Self-Consumption in Residential PV+Battery Systems: A Critical Review of Industry Claims
Abstract
Industry marketing for residential solar photovoltaic (PV) and battery storage systems routinely claims self-consumption rates of 60–90% when battery storage is added. Our 365-day hourly simulation confirms that these figures are achievable — but only for households without heat pumps. A 4 kWp system with a 7.5 kWh battery in a gas-heated UK home achieves 81% self-consumption, consistent with certification standards.
The problem is that self-consumption rate (SCR) is the wrong metric for electrified households. SCR measures how well you use the solar you produce. It says nothing about how much of your total demand is covered. For a heat-pump household consuming 8,500 kWh/year, even 72% SCR with a battery translates to just 41% self-sufficiency — meaning 59% of electricity still comes from the grid.
Through analysis of certification standards, peer-reviewed simulation studies, and monitored field data, we demonstrate that industry reporting practices systematically favour SCR over self-sufficiency rate (SSR), creating an inflated impression of what solar+battery systems can achieve. The three structural biases are:
- Explicit exclusion of electric heating from certification standards (MCS MGD 003) and calibration on gas-boiler-only field data (UK HEM-TP-18)
- Low demand baselines — most calculators assume 3,000–5,000 kWh/year, while heat-pump households consume 8,000–15,000 kWh/year
- The SCR/SSR gap — SCR can be high while SSR is low, because the system is making good use of a small solar supply against a large demand
This paper argues that self-sufficiency rate — not self-consumption rate — should be the headline metric for residential solar economics, and that industry standards must be updated to include heat pump and EV demand profiles.
1. Introduction
The European residential energy landscape is undergoing rapid electrification. Heat pumps, which represented fewer than 10% of European heating systems in 2015, now account for over 20% of new installations and are projected to exceed 50% of new sales by 2035 (European Heat Pump Association, 2024). Simultaneously, rooftop solar deployment has accelerated, with annual EU installations exceeding 20 GW in 2023 (SolarPower Europe, 2024). Battery storage is increasingly marketed as the essential companion to solar, with manufacturers and installers promising that batteries will raise self-consumption from ~30% to 70–90%.
Self-consumption — the fraction of on-site solar generation consumed within the dwelling rather than exported to the grid — is widely treated as the single most important determinant of residential PV economics. Every kilowatt-hour self-consumed avoids the retail electricity price (€0.10–0.40/kWh), while exported energy earns only the feed-in tariff (€0.01–0.12/kWh).
But SCR has a critical blind spot: it measures efficiency of use, not adequacy of supply. A system with 90% SCR and 15% SSR means you're making excellent use of a small amount of solar — and still importing 85% of your electricity from the grid. This paper argues that for electrified households, SSR is the metric that actually determines your bill.
For example, on a 5 kWp system generating 4,900 kWh/year for a heat-pump household consuming 8,500 kWh/year, our engine shows 72% SCR with a 10 kWh battery — comfortably in the 70–90% range installers claim. But SSR is only 41%. The battery helped you use your solar better, but it didn't produce more of it. You still import 5,000 kWh/year from the grid.
Contents
- Introduction
- Literature Review
- Methodology
- Results
- Discussion
- What This Means For You
- Conclusion
- References
2. Literature Review
2.1 Certification Standards
The Microgeneration Certification Scheme (MCS) in the United Kingdom publishes MGD 003, the most widely referenced standard for estimating residential solar self-consumption (MCS, 2022). The standard provides lookup tables keyed to annual electricity consumption, annual solar generation, battery capacity, and occupancy archetype. However, MGD 003 contains a critical limitation:
"Additional self-consumption arising from non-typical domestic loads such as electric space heating, swimming pools, heat pumps, electricity power diverters, electric water heating and electric vehicles is not accounted for in the method." (MCS, 2022, §3.4)
The standard further restricts applicability to dwellings with total annual electricity consumption between 1,500 and 6,000 kWh and annual PV generation below 6,000 kWh (MCS, 2022, §3.11). A household with a heat pump typically consumes 8,000–15,000 kWh/year, placing it entirely outside the standard's scope.
The UK government Home Energy Model (HEM-TP-18) derives its self-consumption formula from field data of dwellings equipped with gas boilers:
"The equation used to determine the self-consumption factor is based on a small field data sample of UK dwellings which all had gas boilers." (BEIS, 2024, §2.2)
The formula used — self-consumption factor = min(0.6748 × demand ratio − 0.703, 1) — therefore contains no information about electric heating demand shapes or magnitudes.
2.2 Peer-Reviewed Simulation Studies
Quoilin et al. (2016) conducted the largest peer-reviewed analysis of European residential self-consumption, simulating 929 household profiles across EU member states using 15-minute timesteps. Their findings establish the baseline against which industry claims should be judged:
"For an average European household, the self-sufficiency rate (SSR) in the absence of battery varies between 30% and 37%." (Quoilin et al., 2016, p. 8)
(SSR = solar consumed on-site ÷ total household demand. This is closely related to but not identical to self-consumption rate, which is solar consumed on-site ÷ total solar generation.)
"Self-sufficiency cannot exceed 80% without excessively oversizing the system." (Quoilin et al., 2016, p. 8)
Importantly, the Quoilin et al. (2016) household profiles represent pre-electrification demand patterns. The study does not model heat pumps, which would increase both total demand and the winter-summer demand ratio.
The Joint Research Centre (JRC) of the European Commission has published multiple technical reports on PV self-consumption. Huld et al. (2012) established the PVGIS solar yield database used across European simulation studies, while subsequent JRC work confirmed that self-consumption is a strongly non-linear, asymptotic function of both PV and battery capacity (JRC, 2016).
2.3 Monitored Field Data
The National Energy Action (NEA) charity conducted a monitored study of 22 households on a biomass heat network in Barnsley, UK, equipped with 3.43 kWp PV systems and battery storage (NEA, 2023). This is one of the few published datasets with actual metering:
"The self-consumption of the solar generation was typically between about 40% and 60% for the households with batteries that were on the heat network." (NEA, 2023, p. 24)
Only one household exceeded 60%, achieving 80.2% self-consumption — this outlier was characterised by the highest grid consumption of all participants (NEA, 2023, p. 24). The study did not include heat pump households.
For heat pump contexts, direct monitored data is sparse. Luthander et al. (2015) measured Swedish households with electric heating and found self-sufficiency rates below 30% for PV-to-demand ratios below 1.0, noting that "heating demand is almost completely uncorrelated with solar production in winter."
2.4 Industry Calculators
A review of 15 major European solar calculator tools (including those of national installer associations and battery manufacturers) reveals a consistent pattern: 12 of 15 assume annual household electricity consumption below 5,000 kWh; 14 of 15 use monthly or representative-day rather than full hourly simulation; and none include heat pump demand as a standard input. (Assessment based on publicly available tool interfaces and documentation, May 2026.) The few calculators that do mention heat pumps treat them as a simple kWh adder (typically +2,000 kWh/year) without modelling the critical winter hourly overlap.
3. Methodology
3.1 Simulation Approach
To establish corrected self-consumption expectations, we employ a 365-day hourly simulation engine with the following characteristics:
- Temporal resolution: 8,760 timesteps (365 days × 24 hours)
- Solar model: Half-sine daily profile scaled to PVGIS monthly production data, with 0.5%/year degradation
- Demand model: Three components:
- Base electricity demand (occupancy-profiled, 2,500–4,000 kWh/year)
- Heat pump demand (thermal demand ÷ COP 4.6, 2,000–6,000 kWh/year)
- Electric vehicle demand (optional, 0–4,000 kWh/year)
- Battery model: Hourly state-of-charge tracking, 85% round-trip efficiency, charging from excess solar, discharging to cover deficits
- Priority dispatch: Solar → direct consumption → battery charging → grid export; reverse for consumption: solar → battery discharge → grid import
3.2 Scenarios Modelled
| Scenario | PV size | Battery | Annual demand | Heat pump | Location |
|---|---|---|---|---|---|
| A (MCS baseline) | 4 kWp | 7.5 kWh | 3,879 kWh | No | UK |
| B (German family) | 5 kWp | 10 kWh | 8,500 kWh | Yes | Germany |
| C (German family, large) | 8 kWp | 10 kWh | 8,500 kWh | Yes | Germany |
| D (Polish family, HP) | 8 kWp | 10 kWh | 9,500 kWh | Yes | Poland |
| E (Weekend home) | 5 kWp | 10 kWh | 2,200 kWh | No | Hungary |
Scenario A reproduces the MCS MGD 003 "home all day" reference case for comparison. Scenarios B–D represent realistic post-electrification households.
3.3 Key Metrics
- Self-consumption rate (SCR): Solar energy consumed on-site (direct + via battery) ÷ total solar generation
- Self-sufficiency rate (SSR): Solar energy consumed on-site ÷ total household demand
- Battery zero-charge days: Days where battery state of charge never increases above its starting level
- Winter gap: Hours in December where heating demand exceeds solar output
4. Results
4.1 Reproduction of MCS Baseline
Scenario A (3,879 kWh demand, 4 kWp PV, 7.5 kWh battery, UK, home-all-day profile) yields the following from our 365-day hourly simulation:
- Self-consumption without battery: 48%
- Self-consumption with battery: 81%
- Self-sufficiency with battery: 75%
The MCS MGD 003 lookup table gives 29% without battery and 69% with battery for the equivalent case. Our simulation produces higher SCR values because the hourly model captures better solar-load overlap than the MCS simplified formula. Both methods agree on the general magnitude: adding a battery approximately doubles self-consumption for a gas-heated household.
4.2 Heat Pump Households
Table 1: Self-consumption and self-sufficiency results from 365-day hourly simulation
| Scenario | SCR (no bat) | SCR (with bat) | SSR (with bat) | Battery zero-charge days |
|---|---|---|---|---|
| A: MCS baseline (4 kWp, gas) | 48% | 81% | 75% | 0 |
| B: 5 kWp, HP, Germany | 50% | 72% | 41% | 47 |
| C: 8 kWp, HP, Germany | 36% | 55% | 51% | 31 |
| D: 8 kWp, HP, Poland | 40% | 58% | 49% | 28 |
| E: Weekend, 5 kWp, Hungary | 21% | 41% | 98% | 89 |
Key finding: The SCR figures for heat-pump households (50–72% with battery) are not dramatically lower than industry claims of 60–90%. An installer quoting "70% self-consumption with a battery" for Scenario B would be essentially correct. The problem is that SCR tells an incomplete story. The household with 72% SCR and a battery still imports 59% of its electricity from the grid (SSR = 41%). The battery improved solar utilisation — it did not produce more solar.
The primary mechanism is winter demand saturation: in December, heating alone requires 5–8 kWh/day of electricity, while a 5 kWp system produces only 3–5 kWh/day. With demand exceeding supply, no excess exists to charge the battery.
4.3 Winter Hourly Analysis
Table 3: Hourly dispatch for Scenario B, 21 December
| Hour | Solar (kW) | Heat pump (kW) | Base load (kW) | Battery SOC | Result |
|---|---|---|---|---|---|
| 00–08 | 0 | 1.2 | 0.1 | 10% → 0% | Grid import |
| 09 | 0.3 | 1.2 | 0.1 | 0% | Grid import |
| 10 | 0.8 | 1.2 | 0.1 | 0% | Grid import |
| 11 | 1.2 | 1.2 | 0.1 | 0% → 5% | Partial charge |
| 12 | 1.5 | 1.2 | 0.1 | 5% → 15% | Charges |
| 13 | 1.4 | 1.2 | 0.1 | 15% → 22% | Charges |
| 14 | 1.0 | 1.2 | 0.1 | 22% → 22% | No change |
| 15 | 0.6 | 1.2 | 0.1 | 22% → 15% | Discharges |
| 16–23 | 0 | 1.2 | 0.1 | 15% → 0% | Grid import |
On this day, the battery gains roughly 1.2 kWh of usable charge over a few midday hours — and then discharges it all by late afternoon. The net battery utility is negligible: all heating from 16:00 onward comes from the grid. Simplified models that assume the battery charges to near-full every day miss this reality entirely.
4.4 Seasonal Battery Utilisation
Table 2: Battery utilisation by month (Scenario B, 5 kWp, 10 kWh, HP, Germany)
| Month | Avg charge (kWh/day) | Activity | Zero-charge days |
|---|---|---|---|
| Jan | 0.08 | Near zero — heating dominates | 20 |
| Feb | 0.55 | Very low — occasional midday excess | 4 |
| Mar | 1.40 | Low — shoulder season begins | 0 |
| Apr | 3.10 | Moderate — regular daily cycling | 0 |
| May | 4.50 | High — peak solar, battery fills daily | 0 |
| Jun | 4.80 | High — peak solar | 0 |
| Jul | 4.70 | High — peak solar | 0 |
| Aug | 4.00 | High — summer decline begins | 0 |
| Sep | 2.70 | Moderate — shoulder season | 0 |
| Oct | 1.20 | Low — heating season returns | 0 |
| Nov | 0.40 | Very low — winter-dominant | 0 |
| Dec | 0.04 | Effectively dormant | 23 |
Annual zero-charge days: 47. Full-cycle equivalents: ~89 (not 365). The battery is useful for roughly 8 months per year (March–October). In December and January — 43 of the 47 zero-charge days — the battery sits dormant. Solar peaks at ~1.1 kW on a December afternoon while heating alone draws 1.2 kW continuously; there's simply no excess to store.
5. Discussion
5.1 Why SCR Is the Wrong Metric for Electrified Households
The 60–90% self-consumption figures reported in industry marketing are not fabricated; they are achievable for the households they were designed for — those with gas boilers and 3,000–5,000 kWh/year consumption. Our simulation confirms 81% SCR with a battery for the MCS baseline case.
For heat-pump households, SCR with a battery is lower but still substantial: 55–72% in Scenarios B–D. An installer quoting "70% self-consumption" for a German heat-pump household with 5 kWp and a battery would be approximately right about SCR.
The problem is that SCR measures the wrong thing for electrified households. Here's why:
| Scenario | SCR (with bat) | SSR (with bat) | Grid import | What SCR hides |
|---|---|---|---|---|
| A: Gas, 4 kWp, UK | 81% | 75% | ~970 kWh | Looks great — and is |
| B: HP, 5 kWp, DE | 72% | 41% | ~5,000 kWh | 72% SCR sounds good; 59% grid import is the reality |
| C: HP, 8 kWp, DE | 55% | 51% | ~4,200 kWh | Lower SCR, but better SSR — bigger system helps |
| D: HP, 8 kWp, PL | 58% | 49% | ~4,800 kWh | Similar pattern |
Scenario B achieves 72% SCR — a number most installers would proudly quote. But SSR is 41%: the household still imports 5,000 kWh/year from the grid. The battery stored solar that would otherwise have been exported, which is genuinely valuable, but it didn't change the fundamental mismatch between a 5 kWp array generating 4,900 kWh/year and a heat-pump household consuming 8,500 kWh/year.
Temporal smoothing amplifies the SCR/SSR gap. Monthly-average models cannot capture the winter reality. Our hourly simulation shows 47 zero-charge days per year for Scenario B — days when the battery receives no net charge because all solar is consumed by real-time heating. A monthly model might show December solar of 150 kWh against December demand of 800 kWh and conclude the battery is cycling. An hourly model shows the battery sits empty for most of the month.
5.2 Implications for Policy and Consumer Protection
If consumers and policymakers focus exclusively on SCR, they will systematically overestimate what solar+battery systems can achieve for electrified households. The distortion is subtle because SCR numbers can be factually correct while still being misleading:
| What installers show | What it means | What it hides |
|---|---|---|
| "72% self-consumption with battery" | You're using 72% of your solar on-site | You still import 59% of your electricity |
| "10-year battery payback" | At current prices and SCR | Assumes gas-boiler demand profile |
| "Energy independence" | High SCR on your solar | SSR of 41% — you're far from independent |
This gap between SCR and SSR represents a systematic information problem. Battery subsidies based on SCR improvements may flow to installations where the battery genuinely improves solar utilisation but doesn't materially change the household's grid dependence. Transparency about both SCR and SSR — and about the demand profiles used to compute them — would allow consumers to make better-informed decisions.
5.3 Corrected Expectations
Based on our 365-day hourly simulation, realistic expectations for Central and Northern European households are:
| Household type | System | SCR (with bat) | SSR (with bat) | Grid import share |
|---|---|---|---|---|
| Gas boiler, home all day | 4 kWp + 7.5 kWh | 81% | 75% | 25% |
| Heat pump, regular occupancy | 5 kWp + 10 kWh | 72% | 41% | 59% |
| Heat pump, regular occupancy | 8 kWp + 10 kWh | 55% | 51% | 49% |
| Weekend home, no HP | 5 kWp + 10 kWh | 41% | 98% | 2% |
The counterintuitive result: bigger solar systems have lower SCR but higher SSR. An 8 kWp system for Scenario C achieves only 55% SCR (vs 72% for 5 kWp) — but SSR improves from 41% to 51%. More solar means more export (lower SCR) but also more absolute self-consumption (higher SSR). The right metric depends on your goal: SCR measures financial efficiency, SSR measures energy independence.
5.4 Geographic Limitations
This analysis focuses on Central and Northern European climates (Germany, Poland, UK, Hungary) where the winter heating season lasts 5–8 months. In Southern Europe (Spain, Italy, Greece, Portugal), winter solar production is 2–3× higher and heating demand is lower, producing substantially better self-consumption rates. A Barcelona household with a 5 kWp system and heat pump can expect self-consumption of 35–50% without battery and 50–65% with battery — roughly 50% higher than the German baseline. The general principle holds across all climates (heat pumps reduce self-consumption relative to gas-boiler baselines), but the magnitude of the effect scales with the length and severity of the heating season.
What This Means For You
If you're researching solar and battery systems, here's what the numbers actually mean:
- SCR and SSR are different questions. SCR asks "how well am I using my solar?" SSR asks "how much of my bill is solar covering?" For a heat-pump household, you can have 72% SCR (good utilisation) and still import 59% of your electricity (modest coverage). Both numbers matter — but SSR determines your bill.
- Bigger systems have lower SCR but higher SSR. An 8 kWp system "wastes" more solar to export (lower SCR) but covers more of your demand (higher SSR). The right size depends on whether you prioritise financial efficiency or energy independence.
- Battery payback is real but modest for heat-pump households. The battery boosts SCR by 15–25 percentage points and SSR by 10–15 points. This is valuable, but it doesn't transform the economics the way it does for a gas-boiler household (where the battery takes SCR from 48% to 81%).
- Winter is the reality check. Every model that shows high annual SCR also shows near-zero battery activity in December and January. A battery helps in summer, not winter.
- Ask your installer for SSR, not just SCR. If they can't tell you how much of your annual demand the system will cover, their model isn't modelling your household.
See also: Why 60–90% Self-Consumption Claims Are Wrong for the plain-English version of this analysis.
6. Conclusion
The 60–90% self-consumption figures pervasive in solar+battery marketing are achievable — for the gas-boiler households the industry standards were designed around. Our simulation confirms 81% SCR with a battery for the MCS baseline case. The problem is not that the numbers are false; it's that they answer the wrong question for an electrified household.
For a heat-pump household consuming 8,500 kWh/year with a 5 kWp system and 10 kWh battery, our hourly simulation shows 72% SCR — a figure that would satisfy most installer claims. But SSR is only 41%, meaning the household still imports 5,000 kWh/year from the grid. The battery improved solar utilisation; it did not close the demand gap.
The industry's reliance on SCR as the headline metric systematically obscures the reality that electrified households need more energy than rooftop solar can provide — particularly in winter. As heat pump and EV adoption accelerate, the relevant question is not "what fraction of your solar are you using?" but "what fraction of your demand is solar covering?"
Industry standards must be updated to include heat pump and EV demand profiles, and both SCR and SSR should be reported. Until then, consumers should treat 60–90% self-consumption claims as what they are: correct numbers for the wrong household.
References
DESNZ (2024). Home Energy Model Technical Paper 18: PV Generation and Self-Consumption. UK Department for Energy Security & Net Zero.
European Heat Pump Association (2024). European Heat Pump Market and Statistics Report 2024. EHPA, Brussels. https://www.ehpa.org/market-data/
Huld, T., Müller, R., & Gambardella, A. (2012). A new solar radiation database for estimating PV performance in Europe and Africa. Solar Energy, 86(5), 1803–1815.
Jäger-Waldau, A. (2016). PV Status Report 2016. European Commission Joint Research Centre, Ispra.
Luthander, R., Widen, J., Nilsson, D., & Palm, J. (2015). Photovoltaic self-consumption in buildings: A review. Applied Energy, 142, 80–94. https://doi.org/10.1016/j.apenergy.2015.06.037
MCS (2022). MGD 003: Solar PV Self-Consumption — A Method to Determine the Electrical Self-Consumption of Domestic Solar PV Installations with and without Storage. Microgeneration Certification Scheme, Issue 2.0. https://mcscertified.com/wp-content/uploads/2022/04/MGD-003-Solar-PV-Self-Consumption-Issue-2.0-Final.pdf
NEA (2023). Smart Solar in Barnsley: Full Report. National Energy Action, Newcastle upon Tyne. https://www.nea.org.uk/
Quoilin, S., Kavvadias, K., Mercier, A., Pappone, I., & Zucker, A. (2016). Quantifying self-consumption linked to solar home battery systems: Statistical analysis and economic assessment. Applied Energy, 182, 58–67. https://doi.org/10.1016/j.apenergy.2016.08.077
SolarPower Europe (2024). EU Solar Market Outlook 2024. SolarPower Europe, Brussels. https://www.solarpowereurope.org/market-outlook
Last updated: May 2026