Gaining one grade on the DPE (Diagnostic de Performance Énergétique, the French Energy Performance Certificate) — moving from F to E, E to D, D to C — is not just about image: it is what takes a dwelling out of the rental-ban timetable, unlocks a purchase loan and restores part of the lost green value. But not all works are equal. Some cost a few thousand euros and jump a whole grade; others cost ten times more for the same jump, because they target a work item that was not the dwelling's weak link.
The useful question is therefore not "which works should I do?" but "which works, in which order, for the best return on investment?". This article reviews the five work items with the best grade-gain-to-cost ratio — loft insulation, wall insulation, window replacement, heating system replacement and ventilation — with, for each, a cost range from ADEME data, the 2026 subsidies, and an indicative impact on the rating and the bill. The figures are sourced orders of magnitude, never promises: the real gain always depends on the building.
Nearly 5 million dwellings are rated F or G in France — roughly 15% of the residential stock according to the ministry's statistical service (SDES, stock as of 1 January 2024). For most of them, one or two well-chosen works are enough to gain at least one grade — provided you target the right item first.
What this article covers
Why gaining one DPE grade has concrete economic value (rental ban, loan, green value), how the 3CL-DPE methodology ranks work items, the five works with the best ROI with ADEME cost ranges and 2026 subsidies, an illustrative example on an F-rated house, the pitfalls that lead to spending without gaining a grade, and the effect of the 1.9 electricity coefficient that can gain a grade with no works.
Why gaining one DPE grade has economic value
The DPE rates dwellings from A (the most efficient) to G (the most energy-intensive) using two primary-energy indicators: consumption (in kWh/m²/year) and greenhouse gas emissions. Its validity period is ten years (French Construction and Housing Code — CCH — article D.126-19). Since the reform of 1 July 2021, it is binding: the displayed grade triggers direct regulatory consequences.
The first is the rental-ban timetable for thermal sieves, set out in Act no. 2021-1104 of 22 August 2021 (the Climate & Resilience Act) in articles L.173-1-1 and L.173-2 of the CCH. G-rated dwellings have been banned from rental since 1 January 2025, F-rated ones will be from 1 January 2028, and E-rated ones from 1 January 2034. Moving a property from F to E pushes its ban deadline back from 2028 to 2034.
The second consequence affects financing: banks now factor the DPE grade into their risk analysis, because an energy-intensive dwelling exposes the borrower to high charges and to a resale discount. A thermal-sieve property can lengthen the time to obtain a loan, or even jeopardise it.
The third is green value, the price gap observed between an efficient dwelling and a comparable thermal sieve. Studies by Notaires de France (the French Notaries Federation) have for several years measured a sale discount on F- and G-rated dwellings compared with D-rated ones, varying by region and property type. Gaining one grade therefore lifts a regulatory constraint, eases a loan and restores part of the market value at once.
| Deadline | Grade banned from rental | Legal reference |
|---|---|---|
| 1 January 2025 | G | CCH art. L.173-1-1 (Act no. 2021-1104) |
| 1 January 2028 | F | CCH art. L.173-1-1 |
| 1 January 2034 | E | CCH art. L.173-1-1 |
How the 3CL methodology ranks the best-ROI works
The DPE grade is calculated by the 3CL-DPE methodology, defined by the ministerial order of 31 March 2021 (amended by the ministerial order of 25 March 2024). It reconstructs the dwelling's heat losses item by item — roof, walls, windows, floors, thermal bridges, air renewal — then weights them by the heating system and its efficiency. This logic explains why the same work can gain a grade in one dwelling and almost nothing in another.
The weak-link principle
The grade gain depends on the dominant heat-loss item. In an old house with uninsulated lofts, the roof can account for a major share of losses, so insulating it produces maximum leverage. In a flat already well insulated at roof level but fitted with single glazing, it is the opposite. The best ROI is never universal: it tackles the weak link of the specific dwelling.
Profile: a detached house from the 1970s, rated F, uninsulated walls, accessible empty lofts and an original oil boiler combines several weak links — this is the case where the order of works makes all the difference to the cost of gaining a grade.
Why the order of works matters
Insulating before changing the heating system allows the new equipment to be correctly sized: a heat pump (PAC — pompe à chaleur) installed in a still poorly insulated dwelling will be oversized, more expensive and less efficient. This is the logic advocated by ADEME and France Rénov' (the public home renovation advice service): the envelope first, the system second. This hierarchy follows from the calculation itself: heating consumption starts from the heat losses of the envelope before the efficiency of the system is applied, as our guide to the 3CL EPC method explains.
The effect of the electricity coefficient on electrically heated dwellings
The 3CL methodology converts final energy into primary energy via a coefficient. For electricity, it has dropped from 2.3 to 1.9 (ministerial order of 13 August 2025, published in the Official Journal on 26 August 2025, in force on 1 January 2026). Mechanically, the primary-energy consumption of an electrically heated dwelling falls — which has moved many properties up one grade, with no works at all. Before any spending, always start from the rating recalculated with the 2026 coefficients.
The 5 works with the best return on investment to gain one grade
Here are the five items ranked by grade-gain-to-cost ratio, based on cost ranges published by ADEME and 2026 subsidies published by ANAH (the French National Housing Agency). The ranges are orders of magnitude excluding subsidies; the grade gain is indicative and depends on the building.
1. Loft insulation: the best ROI almost everywhere
In an old house, the roof is one of the first heat-loss items: ADEME estimates it can account for roughly 25 to 30% of the heat losses of a poorly insulated house. It is also the cheapest item per m². Insulating empty lofts costs, as an indication, around €25 to €75 per m² depending on the technique (blowing or rolling out) and the insulation material, per ADEME ranges; sloped roof sections cost more.
Why it is the best ROI: one of the lowest costs per m², one of the highest heat-loss shares. On an old house, loft insulation is in many cases enough to gain a grade on its own — depending on the case, and subject to the dwelling's initial condition.
2026 subsidies: loft insulation remains fundable through MaPrimeRénov' on the single-action route, combinable with Energy Saving Certificates (CEE — Certificats d'Économies d'Énergie, bonuses funded by energy suppliers) and the reduced 5.5% VAT (ANAH, 2026 conditions).
2. Wall insulation: the big lever, now reserved for full-scale renovation
Walls account for a heat-loss share of the same order as the roof in an uninsulated house. Two techniques exist: internal insulation (ITI) and external insulation (ITE). ITE, more efficient because it better eliminates thermal bridges, is also the most expensive: ADEME puts its cost, as an indication, at around €110 to €200 per m² of façade, against a lower range for ITI. At the scale of a house, walls can represent roughly 20 to 30% of heat losses according to APUR data on the Paris-region building stock.
⚠️ Warning: since 1 January 2026, wall insulation (both ITI and ITE) has been excluded from the MaPrimeRénov' single-action route (ministerial order of 8 September 2025, amending the order of 17 November 2020). It is now only fundable as part of a full-scale renovation, with an energy audit, a Mon Accompagnateur Rénov' supervisor and a gain of at least two grades. A quote offering ITE "funded by MaPrimeRénov'" as a single action is based on outdated rules.
The choice between ITI and ITE is decisive for the cost, the loss of living space and the result. On the funding side, our guide to MaPrimeRénov' 2026: new rules and grant levels details the conditions of the full-scale renovation route.
3. Window replacement: immediate comfort, variable grade gain
Replacing single-glazed windows with double or triple glazing clearly improves comfort and removes the cold-wall effect. But in heat-loss share, windows generally weigh less than the roof or the walls — often around 10 to 15% in an old dwelling. The cost, as an indication, stands per ADEME at several hundred euros per fitted window, a budget that rises quickly across a whole dwelling.
The "windows only" trap: changing only the windows on a dwelling with uninsulated lofts and walls rarely produces a grade gain, because the dominant items remain unchanged. Windows become effective when they complete an envelope insulation, not when they replace it.
2026 subsidies: window replacement remains eligible for the single-action route (MaPrimeRénov' + CEE), subject to performance conditions.
4. Heating system replacement (heat pump): the carbon lever
Replacing an oil boiler or a worn-out electric convector with an air-to-water heat pump (PAC) acts on two DPE indicators at once: primary-energy consumption and greenhouse gas emissions. It is the most effective item to bring down the GHG rating, which can cap the grade of an oil-heated dwelling. Its cost stands, as an indication per ADEME, from several thousand to more than ten thousand euros depending on the power and the installation.
2026 subsidies: the air-to-water heat pump is the most heavily subsidised single action. According to the schedules published by ANAH, MaPrimeRénov' on the single-action route can reach, as an indication, up to €5,000 for very modest income households, €4,000 for modest ones and €3,000 for intermediate ones, combinable with CEE, the eco-PTZ (the zero-rate renovation loan up to €50,000) and 5.5% VAT.
Do it in the right order: install the heat pump after the insulation, otherwise the equipment is oversized. This is a spend where the order of works directly changes the cost and the efficiency.
5. Ventilation: the discreet item that secures the gain
Mechanical ventilation (VMC — ventilation mécanique contrôlée), especially in humidity-controlled or dual-flow versions, is not the item that produces the biggest grade jump, but it is indispensable after heavy insulation: a dwelling made airtight without suitable ventilation accumulates moisture and degrades air quality. Its cost is moderate, and it improves the air renewal accounted for by the 3CL methodology — which is why it often appears in full-scale renovation bundles.
| Work item | Indicative cost range (ADEME, excl. subsidies) | Heat-loss share (order of magnitude) | 2026 single-action subsidy |
|---|---|---|---|
| Loft insulation | ~€25 to €75 / m² | ~25-30% | Yes (MPR + CEE) |
| Wall insulation (ITE) | ~€110 to €200 / m² of façade | ~20-30% | No — full-scale only |
| Windows (double/triple glazing) | Several hundred € / window | ~10-15% | Yes (MPR + CEE) |
| Air-to-water heat pump | Several thousand to > €10,000 | Acts mainly on primary-energy consumption and GHG | Yes (most subsidised) |
| Ventilation (VMC) | Moderate cost | Air renewal | Yes (often in a bundle) |
Illustrative case study: an F-rated 1970s house
Take a 100 m² detached house rated F, with uninsulated walls, accessible empty lofts and an original oil boiler. The goal: gain at least one grade, ideally two. The amounts below are illustrative, built from ADEME ranges to show the trade-off logic, not to predict a result. Only an audit or a simulation on your real DPE gives a reliable figure.
Scenario — "envelope first" bundle targeting a two-grade gain (from F to D)
| Work item | Logic | Indicative cost (excl. subsidies) |
|---|---|---|
| Empty loft insulation | Weak link no. 1, best ROI — often a grade on its own | ~€2,000 to €5,000 |
| Wall insulation (ITE) | Weak link no. 2, big lever — fundable in full-scale renovation | ~€15,000 to €30,000 |
| Air-to-water heat pump (replaces oil) | Lifts the oil GHG cap, sized after insulation | ~€10,000 to €16,000 |
| Humidity-controlled VMC | Secures air quality after insulation | ~€1,000 to €3,000 |
| Indicative bundle total | — | ~€28,000 to €54,000 |
| Out-of-pocket cost after subsidies | Per full-scale MPR + CEE + eco-PTZ + 5.5% VAT | Significantly reduced, to be costed |
The lesson of this table is not the total but the hierarchy. If the budget only allows one item, loft insulation offers the best return: low cost, big effect, often a grade gained. If the goal is a two-grade gain to leave the ban timetable for good, the envelope + heat pump bundle is the way, and it then shifts into the full-scale renovation route where wall insulation becomes fundable again. Stopping at grade E does not put the dwelling in the clear either: our analysis of the 2034 deadline for E-rated dwellings sets out that timetable and the situations that force action well before it.
The mistakes that lead to spending without gaining a grade
Mistake no. 1 — Changing the heating before insulating
Installing a heat pump in a still poorly insulated dwelling leads to oversizing it: the equipment costs more, runs less efficiently, and the grade gain is lower than it would have been after insulation. The envelope first, the system second: this is the basic rule advocated by ADEME and France Rénov'.
Mistake no. 2 — Betting everything on windows
Window replacement is the most visible and tempting work, but on a dwelling with uninsulated lofts and walls, windows weigh too little in the balance to jump a grade. Comfort improves; the rating, rarely. Keep this item for the end of the bundle, once the envelope is treated.
Mistake no. 3 — Starting works before checking the 2026 rating
Since 1 January 2026, the electricity coefficient of the 3CL methodology has dropped from 2.3 to 1.9. An electrically heated dwelling may have gained a grade with no works at all. Starting a project on the basis of an outdated rating can mean paying for a gain the reform has already delivered. Always start from the recalculated 2026 rating.
⚠️ Warning: no professional can guarantee a precise grade gain before studying the building. Be wary of quotes promising "+2 grades guaranteed" before any audit: the gain depends on the dwelling, the climate and the initial condition. The only certainty comes from a simulation on your real DPE or an energy audit.
Mistake no. 4 — Forgetting ventilation after heavy insulation
Making a dwelling airtight without adapting the ventilation causes condensation, moisture and degraded air quality — sometimes disorders in the building fabric. The VMC is not a luxury: it is the condition for the insulation to deliver its promises without side effects.
Simulate your grade gain before signing a quote
OneDpe works and DPE grade-gain simulator
Build your works bundle item by item — loft insulation, wall insulation, windows, heat pump, ventilation — and see the estimated new rating, the expected grade gain and the impact on the bill, with the 2026 subsidies built in (MaPrimeRénov', CEE, eco-PTZ). You immediately see which item gains a grade and which one does not move the rating.
Going further: the 2026 DPE certificate to update your rating for free and the EPC classes A to G guide.
Conclusion
Gaining one DPE grade is first a trade-off problem, not a budget one. The best return comes from loft insulation in the vast majority of old houses; wall insulation offers the biggest lever but now shifts into full-scale renovation; the heat pump unlocks the GHG cap of oil-heated dwellings; windows and ventilation complete the envelope without replacing it. The order — envelope first, system second — decides the gain as much as the choice of items. And before any project, the rating recalculated with the 2026 coefficients is the compulsory starting point.
Before the first quote, the OneDpe works and DPE grade-gain simulator lets you test each item, see which one gains a grade and weigh the indicative cost against the real out-of-pocket cost after subsidies.
