves.
If you want this comparison for your own building, with the element list, pathways and cost ranges laid out before formal assessment, start a project.