Where Retail meets Tech

May 2, 2019

How We Took Our Image Recognition App for Field Reps to the Next Level

Written by Cécile Garrett

When implementing a Retail Execution Excellence program, Consumer Goods manufacturers often face two main challenges internally 

  1. Empowering teams with new tools to achieve higher results
  2. Driving change management to ensure the full adoption of these new tools 

If brands want to reach perfect execution and see a return on their investment, they cannot dissociate one from the other. But introducing new tools often leads to more procedures and internal processes, which ultimately creates more complexity for employees and drives engagement and productivity down (BCG analysis, 2016). 

For the past two years, we made addressing that "double challenge" a key company objective at Planorama. That’s why we’re excited to introduce you to our new mobile app for field forces 

Meet the People at the Origin of Our New Mobile App for Field Forces 

In January 2017, Tamás Lupkovics, Product Manager at Planorama and lead on the project, teamed up with Othilie Nicod, our Chief Marketing Officer, to revolutionize our mobile app. The goal: make it fit our users’ operational activities and daily routines completely.  


Last week, we sat down and talked about the project in more details. Here's what they told me.

Don't have time to read the entire interview? Why not skim through?

  1. How did the project start?
  2. What was the scope of the project?
  3. How did we choose the clients with whom we worked on this project?
  4. What did we discover?
  5. What developments did we make after that?
  6. What are the benefits of our new mobile app for store checks?

PlanoCheck : Collect shelf data in real time and realize the Perfect Store 

How did this project start? 

Othilie: First, what’s important to understand is that the revamp of our mobile app for field sales was part of a global project including our web platform. Initially, we wanted to conduct this project because our apps were a bit behind the latest UI/UX standards. And since both are linked together to connect field and central teams, it didn’t make sense to single one of them out for this project. At that time, we also had a choice: give a facelift to our apps and make them look nicer or make them even more valuable to our end-users. We decided to go for the second option. At Planorama, we’ve always built our products together with our clients, especially at the very beginning. We created our first product with L’Oréal and Fleury Michon, so involving clients in product development is in our company culture. But as we scaled, it became more and more difficult to do so. You know, when you’re a company of five working with one or two clients who have about ten users, it's easy to take the necessary time to understand, analyze, and include their feedback in your roadmap. But when you grow to have hundreds of clients and thousands of users, collecting feedback broadly and proactively, taking into consideration different markets, different types of users, etc. is a whole other story. That's why we set it aside for a while, and why revamping our apps was an opportunity to get back to our culture of innovating with customers.  

Tamás: I remember we had this fascinating discussion before we started this project. We weighed the pros and cons of reworking the apps’ interfaces versus following a specific UI/UX methodology to get an in-depth understanding of our users and redesign our apps accordingly. Getting everyone’s buy-in on this ambitious project wasn’t easy at first for the reasons you explained, Othilie. It represented a much bigger investment, too. But from the moment we agreed onwards, we saw the benefits of the study. It helped us bring customers back into the heart of our innovation process. Now, everything we do starts and ends with our users, from understanding their pain points to detecting their needs and the product opportunity behind them.  
Othilie: Doing this study enabled us to reconnect with an essential part of our DNA and become even more customer-centric. It helped us develop our capabilities to involve clients and users in product development stages at a larger scale.  


Q: What was the scope of the project? 


OthilieSince we had decided to revamp both our mobile application and our web platform, the scope of the project was quite rich in terms of user profile. We had to study field forces to assess the fit of our mobile app as well as the central teams using our desktop application. We also needed to cover different use cases to get a full view of our users' needs across markets. In total, we had 65 participants coming from 14 client brands spread around 11 countries.  

Tamás: We wanted to conduct an ethnographic study to observe and interview participants as they were using our apps. So, we contracted an agency who immersed themselves into our users’ daily routines, shadowing both central teams and sales reps for a total of 150 hours. During this phase, they collected 350 pictures and conducted 22 interviews to document and understand the pains and gains customers associated with our apps. It was a crucial step for us to get an in-depth view of our different types of users and assess our applications’ fit to their needs. 

Othilie: After that, we built two wireframes, which are structural representations of what the new applications could look like in terms of content and functionality. What happened was that participants came to the agency’s lab and started playing around with these wireframes, giving live feedback. From there was the start of a long iteration process where we would update the wireframes as we received new feedback. Then, came the online tests.  

Tamás: Yes. Users would log into a website and see mockups of the mobile app or web platform. Here again, they would test the apps, send comments to the agency who would iterate the mockups, etc., until we reached a satisfying result. 


Q: How did we choose the clients with whom we worked on this project? 

Tamás: We took multiple criteria into account.  One was our relationship with clients and their engagement level. The other one was their project scope as we needed to cover different use cases. For example, we wanted to have clients who were using the solution as a retail execution tool, with live reports in the mobile application, frequent checks being submitted, etc. But we also wanted to review another use case to assess the needs of clients who use the product for data collection purposes, so at rather low frequency and without live reports.  

Othilie: At the beginning, we also hand-picked some markets strategic importance to us. We could not cover all of them due to geographical constraints, but we did our best to include as many strategic markets as possible. We also did that to have insights on different retail environments, which can vary quite drastically according to markets, so we wanted to gain a more profound knowledge on that.  


Brand-New Mobile & Desktop Apps Driven by UI & UX Discoveries  

Q: What did we discover?  

Tamás: We discovered a few new things, but I found the study was more impactful in how it helped us structure our clients’ pain points and highlight the opportunities we could build from that. Before, we had a lot of ideas on how we could improve our apps, but we didn’t really know which block to prioritize to benefit most users at the same time. After the study, what emerged were two topics where we needed immediate improvement.  
The first one was around picture-taking and what we call “picture compliance” (or "picture quality," in familiar terms). Whenever users take pictures in-stores, they must follow specific guidelines which we wrote to optimize product recognition. But sometimes, shelf photos don’t align with these guidelines. It means the system rejects the picture and notifies the user that it’s not analyzable due to poor quality. Before, we didn't provide users with any clue as to why the system rejected their photo, which was very frustrating for them. So, the first improvement we made was to develop features to help field users capture better pictures with the mobile app.  
The second one was around transparency. The study revealed that clients assimilated our Image Recognition system to a black box, which was a source of stress for them. They were wondering, you know, "What's happening to my pictures after I upload them? When will I have my report back? Why doesn't the system recognize my products?", that sort of things. 


Q: What developments did we make after that?   

Tamás: The main one, which pushed us to take this UI/UX approach further, is what we call the "puzzle mode." In markets dominated by convenience stores, there isn't much space in store aisles. So, it's almost impossible to take one picture per section, which is the number one rule for picture taking. We worked closely with clients from these markets and got feedback that the original picture-taking feature we had built in the mobile app was not really working in their retail environment. In truth, it had been designed for markets where modern trade is “the rule” and convenience, the exception. So of course, it wasn't adapted to the traditional trade channel. It was complicated to find a solution that worked for all users, both technically and in all store types, so we also ran a mini UI/UX project with selected clients to get to this puzzle mode feature. 

Me: I had a question about that. In the puzzle mode, the user first takes a picture of a piece of his or her category section, then moves down to capture the second pieceand up and down, and on so on until the puzzle is whole.  

Q: How does the user make sure that no product is missing between the pictures during this process?  

Tamás: There is an overlap assistance. So, in the live camera of our mobile app, we show the side or the top of the previous picture to help users align their photos. 


Q: What are the benefits of our new mobile app for store checks?  

Tamás: In the MVP, which is the Minimum Viable Product or the first version of the new mobile app, the main benefit is definitely the improved feedback we now give to our users on picture quality. We provide users with this feature for live projects, those with a 5- to 10-minute SLA, to provide instant feedback to users. It’s really a matter of a minute or two before the system shows an illustrative description of what the issue is. It helps users correct it before submitting their category check. But we also made sure that they could go back and reopen their check to fix low-quality pictures even after submission. It helps clients get the best quality of shelf data from the least number of pictures or in the least amount of time. 


Tim Wynia_Unilever_Planorama-new-mobile-app-for-field-forces“We have been using Planorama’s Digital Measurement solution since 2014. Earlier this year, our entire field force of over 100 users moved to the new app. This transition was extremely smooth as it did not impede on their daily work. On the contrary, the new mobile app has become a real performance enhancer. Now, sales reps can not only execute their in-store tasks more efficiently but also track outlet and category audit progress as they aim to reach a better performance. Ultimately, the app helps them focus their efforts on adding value to Unilever’s customers, spending more time conversing with them, and obtaining a superior in-store shelf for Unilever as well as Unilever’s customers! So, it definitely impacts the sales performance too."  Tim Wynia, Perfect Store Coordinator at Unilever Netherlands 


If you’d like to know more, contact our sales team by clicking below: 


PlanoCheck : Collect shelf data in real time and realize the Perfect Store


Topics: image recognition, retail execution, planorama