Slower interventions at self-checkout - Host ratios and the 2026 SCO summits

Slower interventions at self-checkout: host ratios and the 2026 SCO summits

Self-checkout intervention times are the measure retailers have quietly started taking seriously, and the most useful thing to come out of ECR’s two 2026 self-checkout summits, hosted by Walmart in Dallas and Waitrose in the UK. Between them they drew around sixty retailers in America and eighty people in June, across store visits, table exercises and the latest academic research, including the Self-Checkout Loss Report 2026.

Two numbers frame the conversation. How long it takes a host to reach an intervention, which runs from fifteen to forty five seconds and sometimes far longer. And how long it takes to clear one, where under five seconds means nobody really interacted, they just cleared the screen and moved on. One Australian retailer deliberately went the other way, cutting the share of interventions cleared in under five seconds from 40% to 10%, and their net promoter score went up.

Professor Matt Hopkins and Colin Peacock also cover what good design looks like, why training hosts across hundreds of stores never quite finishes, the machine to host ratio that nobody has yet solved, and what changes when self-checkout moves out of grocery and into hardware and health and beauty. More on the wider picture at the self-checkout hub.

The full conversation follows.

What the 2026 self-checkout summits found: Colin Peacock and Matt Hopkins in conversation

COLIN:

Welcome to the Sell more, lose less podcast. This week I’m with Professor Matt Hopkins. And we’re basically going to go over some of the highlights, if you like, some of the key takeaways from a summit that was organised in the UK in June, and also draw on some findings from a similar summit that was held in Dallas.

And Waitrose and Walmart were kind enough to host those summits. And the summits were all around self checkout innovation. The two days consisted of store visits, they consisted of table exercises, the latest research findings from academia, including the great new study from Professor Matt Hopkins on SCO loss. But also lots of time for informal catch ups and discussions around items that are important to those whose feet are being held to the fire on managing self checkout participation and managing losses.

In America we had about sixty different retailers, though that was across multiple banners, from Walmart to Target to Giant Eagle to Aldi and others. And then in Waitrose in June, again, we had 80 people, Waitrose, Tesco through to Migros in Turkey, and many, many others, Albert Heijn, et cetera.

So a huge, if you like, a sample of retailers on which to base some of these observations that we’re going to share with you today. So Matt, if I may, come to you and maybe we first start talking about just the discussion that flowed from a retailer who presented a worldwide study of self-checkout design, both the fixturisation and what’s on the screen. And what we could take away from that presentation of results, but also then the discussion that flowed throughout both conferences around design.

SCO architecture, and what good self-checkout design does

MATT:

Yes, thanks, Colin. Yes, SCO architecture, as I often refer to it. SCO architecture, the design of SCO.

The design of SCO is really important. I think both in terms of the environment where SCO is, and really making sure that user experience is a really positive one. And I think there was a lot of talk around this in both of the summits, so in Dallas, and when we came back to the UK as well.

And I think there are some real key themes to take out of this. So I think you’ve got to think about the user experience. Are users getting to SCO areas? Do they know what’s going on? Is the interface easy to use in SCO? Are the instructions clear? So there’s all that sort of stuff.

Does it create that frictionless, positive customer experience there? We know that customer frustrations are a key generator of loss around SCO. They can’t use the systems properly, so they think, I’m not going to scan this.

So that user experience is really, really important. Is SCO accessible to people is another really important thing to think about here. Are the areas spacious? Are they easy to get through? Is it easy just to use them in terms of moving products around and scanning them?

I think what we’ve also got to think about in terms of SCO architecture and that kind of environment, we can’t get away from the reality that SCO does generate loss. We know that from the data that we have around this. So I think we have to think about creating a zone of control, as Professor Adrian Beck would always talk about, zones of control around SCO.

So these can’t be areas where there’s a sense of no risk for those people who do want to engage, possibly in criminality, or those who don’t want to scan products. So you have to make sure there’s an amplification of risk as well built into SCO architecture. That’s really, really important.

And I think what we did see across the two summits, and particularly when we came back to the UK, we saw some great work around, as you referred to, Colin, a study across the world that had been done by one retailer looking at SCO architecture and design. And they were talking quite a lot about flat table design SCOs and the ergonomics around this and the ease for customers. Actually, I’m really keen on this idea.

I kind of like SCOs where you go. They’re easy to use. You’ve got that flat table. There’s plenty of space. It’s easy to move your products around. And also where the instructions are really clear on what you have to do essentially to scan your products.

I think the SCO architecture piece, really important, and I think there’s some really, really useful issues to think about here in discussion going forward.

COLIN:

I particularly was taken aback by the portrait screen, as I think the presenter talked about when they went to Asia and other places. If you go into McDonald’s, you go into Tesco Cafe, the screens where you order stuff, your burgers or whatever, they’re all portrait.

And yet in the world of self checkout we seem to have gone the different way. It’s very much landscape. So again, I thought that was quite interesting, and showing the monitor of you being able to see what you scanned, I actually thought that was quite interesting, and we’ve seen some evidence of that. So I’m with you. I think SCO architecture is really, really important.

Another area you talked about was, and I think is another theme that is really high on the agenda, is around the staff training and investing in staff. And I think we are getting beyond just the challenge of staff turnover, and they often liken, well, if I invest in staff training it’s a bit like the Forth Bridge, because of the high turnover in stores. By the time I get to my 800th store to do the training and deliver the training, the very first store, everyone’s new.

And so no one’s remembered the initial training. I think retailers are getting over that. But what were you hearing in terms of conversations around and initiatives around staff training and coaching and mentoring, et cetera?

Training self-checkout hosts when turnover never stops

MATT:

Yes, I think as an industry, there’s a greater recognition now that staff are an important piece in this, are an important piece in the customer journey, and they’re an important piece as well in terms of loss prevention. I think there’s an acceptance of this. I do think, obviously, we have to think about technology, design and staff together.

I think that staff are a really important piece. If you have good SCO hosts, you know what they’re doing, they’re well trained, I think that will pay dividends for your stores. I mean, also as well, well trained staff can prevent all sorts of problems with SCO where you get those customer frustrations, maybe over age verification, detagging products, for example, customers who can’t work systems. Well trained staff who know how to deal with customers potentially know how to defuse situations where you maybe have got a risk of abuse or even worse, unfortunately, violence against the staff. That’s a really important piece.

Going back though to the delivery of training and the Forth Bridge problem, we might want to call it, where you’re just going round and round and round training. Yes, I think this is an issue where you get high turnover. But I think there are ways around this. I mean, there are technological solutions now to deliver training, where training can be delivered to staff through iPads, through online learning around SCO hosting. Obviously, some of the training has to be done at the coalface, so to speak. But I do think staff training is really important, and I do think retailers realise that this is important.

I think this is something which did come out of both summits very clearly.

COLIN:

Yes. And our data from the SCO loss survey, there was some evidence that it was working. But again, to your point about the industry needs to do a better job of measuring the impact. I think measuring the impact of SCO hosts is another one.

But let’s take it back to some of the basics, which is that if a customer has not scanned an item, you can see clearly they’ve not scanned it, they’ve left it in their bucket. It takes a bit of gumption, a bit of confidence to be able to say to that customer, I don’t think you scanned that item. Do you want me to scan it for you? And that’s hard, because you’re setting up a confrontation. And unless you’re trained to do that well, it probably doesn’t happen.

MATT:

Yeah, but I think if you stand and observe SCO hosts when they’re working, and I’ve done this a number of times, they’re busy people. They’re constantly dealing with issues. And I’ve tended to find that the way they will deal with non-scan behaviours will be, oh, you seem to have forgotten to scan. Are you having a problem?

So you’re not accusing people of theft. Not accusing people of intentionally trying to steal anything. So I think immediately the way they initiate those conversations or corrections tends to be in a very diffusive way, if that’s the right word to think about this. And I think this is really, really important.

And I think, look, a lot of retailers know some customers will want to trigger some kind of confrontation, and you have to just learn how to deal with that. And I think the SCO host training is a really important piece in that. And lots of the SCO hosts I’ve seen are just naturally very good at this. They’re just very good with customers. They’re very good at dealing with customers in the right sort of non confrontational way.

Timing an intervention, and the host ratio nobody has solved

COLIN:

Yeah, and I’m certainly starting to see ways in which the retailers are starting to measure better how long does it take to get to someone who’s created an intervention. We’ve heard everything from fifteen seconds to forty five seconds. I’m sure you’ve experienced it as well, like two minutes.

And then the second metric that I think people are really looking at is how long does it take them to clear that intervention, from the time that there’s the FOPs put there to then clearing it? If that’s less than five seconds, it probably suggests there isn’t a great deal of interaction going on there. It’s the get out my way. I’m just clearing.

MATT:

But this all comes back to the time piece, doesn’t it? There’s a lot of talk about ratios, SCO to host ratios. And I think we need to find out what the sweet spot is here between SCO to host ratios. And I think as well, we need to think about, I mean, you’re right, those interventions, are the SCO hosts just letting things go through because they’re so busy and they’re moving from intervention to intervention? Or are they actually doing something which will reduce loss in some way?

And I think we don’t really fully know at the moment, in all honesty. And for me, this is a part of the whole return on investment piece that retailers need to be thinking about here. So what’s the value of your SCO host? Is the value of them just to get people through the system and get them on their merry way? Or is the value of them actually in terms of recovery and loss prevention as well?

And we need to think about that. And we haven’t got that at the moment. We know that, I mean, the SCO loss report tells us that where you get more transactions, reaction times are slower for SCO hosts. There’s a lot of talk around that machine to SCO host ratio. And we don’t really know what that sweet spot is. And of course, that’s going to be dependent on types of stores as well.

COLIN:

Well, we did have one retailer that gave a really good case study from Australia actually. They showed the impact on the net promoter score when they were able to reduce the time it takes to get to the intervention down from on average less than fifteen seconds, and they increased that percentage. But importantly, they increased the percentage, and this is counterintuitive, of the actual time it took to clear the intervention. They went from 40% being under five seconds to 10% being under five seconds. And that’s what drove the NPS score.

We also saw when we went to Dallas, we saw Walmart, and this is in the public domain, so we’re not telling tales out of school here. But it was clear that in the stores we visited, they’d gone to a one to four ratio. They opened up a bank of four, and there was a SCO host overseeing all four.

So I think there’s some clues emerging around how to think about managing that ratio, but it’s certainly an open question. And right now having visibility to that ratio at any one point in time is hard for a retailer remotely. And we know it ourselves from spending time in stores that suddenly you’ve got a very quiet period. There’s one person at the SCO, one to one. And then suddenly there’s a rush and you’ve got twenty people there.

MATT:

Yeah, we see this all the time, and I see it, if ever I do observations around SCO, quite often you’ll be having a bit of a chat with a SCO host, and the next minute they’re dealing with four or five customers, they’re running around. I think it’s very difficult for retailers to plan out that flow here. So that’s another challenge that we need to think about going forwards.

COLIN:

Well, so let’s just close out with maybe three or four of the top challenges that you heard discussed. I know you’re looking at SCO beyond just grocery.

Self-checkout beyond grocery

MATT:

Yes, absolutely. So hardware stores, health and beauty type stores are obviously engaging this technology. As an industry, we’ve quite often thought about it as being a grocery thing predominantly.

But obviously, we’re now seeing more and more challenges with different types of retailers wanting to implement this technology. And again, I think for those retailers, the challenges come back to the architecture and design of SCO. For big hardware stores, for the B&Qs of this world, for example, or the Dunelms, how do they design SCO areas which are easy to use? Do they even need SCO areas with any sort of tabletop design or whatever?

So for example, somewhere like B&Q, most of the customers will be going through with large items, they can scan in the basket. So it’s got that kind of user experience for those stores as well.

I think also with a lot of the newer uses of SCO technologies is how do they deal with things like tagging? Hard, soft tags on maybe cosmetic products, which tend to be quite small, quite often quite expensive. These are desirable products for would be thieves. They’re craved items. So they’ve got to think about that kind of balance between tagging, loss prevention for a lot of those products, and again, going back to this frictionless customer journey. So there’s a challenge there in relation to this and around those products, customer friction.

And also as well, for a lot of hardware stores, there are age verification issues with a lot of products. So those that are selling, for example, restricted cleaning products under the Offensive Weapons Act, corrosive substances. Those age-restricted products, again, how do they deal with that age restriction, that age verification at the point of sale here? So they do have challenges.

They’re not challenges which are a million miles away from the traditional retailer format at SCO. But I think there are slight nuances and differences with the challenges that they face going forward.

COLIN:

Yes, you’re absolutely right. And I think even within grocery, we’ve got the flexi tills now, where you’ve got someone overseeing self checkouts and also the main bank. You’ve got the twisty turnaround till. So you’ve got all sorts of new challenges for those sitting around cross functional teams to wrestle down that balance between a smooth customer journey and scan accuracy and payments without walkaways, which we’ll come to another stage, but walkaways certainly are a big issue as well.

But Matt, thank you so much for taking the time and joining us both in Dallas and in Waitrose. We’re going to be in Turkey next year and maybe in America as well with these particular innovation challenges, innovation summits. So yeah, big thanks and see you next time.

MATT:

Thank you. Thank you.