Predicting Inventory Inaccuracies Prioritising SKU

Why inventory record inaccuracy matters

Predicting inventory inaccuracies helps retailers focus audits and interventions on the SKUs most likely to be wrong, rather than spreading counting effort evenly.

Colin Peacock speaks with Aris Syntetos, Yacine Rekik and Christoph Glock about ten years of ECR research into inventory record inaccuracy, including how widespread errors can be and why they undermine on-shelf availability, customer service and loss control.

They share headline findings, including the potential sales upside of fixing records and how retailers can use existing stock movement and audit data to predict discrepancies and prioritise action.

COLIN:

Hello and welcome to the ECR Retail Loss Sell More and Lose Less podcast series.

My name is Colin Peacock. I’ve worked in the retail and FMCG business for more than 40 years, working on the dark side of retail on retail loss at every point in the supply chain.

But we’ve been working at ECR with these wonderful people for over 10 years, looking around the problems of inventory record inaccuracy. You’ll hear a bit more about that as they go about sharing their work.

But first of all, let’s go and hear from them. Let’s start with Aris. Who are you?

Meet the researchers

ARIS:

Hi, thank you very much, Colin. I’m Aris Syntetos. Hi, everybody.

I’m a professor of decision science at Cardiff University and I specialise in inventory forecasting.

I’m very much interested in modelling and solving large-scale operational problems, one of which is inventory record inaccuracy.

CHRISTOPH:

Hello. I’m Christoph Glock. I’m based in Darmstadt. I’m with the Technical University of Darmstadt in Germany.

I’m heading the Institute of Production and Supply Chain Management, and I mainly work on inventory control and warehousing.

YACINE:

Hello. Hi, everyone. I’m Yacine Rekik. I’m based in Lyon from EM Lyon Business School.

I specialise in RFID-enabled retailer operations and inventory recording accuracy.

My work focuses on understanding why inventory errors happen, how to measure them properly, and more importantly, how companies can reduce them to improve availability and customer service.

Ten years of research into predicting inventory inaccuracies

COLIN:

Thank you very much. This particular topic, inventory record inaccuracy, is linked to two of my big passions: improving on-shelf availability and reducing shelf out-of-stocks, and the other one being the problem of unknown loss and shrinkage.

So for me, this whole topic is pivotal to the understanding of those two problems, and pivotal to actually the resolution of those two problems.

We have been working with this team for 10 years. This is our 10th year anniversary. It’s been an amazing journey.

I would like now to ask Aris, on behalf of the group, and perhaps others — Christoph and Yacine, do chip in — Aris, what have you been doing for the last 10 years and what difference would a retailer notice if they were to read your work?

ARIS:

That’s a really good question. Thanks a lot, Colin.

As you mentioned, we have been working together for 10 years. I remember when Colin came over to Cardiff University and we had initial discussions about the problem and its implications for grocery and apparel retailing.

We have been doing quite a lot. First of all, through four major projects, we have partnered with literally dozens of retailers across the world — Europe, Australasia, North America and so on.

We have managed to tackle quite a few of the constituent aspects of the problem: what is IRI and how can we measure it; what are the implications of IRI for cost efficiencies and sales; how and when do we need to intervene to alleviate the problem; and what is the impact of those interventions?

Collectively over the last 10 years, it was an amazing journey for all of us. All this experience has given us hands-on insights into what causes inventory errors, why they matter, and how we can manage and reduce their impact.

Headline numbers retailers should notice

COLIN:

If you were a retailer that perhaps hadn’t heard about some of your work, what would be some standout numbers that would make them sit up and take notice?

ARIS:

First of all, people sometimes do not know that up to 70% of the inventory records are inaccurate.

Sometimes it’s difficult to convey this message. Even to people in academia, when you say that 70% of the records are inaccurate, they say, “So 70% of records are accurate.” No — it’s the other way around: they are inaccurate.

The other fascinating thing is the implications for sales. There are implications for cost efficiencies, but for sales we managed to model the whole thing ourselves.

You can show conclusively that if you fix your inventory records, sales can go up by up to 11%. These are powerful messages for retailers across the world to start taking this issue more seriously.

Predicting inventory inaccuracies to prioritise audits

COLIN:

It often feels that when you say, “How can you fix inventory record accuracy?”, people might say you need to adopt and become a Six Sigma-type retailer and go after every single root cause.

But Yacine, you’ve done some amazing work saying we can start to predict wrong inventory records. What are some of the headlines there?

YACINE:

The third number I would introduce is more than 80% or more than 85%.

The data that we have — or any company has — about their stock movement and about their previous audit is material to predict in a very accurate way the discrepancy.

So more than 85% of IRIs could be predicted. This could help prioritise which SKU to audit and what to keep for later.

COLIN:

Retailers do spend a lot of money counting. So if you can predict, there might be operational benefits — not least being able to not do some counting, but count the ones that can’t be predicted.

Organisation and capability: a maturity model for inventory record inaccuracy

COLIN:

Christoph, you’ve done some work in terms of understanding how retailers are organised against this. What have you come up with in terms of insights around building this capability?

CHRISTOPH:

What we saw is that improving inventory record accuracy can improve sales by 4 to 11%.

Retailers perform quite differently when it comes to managing inventory record inaccuracy, so we wanted to understand the drivers of this different performance.

We conducted a large interview study with 25 retail experts, and we came up with an IRI maturity model that shows which factors and dimensions retailers need to look into to make sure they perform well and keep their inventory records accurate.

For example, they need to measure IRI to see how big the problem is. They need to embed that into their organisation, make sure everybody is informed about it, make sure information is shared, and that they collaborate across departments. There are other dimensions included there as well.

COLIN:

As you say, it would seem obvious with a 4 to 11% increase in sales that you would prioritise this, yet that’s not the case.

All this research is available to download free through ECR. We’re a charity based out of Brussels and our mission is to make this knowledge available for free.

Terry Leahy once said, in the hands of consultants it would be worth billions, but through ECR it’s free. So thank you for all the great work.

Training programme: rethinking inventory management with record inaccuracy

COLIN:

Even better news is I think you’ve been able to put this 10 years’ worth of knowledge into a training programme that we’re going to be running the second year down in Cardiff.

Tell me a bit more about the training programme — who should go, what they’ll learn, and when it is.

ARIS:

Sure, absolutely — pleasure, of course.

We ran this course last year in Cardiff. Terrific because we loved it, we enjoyed it, and we hope that people enjoyed it as much as we loved putting it together.

Most inventory forecasting courses emphasise demand forecasting, supply chain forecasting, inventory replenishment, inventory optimisation, inventory management — and even, I dare say, all the textbooks in this area are built upon one important fundamental assumption: that inventory records are accurate.

All courses and books start with that assumption, sometimes explicit, sometimes implicit: the records are correct. If your input and your data are correct, then of course the focus is on optimising policies and processes.

But our research, and research of other academics and practitioners, shows that in reality the records are wrong up to 70% of the cases. This changes everything.

It means inventory management needs to be rethought from the ground up, and this is precisely what this course is about. It is the first of its kind internationally that puts inventory record inaccuracy at the centre of the conversation.

COLIN:

What was the feedback we got from the course last year?

CHRISTOPH:

The course was very well received. We received a lot of positive feedback on almost all aspects of the training.

We also received valuable suggestions for improvement — for example, to introduce further interactive tools to bring the ideas we share to life, trimming back some of the more technical detail to make it more accessible to participants who are not data scientists, and even refining the catering.

We’ll take that on board and use it in the upcoming course in 2026 to enhance the content and the overall experience.

COLIN:

That’s on December the 9th and 10th. You can find details on the ECR website.

It’s been a pleasure working with you for 10 years — absolute pleasure. I think the learnings are immense.

I feel that momentum is building behind the work as well, with many academic journals becoming more interested in what is being said.

We’ll carry on working. We’ve got more research projects up and running and we’ve got more training courses. Here’s to another great and successful 10 years. Thank you for your time.

ARIS:

Thank you very much indeed.

CHRISTOPH:

Thank you.

YACINE:

Thank you.

COLIN:

Well that’s all we have time for this week. Thank you for joining us, and thanks to all the ECR Retail Loss community for the sharing of their knowledge, their research and latest learnings to ensure that we become truly a global platform for learning.

My name is Colin Peacock. This is the Sell More and Lose Less podcast. If you liked it, please subscribe so you automatically get all future episodes.

And if you think your colleagues and friends might also like it, please ask them to follow ECR Retail Loss on LinkedIn and tune in to our next podcasts.

If you didn’t like it, then let us know and we’ll try and fix it. Either way, see you next time. Thank you again.