07-23-2025 05:13
07-23-2025 05:13
We are GoVida, and we are currently using Fitbit to fetch user activity data, specifically Activity and steps data, using the Fitbit Web API (/1/user/[user-id]/activities/steps/date/[date]/1d/1min.json).
We have identified a critical issue:
When a user logs a Manual Walk Activity through the Fitbit Mobile App (by manually selecting a time range and entering a distance), the equivalent number of steps are added to their daily step count in their Fitbit account. These manually added steps are then returned by the Fitbit API and appear as part of their regular step data.
We expect the Fitbit API to either:
Exclude manually logged activities from the step count returned in the API response,
Or provide a reliable way to identify and filter out those steps on our backend.
Our goal is to only count steps recorded by the device's sensors (e.g., tracker-based step counts) and exclude steps derived from manually entered activities like a manual walk or run.
User logs a Manual Walk Activity (e.g., 2 km for 30 minutes).
Fitbit internally calculates and adds steps for this activity.
These calculated steps are reflected in the total daily step count.
When GoVida fetches the user's steps data using the API, these manually added steps are indistinguishable from sensor-based steps and get included in our records.
This affects the integrity of our step challenges and user rankings. Users can inflate their step counts by manually adding activities, which goes against our fair usage policies and impacts the authenticity of challenges.
Is there a way to identify or filter steps originating from manually logged activities (vs. device-tracked)?
Is there a specific field or endpoint where such data is marked as user-entered or non-sensor-based?
If not currently possible, is this distinction something Fitbit plans to support in upcoming versions of the API?
We appreciate the support and would be happy to provide additional details if needed to investigate this further.