Household Appliances Data: Certified Models, Shipments and Smart-Meter Corpora · Head-to-head
EU EPREL — European Product Registry for Energy Labelling vs Pecan Street Dataport — High-Resolution Household Energy & Water
Which household appliances data: certified models, shipments and smart-meter corpora data fits your job: EU EPREL — European Product Registry for Energy Labelling, or Pecan Street Dataport — High-Resolution Household Energy & Water. API, files, or your warehouse. Daily, weekly, or hourly.
EU EPREL — European Product Registry for Energy Labelling
Pecan Street Dataport — High-Resolution Household Energy & Water
Coverage, side by side
| EU EPREL — European Product Registry for Energy Labelling | Pecan Street Dataport — High-Resolution Household Energy & Water | |
|---|---|---|
| Geographic | European Union, with availability flags extending to Northern Ireland (XI) | United States: Austin, Texas primary, plus New York and California testbeds |
| Temporal | Registry live since 2011 for legacy-label models, 2019-2021 for rescaled groups; every record carries first-publication and on-market start/end dates | Enrolments from 2011 onward, with per-home, per-feed availability percentages documenting how much history each house carries |
| Granularity | One record per registered model per version, with per-cycle and annual consumption figures | One row per home per interval: 1-second and 1-minute electricity, metered gas, water and indoor temperature, annual surveys |
What each contains
They tie on 2 attributes. Pick by fit, not by loyalty.
| EU EPREL — European Product Registry for Energy Labelling | Pecan Street Dataport — High-Resolution Household Energy & Water | |
|---|---|---|
| Publisher | European Commission (EPREL) | Pecan Street Inc. |
| Subject lens | Mandatory EU registry of every energy-labelled model placed on the market: energy class, efficiency index, consumption, capacity, noise, dimensions and market dates across 33 product groups | Research panel of instrumented homes: 1-second and 1-minute electricity at whole-house and per-circuit level, plus gas, water, indoor temperature, solar, EV charging and occupant surveys |
| Documented fields | 33 fields, definitions verified during research | 27 fields, definitions verified during research |
| Geographic coverage | European Union, with availability flags extending to Northern Ireland (XI) | United States: Austin, Texas primary, plus New York and California testbeds |
| Temporal coverage | Registry live since 2011 for legacy-label models, 2019-2021 for rescaled groups; every record carries first-publication and on-market start/end dates | Enrolments from 2011 onward, with per-home, per-feed availability percentages documenting how much history each house carries |
| Granularity | One record per registered model per version, with per-cycle and annual consumption figures | One row per home per interval: 1-second and 1-minute electricity, metered gas, water and indoor temperature, annual surveys |
| Scale | 33 product groups; observed live counts include 55,757 refrigerating appliances and 23,178 washing machines in the 2019 groups alone | 2,072 home records in the documented roster across a network of more than 1,000 instrumented homes |
| Secondary industries | None | None |
| Best for | Product questions: which models exist, what they declare, when they entered or left the EU market | Behavior questions: what appliances and homes actually consume, second by second |
Where they're equivalent
More than their shapes suggest. Both anchor on 2011: EPREL's registry reaches back to legacy-label models registered that year, and the earliest Dataport enrolments in the sample are dated 1 February 2011. Both resolve to one stable unit of observation — a model here, a home there — with lifecycle dates attached, so longitudinal pulls behave on either side. Both treat consumption as a first-class attribute rather than a footnote: rated per-cycle energy on one side, metered per-circuit draw on the other. Where they cannot be made equivalent is geography — EU market frames against US homes — and that asymmetry decides most sourcing questions between them.
The verdict
Verdict: sample both — they solve different halves of the same appliance question.
Take EU EPREL — European Product Registry for Energy Labelling if the question is about products: which models exist, what class they claim, how consumption, noise and capacity are declared, when they entered or left the market. Compliance screens, assortment benchmarking, label-taxonomy work and competitor model counting live here. Product and competitive-intel teams usually want this record first.
Take Pecan Street Dataport — High-Resolution Household Energy & Water if the question is about behavior: how much a dishwasher circuit actually draws on a Tuesday evening, how EV charging stacks onto a household load shape, what rooftop solar does to the curve. Load research, disaggregation modelling and demand studies live here. Data scientists tend toward it, while market researchers often end up needing both halves of the rated-versus-measured story.
Sample both, pick by fit. See EU EPREL — European Product Registry for Energy Labelling · See Pecan Street Dataport — High-Resolution Household Energy & Water
Or take both in one feed
Yes — as claim and check. The canonical join is label-versus-behavior: take EPREL's rated figures for a model family and set them against metered circuit data from comparable homes to quantify the gap between laboratory declaration and kitchen reality. Two cautions keep the merge honest. There is no shared key — EPREL identifies registration numbers, Dataport anonymised home ids — so any model-to-circuit mapping is an analytical assumption you document, not a join you run. And the geographies do not intersect: an EU-declared rating meets US-metered usage, so treat the result as a directional benchmark rather than a compliance finding. For the survey layer that enriches the panel side, see survey microdata. Browse the rest of the shelf at the household appliances data hub.
Datadory ships either record alone or both merged onto one schema, delivered daily, weekly, or hourly — your call. Or take both in one feed.
API, files, or your warehouse. Daily, weekly, or hourly.
Fair questions
Is EU EPREL better than Pecan Street Dataport?
Better at different jobs. EPREL wins anything about products: 33 regulated groups, hundreds of thousands of models, declared classes, consumption and noise, EU-wide. Dataport wins anything about behavior: 1-second per-circuit electricity, water, gas and surveys from more than 1,000 instrumented homes. They barely overlap, which is why the verdict is to sample both.
Can rated label figures be checked against metered household data?
Yes, analytically. EPREL stores declared per-cycle energy; Pecan Street meters individual appliance circuits second by second. Setting a model family's rated kilowatt-hours against metered draws quantifies the gap between laboratory declaration and kitchen reality. There is no shared key between the two, so the model-to-circuit mapping must be documented as an assumption.
Do EU EPREL and Pecan Street Dataport cover the same countries?
No. EPREL is framed on the European Union, with availability flags reaching Northern Ireland. Dataport instruments homes in the United States — Austin, Texas above all, with testbeds in New York and California. Any combined analysis crosses continents, which is why combined results read as directional benchmarks rather than compliance findings.
How far back does each dataset reach?
Both effectively start in 2011. EPREL's registry holds legacy-label models registered from that year, with the rescaled label groups arriving 2019-2021 and first-publication dates on every record. Pecan Street's earliest sample enrolments are dated 1 February 2011, with per-feed availability percentages showing how much history each home actually carries.
Can I get both datasets from Datadory?
Yes — sample both and pick by fit, or take both in one feed. Each arrives normalized to its verified field dictionary (33 fields on the EPREL side, 27 on Dataport's) with sample rows attached for inspection before anything ships, delivered daily, weekly, or hourly — your call.