Home Assistant publishes its anonymised survey dataset: what 8,616 users said
The Open Home Foundation has released the anonymised results of its first Home Assistant community survey. Here is what is in it, how it was anonymised, and what it tells you about why people choose local control.
By OwnYourStack Admin · Published
On 26 August 2026 the Open Home Foundation published the anonymised results of the first Home Assistant community survey as an open dataset. It ran for about seven weeks from December 2024 until early February 2025 and received 8,616 submissions. The post, by Annika Schulz, explains what was asked, how the data was cleaned and where to find it.
Where to get it
The dataset is on Zenodo, with a codebook explaining each column, a README, the list of survey questions and descriptive statistics. The anonymisation scripts and k-anonymity analysis are on GitHub, in the foundation’s user-research repository. The post says the data comes with an open licence but does not name it, so read the Zenodo record before you reuse anything.
How it was anonymised
- Names, technical identifiers, timestamps and free-text comments were removed, along with sensitive details about health, disability, religion and other household members.
- A k-anonymity review was run and some answers were grouped into broader categories (for example, “5–6 protocols”).
- About 12% of responses were withheld as a precaution.
- Legal advisers reviewed the approach and each decision is documented.
What the results show
| Finding | Result |
|---|---|
| Why people chose Home Assistant: local control | 89.3% |
| …its open source nature | 84.0% |
| …powerful customisation | 81.7% |
| Typical number of devices | 11 to 200 |
| Live in a home they own | 75% |
| Primary users working or studying in technology or engineering | 78.3% |
| Started their smart home with Home Assistant | About one third |
Those top three reasons line up with how the project has behaved lately: the move to Home Assistant Link, local-first partners such as LEDVANCE and the privacy-minded 2026.10 release.
Read it with care
The foundation lists the limits itself. The survey was too long, with an average completion time of more than 40 minutes. It leaned towards committed users, so it says little about people who tried Home Assistant and left. It was English-only. A heavily technical, owner-occupier sample is not “the average smart-home owner”, and the numbers should not be read as one.
A follow-up community survey is planned, as is a report on how users arrive at Home Assistant. Researchers who want to work with the data can contact the foundation through its email address or GitHub discussions.
Source: Home Assistant blog and Open Home Foundation, 26 August 2026.
