S1 Ep 5
Self-checkout errors and 'not-on-file' scans create frustration, temptation and add friction to the shopper journey.
In this podcast Colin Peacock and Professor Adrian Beck look at the scale and nature of this problem, and the key drivers of non-scanning (hello multiple QR codes on packaging!)
They consider the interventions that retailers are adopting to address these problems, to reduce the number of interventions and the losses associated with it. As well as how to improve the customer experience.
Podcast summary
Self-checkout errors can be driven as much by process and packaging as by malicious intent, especially when items won’t scan or appear “not on file”.
Colin Peacock and Professor Beck share insights from an ECR Retail Loss self-checkout working group, including how QR codes placed near barcodes can trigger missed scans at scale.
They discuss practical responses: clearer on-screen prompts, better reporting loops to merchants and head office, and working with manufacturers on SCO-friendly packaging.
Cleaned transcript
Setting the scene: self-checkout errors and item not on file
COLIN:
Hello, and welcome to the ECR Retail Loss Sell More and Lose Less podcast series. My name is Colin Peacock.
This week I’m joined by Professor Beck and we’re looking at the problem of item not on file at the self-checkout. What can be done to prevent it, and when it happens, what can be done to minimise loss.
Yesterday we had our self-checkout working group meeting and we were talking about the problem of perhaps unknown items. Unknown because they might not be on file, or unknown because they couldn’t be read, or they’re reading something else at the self-checkout.
That meant an intervention which took between 12 and 24 seconds to resolve, if it was resolved correctly.
We had about 50 or 60 retailers join the call from around the world, including our friends from Australia and America and the West Coast, and it was a really great discussion for about 75 minutes.
We heard from a lot of the retailers what they were thinking, and we had some benchmark data shared earlier in advance. A very productive session, I thought.
We’re going to revisit it in 2025, but what were your notes and what should we be looking out for in 2025?
What’s really driving self-checkout errors at fixed SCO
BECK:
Yes, initially on paper it looked a very dry session, but it generated some really interesting debate.
We know from one of our earlier surveys on SCO that retailers think about 48% of the losses at fixed SCO are probably non-malicious. Things are going wrong, and we’ve never really been able to unpick much data around what is going wrong in that space to generate that level of loss for the retailer and, obviously, the shopper.
This was an interesting insight. First of all, is it because the item hasn’t been registered on file, and therefore when the customer comes to scan it, nothing happens?
They then have to try and get a member of staff to come over and key in some sort of generic code, or the right code depending on what it could be. That can mean you lose margin.
What was really interesting was the extent to which retailers began to talk about it not necessarily being “not on file”, but that the barcode wasn’t being read easily.
Self-checkout errors caused by QR codes vs barcodes
BECK:
There’s this incredible mix-up with QR codes that we really weren’t aware of. The scanner is picking up a QR code instead of the barcode, because they’re often placed in very close proximity on packaging increasingly.
What seemed to be happening is the shopper probably hears the bleep because the scanner makes a noise that it’s seen a QR code. But the consumer thinks it’s scanned the item and off they go.
It hasn’t actually registered.
We heard one retailer saying this happened 42,000 times a week where the scanner was doing this.
It was really insightful in terms of helping us think about what’s going wrong sometimes. Not only this QR code issue, but also the classic problems around packaging itself: how easy is the positioning of barcodes, the type of packaging, and this issue around the wrong barcode being read by the consumer.
Particularly pack barcodes and distribution barcodes, which can easily confuse consumers. There have been great examples where the consumer thinks they’re scanning the right barcode, but they’re not, and it generates loss.
So it was really useful for helping us think through what may be going on to explain why these mistakes are being made at self-checkout.
Reporting and fixing self-checkout errors across stores
COLIN:
We heard a lot of retailers had put in place a process whereby they could communicate these errors to the merchants and to the head office people.
We also heard from one retailer that was really leveraging the SCO hosts—those right at the front line—to report these up as they happen.
We’ve heard about retailers who’ve got their own software that can automatically update every single store if they pick up on an error.
But these QR codes are a big one. I think it’s going to get bigger because there are going to be more of them.
BECK:
Yes. I actually went away and started looking at some packaging in my house, and quite frequently the barcode and the QR code are very close to each other.
I presume it’s a classic example of people not talking to each other in terms of the introduction of new products and packaging. It probably makes a lot of sense to the designer to put the QR code near the barcode, but they’ve not spoken to the people who have responsibility for managing loss in stores.
We saw a comment in the chat that sharply observed exactly that—no one’s talking to each other.
COLIN:
That was very sharply observed.
BECK:
I wouldn’t let this go, but it also struck home that you need to be much more joined up in thinking this through.
Because SCO is becoming so dominant, particularly in grocery, you’ve now got to go back and say: to what extent is product packaging SCO-friendly?
It generates so many potential problems within retail businesses that we really need to go back to the root cause, rather than sticking plasters on it, and say: manufacturer, you need to think this through now. It’s causing these problems.
Here are the top 50, top 100 lines that are consistently causing problems at the checkout—what can you do to minimise this issue through the design and layout of your packaging?
COLIN:
Yeah, that’s really good.
Reducing friction when self-checkout errors happen
COLIN:
You also made a point about friction in the call, and from a consumer perspective—how does it create confusion?
BECK:
We’ve seen this with walkaways and the issue of contactless payment. Sometimes it can be tricky for the consumer to know whether it’s worked or not.
The thing bleeps, off they go, and they think it’s happened because it’s contactless. That’s proved challenging for some retailers.
I think it’s the same here. How do you let the consumer know something’s gone wrong?
We’ve seen good examples where retailers make maximum use of the screen. It changes colour. It says to the consumer: hold on, payment hasn’t happened, we’re going to get you some help.
I think it’s the same when items don’t scan. They need to use that interface to inform the consumer that something needs to happen. We’re going to get a member of staff over. This needs to be fixed before you can proceed.
We heard from one retailer that was doing that, but for a lot of others, the system bleeps and things just move on. There’s an expectation the consumer will keep an eye on all this.
Really, why should they? It’s down to you to make sure packaging works properly and creates a seamless experience for the shopper.
COLIN:
Fascinating.
We’re going to get some really good insights into the non-malicious as we learn more about this. We’re planning to revisit this meeting, and I think we’re also going to do some work to find out what those top items are where you’re not actually scanning.
So that’s all for now. Thank you very much for your updates and key insights.
We’ll revisit this meeting in February 2025 and see where we’ve got to: have the reporting and dashboards been put in place, what is the actual rate—is it 0.8%, is it 5%—and we’ll go from there.
Thank you very much indeed, and we’ll see you next time. Cheers.
BECK:
Thank you very much. Bye bye.
