Ask what secQR produces, and the honest answer is a stream of events. Every scan writes a record: which code, what state came back, when it happened, roughly where, and what class of device asked. Taken one at a time these are trivia. Taken at territory scale over a quarter, the stream behaves like instrumentation for a channel you otherwise cannot see into.
The platform ships with dashboards over this stream, and for most teams that is where the work starts. The more interesting questions are what to plot first, and when to move the stream into the BI stack your analysts already live in.
What a scan event carries
- The code identity, and through it the product, batch, and work order it was generated against
- The state the platform returned: active, revoked, expired, or not found
- A timestamp, and the interval since the previous scan of the same code
- A coarse location when the scanner permits it, a network-inferred region otherwise
- Device class and request pattern, the raw material for duplicate and bot detection
None of this identifies a person unless they chose to identify themselves, for a warranty or a grievance. It is operational exhaust, and reading it is the point of the exercise: the codes exist to be scanned, and the scans exist to be read.
One scan is an anecdote. A season of scans is an instrument.
Three plots to build first
Duplicate rate by territory. When two scans of one code arrive from places a single pack could not plausibly travel between, the platform flags a duplicate. Plot the flag rate per territory per week. A flat low line is reassurance. A rising line in one territory is a lead, and it tends to arrive well before the customer complaints do.
Scans versus declared production. Compare scan counts per SKU per territory against what your ERP or WMS says you shipped there. If a territory scans more units than you sent it, you have learned something no survey will tell you: product is arriving through a channel you do not control, or codes are being copied at scale. Either answer is worth a meeting.
First-scan lag. Measure the time from a code's activation to its first field scan, as a distribution per batch. The median tells you how fast stock moves through a channel. The tail tells you where it sits. A batch whose first scans never arrive deserves a phone call to the distributor before it becomes a write-off.
The dashboards you get and the stack you keep
BI teams meet vendor dashboards with justified scepticism, because every vendor dashboard eventually loses to the tool the team already lives in. We agree, and we built for it. The included dashboards cover the operational views above, refresh without engineering effort, and are the right home for the people who act on alerts. For everything past that, the stream exports: scheduled file drops or API pulls into your warehouse, on a cadence you set.
The export path exists because the joins are where the value compounds. Scan data alone tells you about scanning. Scan events joined to shipment records expose diversion. Joined to sales data, they show where verification behaviour tracks purchases and where it does not. Those joins belong in your warehouse, next to the rest of your operation, under your own access controls.
A practical cadence: daily file drops for scan events, weekly summaries for state changes, tightened during a launch or an investigation. Files suit warehouse teams that already run nightly loads. The API suits teams that want events sooner than the next file. The two paths carry the same records, so moving from one to the other later is a schedule change and nothing more.
Reading the numbers without fooling yourself
Two cautions before the first review meeting. Repeat scans do not automatically mean fraud. Packs get scanned in warehouses, at retail intake, and by curious customers, so a healthy channel produces plenty of multi-scan codes; the duplicate flag weighs distance and timing, and a human should still read the flagged cases. The other caution: territory metrics built on coarse location are directional, so treat regional numbers as a compass, and reach for precise instruments only where consumers opted in.
Start with the three plots, inside the platform, this quarter. The first dashboard worth keeping is the one your territory managers argue with, because an argument means the numbers touched something real. When the arguments outgrow the built-in views, that is the day to open the export tap.