Reducing Internal Theft and Staff Dishonesty - How to tackle the taboo and really reduce shrink

Tackling staff dishonesty and reducing internal theft is increasingly top of mind for retailers, and this episode unpacks why staff dishonesty is both more complex and harder to spot than many teams assume.

Colin Peacock and Professor Emmeline Taylor explore how internal theft now ranges from discount abuse and return manipulation to risks emerging from e-commerce models like BOPIS and BORIS.

They argue that because only a small proportion of offenders are ever detected, low-cost, communication-led deterrence and better evidence from test-and-control experiments could make a meaningful difference at scale.

Cleaned transcript

Setting the scene on reducing internal theft

COLIN:

In this week’s podcast, I’m with Professor Emmeline Taylor, and we’re going through an update on how retailers can tackle and bring down the cost of staff dishonesty in their business.

I wanted to check in with you on staff dishonesty, or internal theft, because it does seem very much top of mind with a lot of retailers right now that we speak to in our community of 400 or so. It’s really popping up there.

And I think perhaps a good thing is that it’s not a problem that requires a lot of capital investment, unlike some of the problems we deal with, like self-checkouts and shopping trolley theft. Perhaps there’s a way of tackling this that doesn’t cost an awful lot of money, but it’s actually a very complex and sensitive problem.

I just wanted to touch base with you on some of the work you’ve been doing over the last six months or so in this space, just to get us up to speed with the latest thinking and where we’re heading in terms of new research.

So just start us off by taking us through what we think we mean by this sort of internal theft, staff dishonesty. How do we think about it? How big is it?

Defining internal theft and how it’s changing

EMMELINE:

Internal theft has always been, and continues to be, a sensitive topic. It’s relatively taboo.

We spend a huge amount of time and energy focusing on customer theft, but a lot of the data points suggest that internal theft is not that far behind in terms of the level of losses and shrinkage that is attributed to it.

But first of all, what is internal dishonesty? I think this is really key, because nowadays we’ve seen quite a diversification of how this manifests. It’s no longer somebody helping themselves to a sandwich at lunchtime or taking something that was broken and then using it themselves. There’s a whole range of ways that still happens, of course.

There is misappropriation of cash and tangible assets. There’s manipulating returns, faking voided items, changing labels, misuse of discounts and sharing them with friends and family. There are so many ways that this shows itself.

Reducing internal theft risks in BOPIS and BORIS

EMMELINE:

Two really key developments in recent years around e-commerce have begun to make this even more complex. So buy online, pick up in store – BOPIS – and buy online, return in store – BORIS – are creating further opportunities for collusion and maybe even coercion at times.

I don’t think we necessarily know enough about the offender profile. Who’s doing this? How are they doing it? And how do we plug those vulnerabilities?

The scale of the problem and some striking survey figures

EMMELINE:

You mentioned: just how big an issue is this? In the report that ECR will be publishing shortly, we pulled together some of the recent survey findings, and there are some caveats around that – these are based on industry surveys.

Some of the figures are really quite shocking. For example, in Australia, 24 percent of losses are attributed to internal crime. In New Zealand, it’s 32 percent. And then in the UK, 40 percent of thefts – and we should note there are slightly different measurements here – are being attributed to internal.

There are also other data points that suggest internal thieves steal at a higher value than external thieves. Some are saying up to four times on average per incident compared to an external thief.

So there certainly needs to be more attention on internal theft, and I think there’s some low-hanging fruit that could be looked at to begin to encourage employees to be more honest.

The detection gap in reducing internal theft

COLIN:

You mentioned those numbers. If you think about what you’ve just said, that could be an extraordinarily high value associated with internal theft.

But whenever I talk to retailers, they sort of say, “Well, we only really catch a small part of that” – maybe 2 percent, perhaps, of what they think is happening are they able to detect.

That can either mean that their current systems for detecting those internal cases – whether that’s relying on a whistle-blower line, or an analysis of EPOS behaviours, or maybe other rules they can set up in their systems – aren’t working hard enough. Or it’s that this is really difficult to detect.

Detecting skimming from a lorry, the truck driver taking off two or three items just for themselves, some stores not checking – how big a problem is that? Collusion at self-checkouts – someone once said that when self-checkouts were introduced, previously you had one person who was sweethearting all the time, and now that person can sweetheart four people at a time. What a great multiplier effect that is.

What’s your thought about this big gap? Do you see that as well? Is that something that you’ve come across in your discussions?

EMMELINE:

Yeah, absolutely. Pulling together some of the industry figures and beginning to look at those apprehensions – the number of internal cases that are identified and proceeded against – you very quickly realise that they are a drop in the ocean compared to the scale of activity.

Just crunching the numbers in terms of the typical levels of shrink, how much is attributed to internal, and the average case value of an incident, you can begin to see the scale of this problem.

Looking at that, it’s estimated that probably only around 2 percent of internal thieves are actually detected and then any sort of action taken based upon that.

So what do you do with the other 98 percent? How do you appeal to them and try and reach them?

There was a great ECR study on this.

Recruitment, integrity checks, and what gets missed

COLIN:

Certainly communication – we’ll come onto that – is one of them.

But also just how you recruit. I think that changed dramatically during COVID. I remember hearing you talk about stories of old when you tried to get a job yourself. How was it? Was it Pizza Hut or something? Those sort of integrity tests that perhaps used to happen face to face maybe don’t happen any more.

There’s much more online recruitment. Maybe the general store manager is not involved. And of course, you’ve just got to get bodies, especially at peak times. So as long as you’ve got two arms and two legs, happy days.

Some of the background checks probably don’t happen. It’s hard. So it’s really challenging and, especially as you say, at those peak times where you’re having to recruit thousands of people at pace, I think sometimes some of those processes are missed.

EMMELINE:

Yes, exactly. Some of those processes are missed in terms of background checks or whether the psychometric testing that I did at Pizza Hut – that I told you about – that unfortunately I didn’t pass. Obviously a long time ago.

COLIN:

Yeah, let’s not dwell on that. You didn’t inhale.

Why internal theft is hard to target with interventions

EMMELINE:

There’s a whole range of different ways that retailers are trying to deter and then detect potential offenders.

Through surveillance and monitoring, through randomised screening or screening the whole workforce as they leave a shift, through exception reporting – there are all these tools at people’s disposal.

But the thing with internal theft is it’s not specific to any role. It doesn’t matter whether you’ve worked in the business for one day or 20 years. We see examples right across different levels of seniority and all different roles of people being dishonest.

So it’s very difficult to target interventions.

Now, there are some successes. The difficulty as well is that it’s always hard to know just how many people were deterred, because you can’t measure what didn’t happen. You can when you bring in a new intervention and compare it to what was happening previously, but you still can’t see everything that’s been prevented.

Reducing internal theft with communications and signalling

EMMELINE:

A really cost-effective way that this project that was run at Loughborough University looked at was whether concepts taken from marketing and criminology could be applied to signal to the 98 percent that the business has eyes on them; that they have ways of knowing whether there is dishonest practice taking place.

What was interesting from this project is that it wasn’t necessarily punitive. It wasn’t, “We’re going to find you and clamp down on you.” It was more about appealing to people’s morality, to their sense of family in terms of their work family and their colleagues, and then thinking about the potential of what they might have to lose.

Is it really worth taking that sandwich and then potentially facing reputational damage or the loss of employment that might follow?

I think this is quite an exciting space. I think we’ve moved on from those posters that just become part of the wallpaper. So I think maybe having a rejuvenated way of looking at marketing and communications could actually be the tipping point for some of those 98 percent.

Opportunity, deterrence, and the “speed camera” effect

COLIN:

If I just go back to some of your points about criminology, I think I remember the sort of spectrum of criminology: everyone can be a criminal if they’re presented with the right opportunity, the low likelihood of being caught, the rewards, and no sanctions.

Retail is pregnant with opportunity. People working late at night, working past their shift, doing an extra shift – removing that opportunity is going to be really hard for retailers to put into practice.

So I suppose that perhaps puts, to your point, the role of communication a bit like a speed camera. Just to remind people that, “Hey, you could get caught here. There’s a set of eyes watching over you.”

I think in the report we’re going to put some examples from the students. I’m not sure we’ll ever see those exact artists’ impressions or posters in reality, but we hope they can inspire some of the readers.

What’s the best-case scenario in your mind from where the research takes us next?

Revitalising the conversation and small gains at scale

EMMELINE:

First and foremost, it’s about trying to revitalise those conversations about internal theft and recognise just what a huge problem it is. It does deserve to have more attention.

There’s a whole range of different solutions that can be used – from training staff, from having hotlines, through to all these different technological ways that have been introduced to deter and detect where there is dishonesty.

What I’m excited about with this particular project is not only revisiting and revitalising those conversations, it’s about whether this could make a difference.

The scope is huge, because if we think 98 percent of incidents are not detected, even if you just change 2 percent of your workforce, that’s a big win when you then look at the average value of these cases, and when you look even bigger at the picture of the costs that internal theft is causing businesses.

Building evidence for reducing internal theft via communications

COLIN:

You make a really good point with this last point about recording it. I don’t think you could go onto Google Scholar and find out what is the impact of communications on staff dishonesty.

I think you can look at wages and other things like that, but I don’t think there’s any real evidence out there to support communications. And of course, that makes it really easy for senior managers, who are very trusting of their staff, to say, “Well, we don’t need to do this because there’s no financial benefit. We’d just upset more people and make it harder to recruit people. So let’s keep away from this stuff. We’re a great place to work.”

I think if we can bookend the point around staff dishonesty communications along the lines of: “We’re watching over you” and “We are a family”, and show that it makes an impact on the bottom line, measured by the shrink results, then perhaps we might stop not using communications, if you like.

EMMELINE:

Exactly. There’s been a whole range of studies within criminology that have evaluated the effectiveness of having posters and communication campaigns.

One of the key ones found that simply having a pair of eyes can change behaviour for the positive. It was found that bike theft was hugely reduced at a university campus when they had this big poster overlooking the bike racks.

It’s also been shown to encourage people clearing their tables in a canteen, people putting their doggy poop bags in the bin. There’s a whole range of ways that it can be used to nudge people to do the right thing in the moment.

But I think, translating it into retail, it’s really about the messaging – the optics of this – that those in the boardroom will be concerned about.

When I speak to staff, one of the issues that comes up is not that they feel these posters would be pointing the finger at them, but that it gives them support. If they see somebody else doing wrongdoing, they feel more comfortable that, “Oh, OK, so that is wrong. There’s a clear message around it, and now I feel more empowered and more comfortable to alert senior staff or to do something about it.”

So I think it’s actually about empowering your workforce rather than pointing the finger.

COLIN:

Wow, what a great point. Yeah, that’s really, really powerful.

Test-and-control experiments and closing the loop

COLIN:

Hopefully we’re going to get, in the next 12 to 18 months, a tsunami of test and control store experiments on communications in all different forms. That will allow us to say with some degree of certainty, “We have got, to your point, evidence out there already from outside, plus some experiments we’ve learned of that have made some difference” – 15 percent, I think, was one of the examples.

But to get that tsunami of test and control experiments out there would be really good for us, and something we could write up and put in the documentation around this particular topic.

But for now, thank you very much. And safe travels, or welcome back if you’re… I’ve been in God already. That’s a time zone. Weird one. Anyway, enjoy.

EMMELINE:

Thank you very much.

COLIN:

Thanks, Colin. Bye-bye.