Most advice about customer feedback assumes you can email the customer. For a brand that sells through a distributor, to a retailer, to someone who paid cash at the counter, that assumption is simply false. You know how many cases left the warehouse. You do not know a single name.
So the usual pitch for scan-to-respond feedback — that it beats email on response rate — misses the point. A better response rate on a list you do not have is still nothing. The pack is the one thing that reaches every buyer, and a code printed on it is the only channel that does not depend on knowing who they are.
Twelve seconds, start to finish
No app, no account, and nothing typed that the code already knows.
A code, printed per unit
Every pack carries a different code, applied on the line you already run.
They scan it
Straight from the phone camera. Nothing to download, nothing to sign into.
Give first, then ask
Show the authenticity result or the warranty. Ask one question after that.
It reaches your systems
A webhook fires on submit. Typically inside a second.
A complaint becomes a diagnosis
With the same QR image on a million packs, a complaint tells you that somebody, somewhere, was unhappy. With a code per unit, eleven complaints resolve into one sentence — and the customer typed none of it:
The same identity underpins TraceGuard and SupplySight. A feedback response, an authenticity check and a movement record are the same scan read three ways, not three systems to buy.
Into the CRM you already run
Feedback that lives in a dashboard nobody opens is feedback you did not collect. On submit we POST a JSON payload to an endpoint you control; from there it is an ordinary integration. The rating and comment become a case record, the code, batch and SKU become fields on it, and anything at or below your threshold opens a ticket and alerts someone. Three things worth knowing before you scope it:
- Not every CRM accepts feedback objects directly. HubSpot’s feedback-submission endpoints are documented as read-only, so the usual answer there is a ticket or a custom object. Check your vendor’s current API docs before anyone commits to a design.
- Deduplicate on the code, not the scan. People scan the same pack twice. Without a rule you get two cases and two people ringing the same customer.
- You do not need IT for version one. The same webhook drives Zapier, Make or Power Automate. Prove the loop with a spreadsheet and an email alert first.
A low rating should not produce a thank-you screen. It should open a ticket, with the batch and the photograph already attached.
Scan rate is a design problem
You will see claims of 40 to 60 per cent response rates. Those come from companies selling QR survey tools. Independent survey-methodology research puts the effect of adding a QR code somewhere between about one percentage point and nothing at all. What published benchmarks do agree on is more useful:
For comparison, CustomerGauge puts emailed NPS at about 12 per cent — measured against people you can already email. Multiply that by the share of your buyers whose address you actually hold. So the realistic promise is not a spectacular response rate. It is a steady stream of responses from buyers you could not otherwise reach, each one naming its batch. For a quality team, a hundred responses that name a batch beat ten thousand that do not.
Not a separate product to buy
Feedback capture is what these modules already do with a scan of the code on your pack.
Benchmarks above are drawn from published vendor reporting and from peer-reviewed survey-methodology research, and they disagree with each other by an order of magnitude. We have said which is which rather than quoting only the flattering number. The NPS figure is CustomerGauge’s published average. None of these are PackSense measurements — the only number that will tell you anything is the one from your own first SKU.
Frequently asked questions
What response rate should we actually expect?
Lower than the brochures say. QR survey vendors publish 40 to 60 per cent; independent survey-methodology research puts the effect of adding a QR code at somewhere between about one percentage point and nothing at all. Published benchmarks agree only that placement and a specific promise decide it — a code on the back with no reason given sits under one per cent. None of these are PackSense measurements.
Where does the feedback end up?
Wherever you already work. On submit we POST a JSON payload to an endpoint you control, so the rating and comment can become a case or ticket in your CRM with the code, batch and SKU as fields on it. Check your vendor’s current API docs first — HubSpot’s feedback-submission endpoints, for one, are documented as read-only.
Do we need IT involved to start?
Not for version one. The same webhook drives Zapier, Make or Power Automate, so you can prove the loop with a spreadsheet and an email alert on one SKU before anybody scopes a real integration.
Is feedback capture a separate product to buy?
No. It is what LoyaltyLoop, PromoPulse and TraceGuard already do with a scan of the code on your pack — one platform reading one code, not a seventh module.