In-Store Analytics for Bricks and Mortar Retail

Watch our retail webinar now to hear from James Poppleton, General Manager of MUJI Australia and Alex Tweddle, Head of Sales at WingArc Australia.

You will hear how MUJI uses WingArc's in-store analytics platform and heat-mapping capability to:

  • increase in-store conversion rates from under 30% to over 50%
  • increase sales by 30%
  • measure and optimise product positioning and promotions
  • use a sales-driven staff rostering approach to achieve efficiencies and increase productivity

Get In Touch

After you submit this form, one of our representatives will be in touch to discuss your corporate needs and organise a limited period trial of our products, if suitable. We won’t share your details and you can opt out of communications from us at any time.

Webinar Transcript

Matt Armstrong, WingArc Australia: Hello, welcome and thanks for joining us everyone for today's Retail Webinar. We've got two great speakers coming up and lots of great insights for you, so in a moment I'll introduce you to James Poppleton from MUJI, who's going to talk about MUJI's experience introducing In-Store Retail Analytics across their Australian network, and then you'll hear from Alex Tweddle from WingArc who's going to talk about in-store analytics in a little bit more detail, but first just before I hand over to James just to run through a bit of housekeeping about how the webinar is going to work.

So firstly we are recording the session, so if you miss anything or you want to watch it back later, share it with your colleagues, we will send out the link over the next couple of days, and secondly we are going to run Q&A at the end of the session today so somewhere down the bottom of your Zoom window there should be a little Q&A button so if you want to ask any questions, hit that button and type them in there and we'll try and get through as many as we can before the end of the session. But with that I'll hand over to James to talk about analytics at MUJI.

James Poppleton, MUJI Australia: All right well thank you Matthew. It's a great pleasure to be here and have been invited by yourself and WingArc to speak on behalf of MUJI and our managing director Mr Takeshi Fujimoto and to give some insight into what MUJI Australia in particular and WingArc have been working on over the last sort of 12 months, with the purpose of developing a locally formatted customer analytical system. So much appreciated.

Who is MUJI?

James Poppleton, MUJI Australia: To give everybody a bit of a run through on who MUJI is, our origins begin back in 1980 in Japan. We're a Japanese retailer and we see ourselves not just as a retailer but an institution which endeavours to propose or develop the idea of a simple pleasant life. And to give a bit of an insight into what that is, the vision is on the background of living a compact life. It doesn't so much involve living without material goods or living in a small way. It's about achieving a simple pleasant life that embodies the personality of the individual and the lives in which that particular individual lives and offering versatile products that are designed to provide what is necessary, so that's the idea behind MUJI.

We currently have about or over 900 stores globally and we have a range of products that are quite vast but locally we run with garments, so men's, women's, kids and through the accessories range and travel, homewares, stationery, health and beauty, food... With food it's drink at this stage but overseas there's a big range of food that we have and obviously much, much, more.

MUJI established... we established our first store back in 2013 at the Chadstone shopping centre, and we've now grown that to five stores, expanding into New South Wales where we have two stores, ACT and then in Victoria we have our big footprint Chadstone store now, as well as our CBD store at the Emporium.

What were our Analytical Goals?

James Poppleton, MUJI Australia: So when, or about 12 months ago myself and Fujimoto-san our Managing Director, we started to consider what's the best way to understand or go about understanding who our customers are. So the purpose of having an analytical system was to have an understanding, so to have better visibility and transparency on our customers' behaviours because at that time we didn't really have anything or any processes in place or data that would enable us to understand and identify opportunities of our customer base, which would in turn allow us to grow our customer and brand awareness.

So, the customer trends were definitely a main principle behind what we were looking for. So we needed a system or a platform that was simple to work with, that would gauge those relevant trends, and in the retail industry as many know things move quite quickly so tomorrow is different from today and definitely in the current market under the COVID-19 scenario there's not a better time than now to understand easily and to be able to make quick actions in regards to customer trends, and what they're doing.

So we were chasing a reporting system that enabled us to identify those opportunities efficiently and effectively and to be able to develop a wide range of strategies that would enable quick, decisive systems and actions that would in turn improve our sales. And then another key point the last sort of main key point was coming from the store operations.

So we desired to enhance the means of the store managers and our leadership teams at the stores to construct their rostering systems, their own internal analysis of what's going on to ensure that we had the right team at the right time ingredient to help improve and maximize the customer experience or the experience for our customers when they came into our stores.

What were our Analytical Goals?

James Poppleton, MUJI Australia: So to delve a little bit further into the detail of what we were looking for and which ultimately led to us building our relationship with WingArc: We wanted to understand the number of passers-by so the number of people actually walking past our stores. At the moment we're just in shopping centres so it was important for us to understand that, but it also enabled us to improve our communication with the relevant landlords around specific discussions or conversation on how to develop the customer traffic numbers in and around the locations we were positioned in their centres. We wanted to know the number of entrants, so the capture rate, so the percentage of people that are actually walking past and then coming into our stores. We wanted to understand conversion rates so entrants into actual transactions. We were sort of driving a little blind for or prior to this, so there was a lot of guesswork involved but having that visibility has certainly helped us, which in turn enabled us... we wanted to build strategies around improving our units per transaction and average transaction value.

We also wanted to have a system, the heat mapping operation, where we could understand where our customers were going when they entered the store, which in turn is designed to improve the strategies and the adjustments that we may need to take in accordance with the strategies that we have in place, and that goes beyond just supporting the operations team. We wanted to be able to build our strategies for the visual merchandising teams, marketing, and our buying teams as such. Another element is, we wanted to understand the dwell time of our customers. So how long are they spending in the stores?
How can we make it more interesting for them, and so forth to develop that sort of KPI.

And then as previously mentioned in the last slide to ensure that we've got that right number of team members on at the right time so as to maximize the relevant points above.

And then lastly something that we're sort of leading towards and working with WingArc on is to understand the demographics, which in turn will help us beyond just store operations but also into the business development side of our future strategies, whether it be sort of 5, 10, 15 years down the track.

What attracted MUJI to WingArc?

James Poppleton, MUJI Australia: So, what attracted us to WingArc? Well, look, they're an established operation, they've got a strong platform in Japan already so they're working with some very strong brands globally, so therefore we definitely had a lot of trust in the WingArc brand before we delved into too much longevity of our relationship with them.

Some of the things that definitely stood out for us when we opened up the conversations with them: they were open to understanding our needs and desires, and were willing to listen to what we wanted. They weren't telling us, "all right, this is what we've got, this is what you're going to need". It was more sort of understanding what our needs were, and then building the platforms and the processes around what MUJI needed. They had a strong desire beyond that to also understand the MUJI business, so which was or is something very important culturally to MUJI, definitely MUJI Australia that our relationships with partners or partnering up with other institutions and organizations is that people take the time to understand who and what MUJI is and WingArc definitely did that and that certainly helped them build and understand what we're trying to do. And then lastly, look they're a great bunch of people, all our dealings with them, everybody involved in the WingArc team are very friendly, nice and supportive and willing to work with us, so it's been great.

What have been the Outcomes?

James Poppleton, MUJI Australia: So: what have been the outcomes?

So now there's a lot more transparency for us surrounding customer trends. We have a better insight into the opportunities to develop and improve on the sales. So this has definitely led to some really good quick outcomes and improvements for us. We understand how our customers move in and around the shop. One of our stores has two levels, so we've been able to have a better understanding as to how the customers are coming in and whether they're staying on ground level and going upstairs or how they're moving, so as a result this has enabled us to adjust our strategy and look to improve both the KPIs and the sales and the customer experience.

And as I've mentioned, it's improved the sales, definitely improved conversions, since the start of the year, with the entrants into customers, and there's been some really outstanding results in some of our stores around that, which has been good. And the visual merchandising and MD teams have been able to make clearer and more decisive actions towards their strategic operations.

The BIG question, "has it helped, and how quickly"?

James Poppleton, MUJI Australia: So the big question is: has this helped, and how quickly? So I was thinking when I was sort of putting this together, what would be the best example to show everybody, and the graph on the left side that you can see... this is a section of data that is from one of our stores back in April - start of April - when the analysis of the conversion rates went live into all the stores, and you can see from what you're looking at that within three weeks we had an increase from an average of less than 30% in conversion rate to an average of above 50%, so just having that pure transparency for the store manager to see okay well this is where you're at, they were able to make decisions, actions sharing that information, and had that very quick and immediate improvement in that single KPI.

So if I looked at it when I looked at it based on the average transaction value at the time this arguably resulted in a 30% uplift in sales for the store. So when that first wave of COVID came through, you know this played an important role in ensuring that we were able to improve our sales and definitely the store manager herself took some very good internal actions, which supported that and they really haven't looked back, and they've been able to sustain that high conversion rate since since the middle of April.

Heat Maps

James Poppleton, MUJI Australia: So the other important or main thing that we've been working on to implement has been the heat mapping, so this has definitely been important or is an important aspect to any retail business, basically so that you understand where your customers are going in your store, where your high traffic locations are versus your low traffic locations and helps in making decisions to actually put your products in the right spot, and to make the most of the layout within the stores, because sometimes certainly from our own experience you might think a location is working well for you but in actual fact it's not.

So having this really visible clear understanding definitely helps and as I've touched on a few times already this has been a very important part for the visual merchandising team to understand the effectiveness of their actions and their strategies within the stores. It's enabled us to assess the impact on in-store promotions, so as I mentioned both good and not so good! We can monitor the progress of visitors to specific store locations, so the store managers can review and assess specific setups and change as necessary, and then to optimize product positioning to ensure or to deliver on a better return of investment. So heat mapping for us is somewhat a more recent introduction into the stores since our journey began with WingArc, but yeah there have definitely been some or probably about half a dozen or so key changes within stores that we've already made and understood based on this information. So highly recommend any retailer being able to have access to this type of data.

What the Future Holds: Sales Driven Rostering

James Poppleton, MUJI Australia: So there are many things that we're working on with WingArc at the moment. You know we've got a long list of them, but one of the key things that we're pushing through with them is what we call the sales driven rostering. So based on sales, we needed a system or we need a system that will enable the store managers and leadership team to build the correct roster, so around that right team right time operation again.

At the moment we somewhat have, when historically we've had a very manual time-consuming piece of, I guess, Excel spreadsheets that eats into the productivity a lot of our store managers, so we're developing a system on the dashboard to tie in with this "right team, right time" operation on what we need on a day-to-day basis, so as to improve that customer experience, the productivity, and to reduce that amount of time that our store managers are spending on putting the roster together. So we're looking forward... MUJI is definitely looking forward to developing this and we can certainly see a major uplift in sales and productivity from this single operation in itself.

And yes, so thank you Matthew. I hope everybody got something out of that but definitely there's a long road for MUJI and WingArc, and a long journey to take place with many things to develop and under the umbrella of what you guys are putting together so thank you for this opportunity for speaking.

Matt Armstrong, WingArc Australia: Thanks very much James. So yeah, it's really great to hear that you're making great progress with the solution. It seems like there's some really good developments coming on both sides through the solution.

So our second speaker today is Alex from WingArc and he's going to talk in a little bit more detail about how the solution actually works and just going to go into that next level of detail, so Alex over to you.

Common Retail Challenges

Alex Tweddle, WingArc Australia: Thanks Matthew and thanks very much James for a great explanation there of exactly what's been happening at MUJI over the the last 12 months or so. I just thought I'd take a step back and just talk about what are the typical retail challenges that we're talking to customers about, and what are the things that we can help with, and I think James has touched on many of these. First of all: how is it possible to assess exactly what's happening in store from a marketing effectiveness perspective. What's happening from a store layout perspective where... how do we know where are the best places to place particular products?

I think typically for organizations that don't have a solution such as this, there's a lot of anecdotal discussion but not a lot of data and evidence to back up how decisions are made. Another typical scenario we get is the, I guess, the comparison with the online world. So typically e-commerce providers have a huge amount of information on everybody that visits their site and they get that complete sort of journey view of what customers are doing, whether they're just browsing, how long they're on the site for, what their conversion rates are, how much they're spending. But that's much more difficult for retailers without this kind of solution. And again, James touched on it around staff and rostering, which typically is a very time consuming process and maybe there's some educated guesswork often involved in that, but if by making the decisions and the effective rostering much more evidence-based, it provides a lot more confidence in that you're going to have the right people on the floor at the right times.

Video Analytics and Point of Sale Integration

Alex Tweddle, WingArc Australia: So the way that WingArc works is really integrating a number of systems to give retailers a complete all-round view of what's happening in their stores, and key to this is linking to a digital video analytics platform, so typically this would be your security system, and we are able to track movements of people through that process. And then also by integrating with point-of-sale systems provide that all-round view of what's happening within the store.

In addition to that, and again going back to the staffing side of things, by integrating with HR or or rostering systems, it really becomes the hub of the store to understand exactly what's happening from customers, from people who are passing by the store, to what's actually happening in terms of the purchasing process within the store.

Heatmap. Sales Performance. Rostering.

Alex Tweddle, WingArc Australia: So moving on, what that actually means is the development of heat maps, the provision of dashboards and reports that will give you a much more detailed breakdown of sales performance, and as we've discussed the whole rostering process.

Compare Performance Across Store Network

Alex Tweddle, WingArc Australia: So just to give you a, I guess, a look at the the sort of insights that James has touched on in his presentation, he's covered very nicely the benefits and what that or certainly what MUJI were looking for, and the benefits that those have gleaned. It's interesting just to take a look at what does that actually look like in practice? I think one thing to say is that the platform is very dynamic, so it gives you the opportunity to see what's happening in individual stores, it gives you the opportunity to see what's happening across all stores. There's opportunities to look at comparisons between stores, so by different time frames. There's a heap of insights that you can generate that wouldn't normally be readily available to you. Looking at the the left-hand side of this particular slide, this is looking at an all-round view of all stores for a particular period of time so you can see that you can look at what the overall sales values are, the average transaction value (or ATV).

James talked about passers-by and entrants, so by understanding the number of people that are passing by and the number of entrants that are actually coming into the store, you get that capture rate and then again by looking at the number of unique purchasers you get that conversion rate, that it's often very difficult for retailers to get. And then by understanding the number of unique items that are purchased on each transaction, you just get that real understanding of what's actually happening within each of those stores.

The comparison on the right hand side shows what's happening in a particular store. So this one is a store in Fountain Gate, when you look at comparing that against all your other stores you can very quickly see okay average transaction value is significantly higher, but I'm actually getting a much lower capture rate, so that's a lower number of customers are coming into the store than I'm seeing elsewhere, and also getting a lower conversion rate, so from a management perspective if there are things that you can put in place to encourage more people to... if there are things that are happening in other stores, maybe that are happening from a staffing perspective or a merchandising perspective, there's definite opportunity to grow the revenue in that store through an increase in capture rate and conversion rate.

Compare Performance Across Store Network

Alex Tweddle, WingArc Australia: We can also look at what's happening in terms of daily and weekly trends. What we're looking at on the left hand side is about four or five weeks' worth of data on a daily basis. The yellow line is showing the actual value of sales, the light blue line is showing the number of people that are passing by the store, the slightly darker blue line shows the number of entrants, so the people that are actually coming into the store, and then the dark blue line is showing the number of people that are actually transacting.

So you very quickly get that all-round view of what's happening on a cyclical nature and again when I talk about dynamic, it's very easy to sort of vary the time frames that you're looking at and narrow down to some of the detail that you may like to look at. In that daily trends section, interestingly, in about the fourth week it looks like sales are significantly down even though it's probably the, looks like it's the busiest time or busiest week of the period in view there. There might be something that happened around that store around that time to result in that, but from a management perspective it gives you the ability to understand what's happening in store, put measures in place to make improvements.

On the weekly side, so this is looking at around about a 20-week view, similar sorts of data in terms of sales and passers-by and entrants and so on, but it can give you that much wider view of what's happening, maybe more from a seasonal perspective, and help you make plans for future events or future promotions and identifying where are the gaps, where do we have opportunities to maybe put promotions in place to maximize sales when at times maybe we see our sales traditionally dipping.

Use Data to Optimise Rostering

Alex Tweddle, WingArc Australia: Again, James talked around staff rostering. So what the platform enables you to do is to really look at breakdown times of day to understand what's the relationship between the number of store entrants versus the number of sales that you're making, and then the overall value of those sales, and help you identify other potentially missed opportunities to make sales, and if there are, take that into account from a rostering perspective, so what you're looking for there really is: is there a disconnect between sales and number of customers that are coming in?

And again from rather than just looking at charts, it's very much possible to break down historical data on a much more detailed view. So the chart on the right is looking at breaking customer numbers down right down to hourly time chunks, mirroring that across to sales data gives you that opportunity to understand exactly what's going on and develop more of an understanding about how you maximize the usage of your staff across both days and times of the week.

Use Data to Optimise Opening Hours

Alex Tweddle, WingArc Australia: I guess in a similar way by understanding exactly what's happening around your store, are there changes in consumer behaviour that maybe makes you start thinking about changing opening hours? I think as we hopefully come out of - certainly in Victoria - come out of the current restrictions, it's very likely that shopper behaviour is probably going to be uncertain for some time.

I think we've seen a clear indication that there's a big appetite for people to get back to shopping in the last few days, so I think getting an understanding of different times of day that maybe there's opportunities to extend opening times depending on the external factors, and that could obviously be around the external factors related to a shopping centre location, maybe what's happening within CBDs based on, maybe people coming back to work over the next 6 to 12 months, and also in our high streets. So really there's a lot of dependencies around the actual physical location of stores.

Comprehensive Reporting

Alex Tweddle, WingArc Australia: I guess from a high-level reporting view, it's great to have the dashboards and the charts that give you that very quick and dynamic understanding of what's happening, and help with that decision making. Very detailed reports can also be accessed through the platform, broken down by different time periods and in many circumstances it's necessary for reports to go to named, approved groups within the organization.

So whether that's on a daily or weekly or monthly basis, there's a lot of very rich data that can help with that, enhance the overall reporting of the business and just get that genuine understanding of exactly what is going on within stores.

Store Heatmaps

Alex Tweddle, WingArc Australia: Finally, looking at heat maps and I think James has probably done, well he's done a fabulous job on explaining what heat maps do, and how they've enhanced what MUJI is doing, and how they're understanding what is going on in-store. So there's probably not a lot I can add to James's excellent explanation, but really what we're getting at is understanding exactly what's happening once customers come into store, identifying the areas where customers are going.

That dynamic piece is very important here, so we can look at very specific time periods. We could look at weekdays, weekends, we can look at certain times of the day to understand what's happening in those time periods. I think an important element, and probably more what you see on the right-hand side, is doing direct comparisons between different time periods. So it could be as simple as looking at what are the differences between mornings and afternoons? Are people going to different areas at different parts of the day? It could be weekends versus weekdays, it could be the Christmas period versus other periods of the year. Really to give you that much better understanding both from a merchandising point of view and positioning point of view, or marketing/promotion point of view, as well as understanding maybe from a staffing perspective where you should be focusing your staffing efforts at different times of the day.

So that's just I suppose an overall summary of what you actually see within the platform, and I hope that backs up some of the excellent examples and results that James explained in his presentation about MUJI. Back to you Matt.


Matt Armstrong, WingArc Australia: Thanks, Alex. So with that I'll head over to the Q&A. So as I said at the start if you do have any questions, find the Q&A button on your Zoom window and type them in, and we'll try and get to as many as we can. We've just got a couple so far, so first one I think probably a question for James: so are there any examples where the data from the analytics or the heat mapping has really surprised you? Something that showed you something that you wouldn't have expected?

James Poppleton, MUJI Australia: So the best example on that is one of our stores in Sydney in regards to the heat mapping somewhat again. There was an area on one of the main spans or fixtures that we'd actually anticipated or expected to be a high-traffic area, just based on its positioning. Now following the introduction of heat mapping, we were able to analyse or see this I guess you'd call it a "dead zone" that we actually thought was high traffic.

In turn, we could see that where we thought the customers were going, they're actually going in a different location so as an affect, or as a decision, we were able to change the positioning of particular key products around, which in turn had a positive result, so that is definitely a strong benefit or an example as to how the heat mapping has assisted.

Matt Armstrong, WingArc Australia: Great. So next question, I think it's probably an Alex question: how accurate is the data?

Alex Tweddle, WingArc Australia: Yeah, so through the development process, we've looked to run tests that we can validate the data against, and we look to sort of be working in that sort of 95-ish percent range. There are clearly, because of the way that we track movements and images of people moving around the stores, there can be some situations where there are slight inaccuracies in terms of maybe a large group of people huddled together, coming into the store at once, but typically there's an understanding that if that particular situation happens at particular times of the day or the week, we can work around mitigating those, that particular factor, but I think James would back up a high degree of accuracy based on the work that we've done with MUJI.

Matt Armstrong, WingArc Australia: Yeah, we've certainly done a lot of manual checking of some of the data to sort of validate that it is pretty accurate. Next question has come through: Is there a use case for department store counters, especially monitoring traffic in department stores? Alex do you want to take that one?

Alex Tweddle, WingArc Australia: Absolutely, yeah, it's absolutely applicable to department stores, so I think what we're working on is, as long as there's camera coverage of the store, then, of the entire store, then we can work with that, and also monitoring cameras for the entrances and exits to manage the number of people coming in and out.

I think working on multi levels is absolutely possible as well. James mentioned in his presentation, they've got one of their stores in Sydney that's on multiple levels, and the platform can manage that as well, so I think from a department's store perspective, given that you've got different brands with their own spaces, I can see some particular benefits in understanding where foot traffic is going within those stores, and some insights that you can generate both for your own benefit, but also for the brands that are in the store as well.

Matt Armstrong, WingArc Australia: Sure. So one more question, I think this is probably a James one: how has customer behaviour changed since the pandemic, where your stores have been able to stay open? Have you seen any particular trends that you can share with us?

James Poppleton, MUJI Australia: Yeah, look, one key trend which has shown up and we've been able to gauge using the Units per Transaction is, perhaps earlier in the year, when it all first sort of started to come into play, and probably ran for about three or four months, was we could see customers when they came into the shop the conversion rates were good, but the UPT increased, because what we were able to gauge was customers, they were coming in, they'd shop a whole lot, they'd buy a whole lot of things like going to a supermarket, and then go home, because they didn't want to come too regularly back into the CBD.

So that is definitely a change in customer trends than somewhat sort of last year, and prior to the Coronavirus. And this in turn has meant we've changed our trajectory of strategy as well. So trying to focus around enabling the customers to pick up more companion items that related to our base items, as well. So there in turn it had a further flow on effect with our sales so that's sort of one key trend that we've seen, Matthew.

Matt Armstrong, WingArc Australia: Fantastic. I think, last call for questions, but I think we're just about out of questions, so that pretty much brings us to the end of... Oh one late breaking question just popped in! There's a question, I guess it's an Alex question, so: do you know any retail clients who use the video analytics for loss prevention purposes, such as face recognition or theft deterrence, in addition to sales analytics?

Alex Tweddle, WingArc Australia: Not specifically face recognition, or not that I'm aware of for theft deterrence. The only one that I'm aware of, and I saw recently was, and it was in the press: 7-Eleven had introduced face recognition not at all stores but in quite a lot, and there's a, I think there's a notice on the entrances to some of their stores, but that's the only one I'm aware of. I think one of the things we've spoken about is, looking at for example heat mapping, that what that can help generate for retailers is actually highlight maybe where there are some blind spot within stores, so if there are areas where there's no sort of reading of foot traffic, there's a, potentially, a clear indication that there are some blind spots within stores that need to be addressed, but yeah that, in terms of facial recognition that's the only example that I've come across.

Matt Armstrong, WingArc Australia: Sure. Thanks very much. All right, I think that we are out of questions now, so thanks everybody for joining us on the session today. I hope you found it useful. If you are looking for more information, head over to our website which is You can get in touch with us through through the website, or alternatively reply to any of the emails that you received from us and we'll try to help you with your Retail Analytics requirements. Thanks very much for your time everyone.

Alex Tweddle, WingArc Australia: Thank you.
James Poppleton, MUJI Australia: Thank you.