Device data doesn't always show up as expected. A user forgets to charge their watch. A ring lives on the nightstand for a week. A smart scale gets stepped on twice a month. An app update breaks a sync. An account quietly disconnects. None of these are highly unusual, particularly when capturing device data at scale. Yet most downstream systems treat device data like it's always complete.
Looking For What’s Missing
Missing data isn’t easy to find. A user who wore their ring every night for a month and a user who wore it on nine of those nights can produce the same average. So when a heart rate trend shifts or a sleep pattern devolves, how can you determine whether the person changed, or just their device habits? And when a device stops syncing entirely, teams usually find out weeks later, if ever, after the gap has already made its way a care decision. So it becomes difficult for teams to answer questions like:
- Is this user's data complete enough to score?
- Has this device gone quiet, or was it never used much?
- Is the account disconnected, or just not producing data?
- Which providers actually deliver dependable data for our users?
- Can we trust this trend? Teams piece together their own rules, and the logic differs across providers. And the moment a new device type shows up, it all breaks.
Introducing Device Reliability
Device Reliability tells you whether Junction has been receiving data consistently from each user's connected devices, for every data type you care about. No more row-counting checks against raw data. You create one Continuous Query, and Junction keeps a daily-refreshed reliability snapshot for every user on your team, broken down by provider, device type, and data type. Each snapshot includes:
- coverage across the last 30 days, gaps included
- recency, or how long since data last arrived
- a reliability status: reliable, degraded, sparse by pattern, stale, or insufficient history
- connection status, reported separately That last one matters. A connected account that's stopped producing data and a disconnected account are different problems -- with different fixes. Device Reliability measures missingness. It doesn't paper over it. Same window, same data, same answer, every time.
What Teams Can Do Now
With Device Reliability, teams can:
- gate model inputs, so risk scores and clinical logic only run on data that clears the bar
- catch silent failures early -- prompt a re-sync when a device goes stale, not a month later
- compare data quality across providers, device types, and data types
- show clinicians and members how complete the data behind a trend actually is
Why This Matters
Device Reliability sets the foundation for trust in device data. Every insight built on device data assumes the data is actually there. Device Reliability turns that assumption into something you can measure, and build on.
Ready to Get Started?
Device Reliability is available now to Junction Sense customers through Continuous Query. Fields, status rules, and examples are all in the Device Reliability docs. Contact: sales@junction.com or your Junction CSM.