Behavior analytics
Applies to:
- BlinkID v7.5 and later
- BlinkID Verify v3000 and later
- BlinkCard v3.20 and later
What are behavior analytics?
Product adoption analytics tell you how much your integration is being used. Behavior analytics tell you how it goes for the end user: where they hesitate, where they retry, and where they give up.
The module turns data into a set of metrics you can act on: completion, duration, drop-off, and the guidance users were shown while capturing. Each metric answers a question you would otherwise have to guess at, and most of them point at a concrete fix.
As with usage analytics, we never collect personally identifiable information from scanned documents or cards. Behavior analytics are derived from session and UX events only.
Which metrics can you observe?
| Metric | Answers | Typical use |
|---|---|---|
| Completion rate | How many users finish what they started? | Headline performance indicator |
| User journey duration | How long does each stage take? | Spotting friction and stalls |
| User journey funnel | Where exactly do users drop off? | Prioritizing what to fix first |
| Live instruction messages | What guidance did users receive? | Tuning capture settings and environment |
| Scans per platform | Which distributions carry your traffic? | Release and support prioritization |
| Top browsers and devices | What do your users actually run? | Compatibility and testing scope |
| Retry rate | How many attempts does a session take? | Frustration and friction analysis |
Completion rate
The percentage of users who initiate a session and reach the end of it.
Completion rate = completed sessions / initiated sessions × 100
This is the single most observed performance metric, and the one to watch after every release. A drop that is limited to one platform, version, or document type is usually a regression. A drop across the board is usually an environment or UX change on your side.
User journey duration
The time users spend in each stage of the session. We report the median per stage, together with the minimum and maximum outliers, so a handful of extreme sessions cannot skew the picture.
The median tells you what a normal session feels like. The outliers tell you whether a subset of users is stuck. Long durations almost always mean the user is fighting the capture, for example because of poor lighting, an unsupported document, or a camera that never reaches focus.
User journey funnel
The stage-by-stage view of how many users move forward and how many leave. It can be visualized either as a Sankey diagram, when you want to see how the flow splits, or as a bar chart, when you want to compare stages directly.
The funnel pinpoints how much drop-off occurred at which stage of the session. This lets you aim corrective measures at the one stage that is costing you users, instead of at the flow as a whole.
Live instruction messages
During image capture, the SDK continuously evaluates the surrounding environment and relays guidance to the user, such as move closer, hold steady, or reduce glare.
This metric shows which of those messages your users actually received, and how often. It is the most direct hint at what can be tuned to improve the experience. A high share of lighting-related instructions points at where and when your users scan, while repeated framing instructions usually point at the camera settings or the scanning UX itself.
Scans per platform
The breakdown of scans between the different software distributions you run, for example the Web SDK, iOS, and Android.
Use it to see where your volume really sits before deciding where a fix or an upgrade is worth the effort.
Top browsers and devices
A drill-down into the same traffic, one level deeper. It shows which browsers and devices your users choose. This tells you what to test against, and whether a performance problem is in fact a problem with a single device family or browser version.
Retry rate
The average number of attempts a user needed in order to complete a session.
A retry rate near 1 means users get it right the first time. Anything meaningfully above that is a strong frustration signal. It is often visible long before it shows up in the completion rate, because users retry for a while before they give up.
How can you filter the data?
The module supports advanced filtering, so you can pinpoint exactly the problematic cohort and fix the problem at its core, rather than optimizing across your entire user base.
You can filter by:
- date and date range
- platform
- product and version
- application ID
- document type
- document country
- document region
Filters combine, and they apply to every metric on the page at once. In practice, this is how most investigations run: start from the headline completion rate, then narrow down by platform and version until the drop is attributable to one cohort.
Where can you access the data?
- Go to our developer portal.
- Open Dashboards.
- Select Behavior analytics.
Behavior analytics data is available exclusively through the graphical user interface and through JSON exports. Every view you can filter, you can also export with those filters applied.
Is it included in my plan?
Similarly to usage analytics, behavior analytics are complementary. The module is available to you alongside the analytics you already have.
Limited usage applies to certain Distribution and Drop-off rate insights. These are metered as exploration credits, which are shared across your whole organization rather than granted per user. Opening one of those insights consumes a credit. The remaining balance is shown in the module.
When the credits are spent, the metered insights are no longer accessible, while the rest of behavior analytics stays available. To restore access, or to discuss a plan that includes them permanently, contact our Support or your account manager.