Smart Plug Statistics 2026
By Axis Intelligence Research
Co-author: Jeff Aaron | Last updated: August 5, 2026 | License: CC BY 4.0
Standby power now costs a typical U.S. household between $94 and $189 a year at 2026 electricity prices, up to 89 percent above the $100 ceiling the Department of Energy has published since 2022. Meanwhile 23.7 million American homes own a smart speaker and use it to control nothing at all. Smart plugs sit exactly in that gap.
Quick Answer
According to Axis Intelligence Research, 47.14 million U.S. households (35.6 percent) owned at least one smart speaker in 2024, but 23.7 million of them did not use it to control a single device. Applying DOE and Berkeley Lab standby-share ranges to 2026 electricity prices, the recoverable standby load per household is $94 to $189 a year, or $12.5 billion to $25.0 billion nationally.
Key Findings
- According to the U.S. Energy Information Administration’s 2024 Residential Energy Consumption Survey, 47.14 million U.S. households owned at least one smart speaker, or 35.6 percent of all occupied primary residences.
- Axis Intelligence Research finds the Dormant Control Gap (DCG) — households owning a voice assistant that control no device with it — stood at 23.7 million homes in 2024, equal to 50.3 percent of all smart-speaker households.
- Axis Intelligence Research calculates the 2026 Standby Recovery Value (SRV) at $94.46 to $188.92 per household per year, using EIA’s 10,380 kWh average annual residential consumption and its 18.2 cents/kWh price forecast for 2026.
- According to the Federal Trade Commission’s November 2024 staff study of 184 connected products, nearly 89 percent of manufacturer product pages did not disclose how long the device would receive software updates.
- Axis Intelligence Research finds the dormant-control share barely moves by housing type, running from 48.2 percent in single-family attached homes to 52.3 percent in small apartment buildings — a spread of just 4.1 points.
How Many U.S. Households Have Smart Plugs and Smart Home Controls?
The honest answer is that no federal statistical agency counts smart plugs directly. EIA’s Residential Energy Consumption Survey, the closest thing the United States has to a census of household devices, asks about smart speakers and about what those speakers control. It does not ask about smart plugs as a product category. Any figure claiming a precise national smart plug installed base is a vendor estimate wearing a lab coat.
What the 2024 RECS electronics tables do measure is the addressable base, and that is more useful.
| Metric | 2024 households (millions) | Share of all homes | Source |
|---|---|---|---|
| All occupied primary residences | 132.54 | 100% | EIA, 2024 RECS HC4.1 |
| Own at least one smart speaker | 47.14 | 35.6% | Axis calculation from EIA 2024 RECS HC4.1 |
| Own one smart speaker | 22.66 | 17.1% | EIA, 2024 RECS HC4.1 |
| Own two smart speakers | 11.17 | 8.4% | EIA, 2024 RECS HC4.1 |
| Own three or more | 13.31 | 10.0% | EIA, 2024 RECS HC4.1 |
| Use a speaker to control any device | 23.44 | 17.7% | Axis calculation from EIA 2024 RECS HC4.1 |
| Own a speaker, control nothing | 23.70 | 17.9% | EIA, 2024 RECS HC4.1 |
That last row is the number this report is built around.
The Dormant Control Gap (DCG), 2020 to 2024
DCG is an Axis Intelligence Research metric with a deliberately narrow definition: the count of households that own at least one smart speaker and report using it to control zero devices — no lighting, no thermostat, no locks, no television. It is read directly from a single RECS line, which is what makes it reproducible by anyone with the same table.
| Year | Speaker households | Controlling ≥1 device | DCG households | DCG share of speaker homes | Source |
|---|---|---|---|---|---|
| 2020 | 40.00M | 18.00M | 22.00M | 55.0% | EIA press release, 2020 RECS |
| 2024 | 47.14M | 23.44M | 23.70M | 50.3% | Axis calculation from EIA 2024 RECS HC4.1 |
Both directions matter. The share fell 4.7 points, so the people who bought voice assistants are slowly wiring them to something. But the absolute gap grew by 1.7 million households, because speaker ownership expanded faster than control adoption in the first two years and only caught up later. Controlling households grew 30.2 percent between the two surveys while speaker households grew 17.9 percent.
Jeff Aaron, Smart Home: Anyone who has done installs recognises this shape. The speaker is a gift, an impulse buy, a bundle throw-in. Commissioning it to control an actual appliance is a separate act of will, and it stalls at the first device that has no smart version — the coffee maker, the bedroom fan, the aquarium heater, the space heater in the workshop. A smart plug is the cheapest way to end that stall, which is precisely why the DCG is the number worth watching. It is not a market-size estimate. It is a queue.
What Does Standby Power Actually Cost in 2026?
Here is where most published smart plug statistics go stale without anyone noticing. The number in near-universal circulation — standby power costs the average U.S. household “as much as $100 per year” — comes from a Department of Energy Energy Saver post dated February 9, 2022. It was reasonable then. Electricity has not stayed still since.
The Standby Recovery Value (SRV): formula and inputs
SRV is the Axis Intelligence Research estimate of the annual dollar value of standby electricity a single U.S. household could target with plug-level control, priced at the current year’s residential rate. The formula, the inputs, and their sources are all on the table:
SRV = A × S × P
A = 10,380 kWh average annual residential electricity consumption per customer
(EIA: 865 kWh/month, 2024)
S = 0.05 to 0.10 standby share of residential electricity use
(DOE Energy Saver; Lawrence Berkeley National Laboratory)
P = $0.182/kWh forecast average U.S. residential price, 2026
(EIA Short-Term Energy Outlook, May 2026)
| SRV scenario | Standby share | Standby kWh/year | SRV per household | Source of inputs |
|---|---|---|---|---|
| Low | 5% | 519 | $94.46 | EIA, DOE, LBNL |
| Midpoint | 7.5% | 778.5 | $141.69 | EIA, DOE, LBNL |
| High | 10% | 1,038 | $188.92 | EIA, DOE, LBNL |
Snapshot date: August 5, 2026.
At the high end, SRV runs 88.9 percent above the $100 figure DOE published in 2022 — not because homes got leakier, but because the price of a kilowatt-hour did the work. EIA’s Short-Term Energy Outlook puts the 2026 residential average at 18.2 cents/kWh, roughly 5 percent above 2025, and EIA’s own bill analysis shows the price rising from 16.0 cents in 2023 to 16.5 cents in 2024 while consumption per customer also climbed 2 percent.
Scaled across the 132.54 million occupied primary residences RECS counts, that is $12.5 billion to $25.0 billion a year in U.S. household standby spending, with a midpoint near $18.8 billion.
What the SRV does and does not claim
SRV is the size of the target, not the size of the prize. It is an estimate, labeled as one. A household cannot cut standby to zero — the refrigerator, the router, the smoke alarm, and the DVR all have to stay energised, and some standby draw is doing useful work. Berkeley Lab’s Standby Power project notes that per-device draw has fallen toward 0.5 watts from the 1 to 3 watts typical two decades ago, while device counts rose enough to leave total standby roughly where it was. LBNL’s own conversion factor is the one to keep in your head: one watt of continuous draw is about 9 kWh a year, which at 2026 prices is $1.64.
That factor is why single-device smart plug savings claims deserve suspicion. Switching off a 2-watt device 16 hours a day saves about 12 kWh a year, or $2.18. The plug itself draws power to stay on Wi-Fi.
Which Household Devices Are Worth Putting on a Smart Plug?
RECS gives the installed base for exactly the categories where standby draw concentrates and where no smart version of the appliance exists. This is the practical shopping list, ranked by how many homes actually have the thing.
| Device category | Households with at least one (millions) | Share of homes | Source |
|---|---|---|---|
| Internet streaming device | 88.08 | 66.5% | Axis calculation from EIA 2024 RECS HC4.1 |
| Printer, scanner, copier or fax | 82.20 | 62.0% | Axis calculation from EIA 2024 RECS HC4.1 |
| Desktop computer | 50.44 | 38.1% | Axis calculation from EIA 2024 RECS HC4.1 |
| Video game console | 47.87 | 36.1% | Axis calculation from EIA 2024 RECS HC4.1 |
| DVD or Blu-ray player | 38.79 | 29.3% | Axis calculation from EIA 2024 RECS HC4.1 |
| Home theater or audio system | 33.15 | 25.0% | Axis calculation from EIA 2024 RECS HC4.1 |
| Cable or satellite box without DVR | 31.65 | 23.9% | Axis calculation from EIA 2024 RECS HC4.1 |
| Cable or satellite box with DVR | 26.82 | 20.2% | Axis calculation from EIA 2024 RECS HC4.1 |
The printer line is the one worth pausing on. 82.2 million households keep a printer, the second most common device on this list, and it is used for minutes a week while drawing standby power for all of the other 10,000 minutes. It has no smart variant. It is the archetypal smart plug load.
Jeff Aaron, Smart Home: The provisioning question that actually decides whether a plug survives to month three is not which ecosystem it joins — it is whether the appliance behind it tolerates hard power cuts. A printer, a lamp, a fan, a slow cooker: fine. A DVR mid-recording, a set-top box that needs a nine-minute boot, a desktop with unsaved work: not fine, and the household learns that once and never trusts the plug again. Cut power to the wrong thing on day two and the plug is in a drawer by day thirty. Half of the DCG is people who tried and got burned.
Do Smart Plugs Save Money? Running the Payback Math
The honest version, with the inputs on the table. Take four dormant loads on one plug — printer, monitor, speaker dock, charger brick — drawing a combined 6 watts and switched off 16 hours a day:
6 W × 16 h × 365 d = 35.04 kWh/year avoided
35.04 kWh × $0.182 = $6.38/year saved
Less the plug's own standby draw (roughly 1 W): 8.76 kWh × $0.182 = $1.59
Net: $4.79/year
At typical retail pricing, that is a multi-year payback on energy alone. Anyone selling a smart plug primarily as an energy-savings device is selling arithmetic that does not close.
Where the numbers do close is at the other end of the load curve. High-draw appliances on schedules — a 1,500-watt space heater running two hours a day instead of six, a dehumidifier, a pool pump, a water heater on a timed circuit — move real kilowatt-hours. A 1,500-watt heater running four fewer hours a day for a 120-day season avoids 720 kWh, or $131 at 2026 prices. That is 28 times the printer-cluster case from a single scheduling decision.
The measurement function is arguably worth more than the switching function. A plug with energy monitoring tells a household which of its loads is actually expensive, and that diagnosis is what redirects behaviour. Households already running the numbers on large loads will recognise the same logic we applied to what it costs to charge an electric car at home: the per-kWh rate is the only thing that turns a wattage into a decision.
One load never to put on a smart plug
EV charging. A Level 1 EVSE pulls roughly 12 amps continuously for 8 to 10 hours on a standard NEMA 5-15 outlet, which is a continuous-duty load that consumer smart plugs are not rated to carry — as covered in our breakdown of Level 1 versus Level 2 charging and how AC and DC charging actually works. Space heaters, window air conditioners, and anything with a compressor deserve the same caution: check the plug’s amperage rating against the appliance nameplate, not against its marketing copy.
Smart Plug Adoption by Housing Type: The Segment Nobody Publishes
Voice-control penetration varies sharply by housing type. The dormant-control share barely varies at all. That combination is the most useful thing in this dataset for anyone sizing a smart plug market.
| Housing type | Speaker households (millions) | Speaker penetration | DCG share of speaker homes | Source |
|---|---|---|---|---|
| Single-family attached | 3.40 | 39.3% | 48.2% | Axis calculation from EIA 2024 RECS HC4.1 |
| Single-family detached | 31.86 | 38.6% | 50.1% | Axis calculation from EIA 2024 RECS HC4.1 |
| Apartment (5 or more units) | 7.54 | 30.1% | 50.9% | Axis calculation from EIA 2024 RECS HC4.1 |
| Apartment (2 to 4 units) | 2.77 | 28.6% | 52.3% | Axis calculation from EIA 2024 RECS HC4.1 |
| Mobile home | 1.57 | 23.6% | 52.2% | Axis calculation from EIA 2024 RECS HC4.1 |
Penetration spans 15.7 points, from 39.3 percent in townhouses down to 23.6 percent in mobile homes. The dormant share spans 4.1 points and shows no clean gradient — small apartments and mobile homes sit at the top, townhouses at the bottom, and everything else clusters near half.
Axis Intelligence Research reads that as evidence the barrier is device-side, not dwelling-side. If the obstacle were wiring, tenure, or landlord permission, apartments would show a much wider gap than detached houses. They do not. What stops households is that the appliance they want to control has no smart version — and that is a problem a plug solves identically in a townhouse and a mobile home.
What people are already controlling
| Control category | Households (millions) | Source |
|---|---|---|
| Lighting | 13.72 | EIA, 2024 RECS HC4.1 |
| Security system, cameras or locks | 11.95 | EIA, 2024 RECS HC4.1 |
| Television, set-top box or streaming device | 11.17 | EIA, 2024 RECS HC4.1 |
| Temperature | 8.23 | EIA, 2024 RECS HC4.1 |
| Other | 1.65 | EIA, 2024 RECS HC4.1 |
The four named categories all have purpose-built smart hardware behind them: bulbs, cameras, TVs, thermostats. “Other” — the residual bucket where a plug-controlled fan or coffee maker would land — reaches just 1.65 million households, 3.5 percent of speaker owners. Every appliance without a native smart version is competing for that single line.
Interoperability: Does Matter Fix the Smart Plug Problem?
The Connectivity Standards Alliance released Matter 1.6 on June 17, 2026. The headline additions speak directly to the friction in the RECS numbers: NFC-based commissioning, which lets a device be provisioned by tapping a phone against it before it is even powered on, and Joint Fabric, which lets multiple authorised controllers co-administer one shared Matter network rather than each ecosystem holding its own copy of the device.
For plugs specifically, commissioning is the whole game. On-off switching is the simplest device type Matter defines; nothing about a plug’s function needed standardising. What needed fixing was the setup path, and the ledger of Matter releases since 1.0 in October 2022 reads as a sustained assault on exactly that.
Jeff Aaron, Smart Home: Standards documents do not install themselves. Matter 1.6 shipped in June; the plugs on shelves this holiday season were designed against 1.3 or 1.4, and firmware that adds NFC commissioning to hardware already in a drawer is not a given. The month-three test for a plug is unglamorous — does it still respond after a router reboot, a firmware push, and a password change — and no specification version answers that. Ask what the device does when the cloud is unreachable. A plug that keeps its schedule locally is a different product from one that becomes a dumb always-on adapter the moment the vendor’s servers hiccup.
That is not a hypothetical concern. The FTC’s November 2024 staff study of 184 connected products found nearly 89 percent of manufacturer product pages disclosed nothing about how long the device would receive software updates — for a class of product the Commission noted may lose its smart functionality, become insecure, or stop working entirely when updates end.
Are Smart Plugs Safe? What the Recall Record Shows
The federal recall record for smart plugs specifically is thin, which is itself worth stating plainly rather than inflating. The clearest entry: on September 7, 2023, the Consumer Product Safety Commission announced a recall of approximately 80,000 Emporia North America smart plugs (recall 23-274) that were not adequately grounded, posing a shock hazard. The firm had received two reports; no injuries were reported.
The remedy detail is the part that generalises. Emporia disabled the recalled plugs remotely, and consumers whose plugs could not reach the internet had to ship them back for the refund to process. A device whose off switch lives on someone else’s server is a different category of household hardware from a power strip, and buyers should price that in — alongside the amperage rating, which is the specification that actually determines whether a plug can carry the appliance in front of it.
Methodology
Collection. Every figure in this report was retrieved from a primary source document on August 5, 2026, and logged with its URL and retrieval date in the accompanying CSV. Household counts come from EIA’s 2024 Residential Energy Consumption Survey, preliminary housing characteristics release of March 2026, Table HC4.1 (Electronics in U.S. homes, by housing unit type), collected from nearly 17,000 households across all 50 states and the District of Columbia. The 2020 comparison points come from EIA’s March 30, 2022 press release accompanying the 2020 RECS, the first RECS cycle to include smart home devices.
Derived household counts. RECS reports “None” categories rather than ownership counts. Ownership figures in this report are computed as total households minus the “None” line — for example, 132.54 million minus 85.40 million reporting no smart speaker gives 47.14 million speaker households. Each derivation is recorded in the method_note column of the CSV.
DCG (Dormant Control Gap). Households owning at least one smart speaker that report using it to control no device, read directly from the RECS line “Does not use speaker to control devices.” Expressed as an absolute count and as a share of smart-speaker households. Snapshot dates: 2020 and 2024 survey reference years.
SRV (Standby Recovery Value). SRV = A × S × P, where A is EIA’s average annual residential electricity consumption per customer (10,380 kWh, derived as 865 kWh/month × 12, 2024), S is the standby share of residential electricity use (5 to 10 percent, per DOE Energy Saver and Lawrence Berkeley National Laboratory), and P is EIA’s forecast 2026 average residential price of $0.182/kWh. Published as a range with a midpoint at S = 7.5 percent. National totals multiply SRV by the 132.54 million occupied primary residences in the 2024 RECS. Snapshot date: August 5, 2026. SRV is an estimate, not an observation, and is labeled as such throughout.
What this dataset cannot tell you. No federal survey counts smart plugs as a device category, so nothing here is a smart plug installed base. RECS measures smart-speaker control, which is a proxy for smart home engagement rather than a measure of plug ownership; households that control devices exclusively by app and never by voice are invisible in these tables. The 5 to 10 percent standby share is a national range from published research, not a per-household measurement, and any single home may fall outside it. The 2020 comparison uses figures EIA reported as rounded (“about 40 million,” “nearly 18 million”), so the 2020 DCG carries that rounding. Commercial market-size and shipment estimates were reviewed and excluded: published 2026 global smart plug market values from major research firms ranged from $3.46 billion to $5.29 billion, a spread too wide and too methodologically opaque to merge with survey data. Where independent estimates disagree by 50 percent, publishing the disagreement is more useful than averaging it away.
About This Dataset
smart-plug-statistics-2026.csv contains 68 rows covering U.S. smart-speaker and smart-home-control adoption, device installed base by category, housing-type segmentation, standby power economics, product-support disclosure, and safety recalls. Every row carries the metric, dimension, value, unit, as-of date, source organisation, source document, source URL, retrieval date, primary-source flag, Axis-calculated flag, and a method note for every derived figure. Temporal coverage: 2020 to 2026. Spatial coverage: United States.
License: Creative Commons Attribution 4.0 International (CC BY 4.0). Free to reuse, redistribute, and build on, including commercially, with attribution.
Citation line: Axis Intelligence Research, Smart Plug Statistics 2026, 2026.
APA — Axis Intelligence Research. (2026). Smart plug statistics 2026: Adoption, standby savings, and the dormant control gap [Data set]. Axis Intelligence. https://axis-intelligence.com/smart-plug-statistics/
MLA — Axis Intelligence Research. Smart Plug Statistics 2026: Adoption, Standby Savings, and the Dormant Control Gap. Axis Intelligence, 2026, axis-intelligence.com/smart-plug-statistics/.
Chicago — Axis Intelligence Research. “Smart Plug Statistics 2026: Adoption, Standby Savings, and the Dormant Control Gap.” Axis Intelligence. Last modified August 5, 2026. https://axis-intelligence.com/smart-plug-statistics/
Frequently Asked Questions
How many watts does a smart plug use just sitting there?
Berkeley Lab’s standby research puts most modern devices below 0.5 watts, though a Wi-Fi plug maintaining a network connection typically sits above that. Use LBNL’s conversion factor to price it: one watt of continuous draw is about 9 kWh a year, which at the EIA’s 18.2 cents/kWh forecast for 2026 costs $1.64. A plug drawing 1 watt to save a 2-watt device that is idle 16 hours a day nets out positive, but not by much.
Will a smart plug still work if the manufacturer shuts down its servers?
It depends entirely on whether the plug executes schedules locally or relays every command through the vendor’s cloud, and manufacturers are poor at telling you which. The FTC found that nearly 89 percent of the 184 connected products it surveyed in 2024 disclosed nothing about how long software support would last. Matter-certified plugs commissioned to a local controller degrade more gracefully, because the control path does not leave the house.
Can I put a space heater or an EV charger on a smart plug?
No for EV charging, and check the nameplate carefully for heaters. A Level 1 EVSE draws roughly 12 amps continuously for 8 to 10 hours, which is a continuous-duty load consumer smart plugs are not rated to carry. Space heaters commonly draw 1,500 watts, at or beyond the rating of many plugs. Match the plug’s stated amperage to the appliance nameplate before it goes in the wall, not after.
Does energy monitoring on a smart plug measure accurately enough to act on?
For relative comparison between loads, generally yes — accuracy is sufficient to tell you which of your appliances is expensive. For low-wattage measurement it degrades; LBNL notes that meter error grows proportionally larger below about 1 watt. Treat a plug’s reading of a 0.4-watt charger as directional and its reading of a 900-watt dehumidifier as reliable.
What happens to my automations when the router reboots or the Wi-Fi password changes?
This is the single most common failure mode in month three, and it is why the commissioning path matters more than the feature list. Wi-Fi plugs bound to a specific SSID and password generally need re-provisioning after a credential change; some recover automatically, many do not. Matter 1.6, released in June 2026, added NFC-based commissioning to reduce the pain of re-provisioning devices in awkward physical locations, but firmware support varies by product.
Is Matter certification worth paying extra for on a plug?
For an on-off plug the functional gain is small, since switching is the simplest device type in the specification. The gain is in ecosystem portability and in Matter 1.6’s Joint Fabric feature, which lets multiple authorised controllers co-administer one shared network — relevant for households running more than one platform, rentals, or new-construction handovers. If your home has one ecosystem and no plans to change, certification buys optionality rather than function.
Have smart plugs ever been recalled for safety reasons?
Yes, though the federal record is short. In September 2023 the CPSC announced a recall of approximately 80,000 Emporia North America smart plugs that were not adequately grounded, creating a shock hazard, following two reports and no injuries. The recall’s mechanics are instructive: units were disabled remotely, and plugs unable to reach the internet had to be shipped back before a refund would issue.
How much can a household realistically save with smart plugs in a year?
Axis Intelligence Research estimates the total standby load available to target at $94 to $189 per household per year at 2026 prices, but no household captures all of it — refrigerators, routers, and alarms have to stay on. Realistic savings concentrate in scheduling high-draw appliances rather than eliminating small standby loads: cutting a 1,500-watt space heater from six hours a day to two over a 120-day season avoids about 720 kWh, or roughly $131.
Why does no source publish a smart plug installed base for the United States?
Because no federal survey asks. EIA’s Residential Energy Consumption Survey covers smart speakers and what they control but has never included smart plugs as a device category, and commercial market-research estimates for 2026 span $3.46 billion to $5.29 billion in global value using undisclosed and incompatible definitions. Axis Intelligence Research publishes the addressable base from survey data rather than an installed-base number that cannot be sourced.