Target Audience




Finding new opportunities creates exciting new challenges. There's an increasing number of food delivery services appearing in Phnom Penh, yet most (if not all) suffer from various customer pain points such as extensive delays, cumbersome app designs, lost drivers, and a significant number of headache-inducing errors.

RMA Group is one of the largest diversified conglomerates in Asia, with over 7,000 employees and many business units. They have a food division with over 100 stores across Singapore, Cambodia, Laos, and Myanmar and employs over 2,000 people with brands like The Pizza Company, Krispy Kreme, and Swensen's DQ, and others.

When RMA Group approached us to examine how the market was changing across the region, we instantly knew we had a fantastic opportunity to take some of our daily frustration with existing solutions and create something better.

The Idea.

EFG had previously integrated Microsoft Dynamics and LSNAV with one of our other customers, Votiva, and so they then came to Mäd with one key question in mind:

Based on the changes in consumer behavior regarding food delivery that has already been in more developed markets, what are the critical threats and opportunities for EFG Group in the coming years?

Our first engagement was a Design Sprint. Various senior team members from across the region, representing multiple functional departments such as finance, operations, and marketing, flew to our office for a two-day workshop with numerous planned sessions to understand the implications.

We were fortunate to be armed with real-world data, gained from various partners of Vietnam and Singapore, to see the specific impact that aggregators such as GrabFood and FoodPanda had on local delivery players in new markets.

We noticed that while these international aggregators tend to grow the overall market, they also tend to take a disproportionate market share. This is due to better technology use, a smooth user experience, well-defined customer support processes, and large marketing budgets.

Knowing that Cambodia, Myanmar, and Laos are not immune to international players coming in (which they did), we had to understand the core impact on EFG's delivery business made thousands of deliveries per day.

The core problem is that the % of the commission that the aggregators take from the delivery business essentially takes a large amount of the net profit on the food sold. We determined that having an alternative solution for the end-customer to order their favorite brands directly from EFG was required. However, we needed the right value propositions to ensure that the project was successful.

The key value propositions that we developed built on the unique end-to-end vertical integration of EFG, including full-time drivers, bikes, equipment, kitchens, and even raw materials delivered to the kitchens. We wanted something that would be incredibly difficult, if not impossible, to copy with the typical aggregator business model.

  1. Multiple Brand Orders. By building  CPUs (Central Production Units), where multiple brands are prepared in close vicinity, this enables the business model to offer delivery from numerous brands within the same order. This is impractical if you have ambulating drivers with disparate locations that need to be picked up sequentially.
  2. Super Fast Delivery. The other core idea was to offer super fast delivery (typically 30 minutes or less). HungryApp achieved this by placing CPUs close to critical residential and business areas and having drivers always on standby at the CPU; it further cut the time required for delivery.
  3. Free Delivery. The decision to offer always-free delivery for end customers and absorb this as a business cost was controversial. However, the overall higher profit margins with the vertical integration and the various cost savings made this feasible.
App icons.

The critical question that we asked ourselves when developing these USPs (Unique Selling Points) is what wasn't going to change in the future?

And if you think about it, customers in the future are not going to wish for a smaller product selection, slower delivery, or more expensive delivery. So, these things are always going to be valuable and appreciated by end customers. So we decided to invest heavily in the operational and technological excellence required to make this a reality.

We did not decide this by locking ourselves as a group in a meeting room for a few days. We interviewed a large number of stakeholders across various functional units of EFG and existing and potential customers. We even developed a prototype of the future application to test with users to understand what would be useful to them.

Having this user-centric approach to developing a new business and product is incredibly important because the most significant strategic mistake any company can make is to spend time and money building something that nobody is going to use.

An Integrated Approach.

One of the key advantages of working with Mäd is that we offer our clients a fully integrated approach. We can work on the high-level strategic elements while simultaneously working on the nitty-gritty executional details. Typically, that often requires many different partners and causes a large disconnect between strategy and execution.

At Mäd, we don't just sell pretty decks and reports; we execute and take our customer's ideas to the market.

EFG recognized this and then engaged Mäd for numerous simultaneous services, including:

  • Brand Development. Building a brand that would be instantly recognizable and loved while easy to pronounce across multiple markets regionally.
  • Project Management. We are ensuring that all the various moving parts of the project are moving forwards at all times.
  • Product Management. Defining the precise features and user experience of the product and prioritizing and managing the development.
  • Development. Building the entire backend and core functionality of the application, ensuring a scalable solution that can work across the whole region.
  • Strategy. Working with the senior management at EFG to ensure that there is a clear and well-defined strategy and communicated to each team member.
  • Vendor Selection. For some of the work that had to be handled by subcontractors, we ran transparent vendor selections to ensure that we found the best possible partners.

The core objective was to launch an MVP (Minimum Viable Product) that could be launched to the market to gather operational experience, customer feedback, and real-world data on some of the hypothesis that we built in the Design Sprint.

The Brand.

When constructing any brand, it is vital to convey the brand values and personality through communications to ensure an authentic and relatable image that will resonate with customers.

Primary Logo.

After defining the core brand essence, we aided the EFG team in constructing a distinct persona for HungryApp™, which helped build the foundation of media assets and printed materials.

Having defined the Brand Pyramid, we developed the project with a clear vision by creating a brand book to help guide all current and future team members in supporting the HungryApp™ mission.

Our brand books cover many essential topics, such as social media guidelines, typefaces, co-branding rules, and color schemes. The best brands create a core identity and can create customer awareness from their known style without plastering their logos everywhere. Think of Coca-Cola as a leading example. Their messaging and color scheme can be a hint of their product without displaying a single drink or logo in adverts.

Brand Guidelines.

Building an MVP.

When we started with the HungryApp project, we had one clear goal in mind: launching and getting real orders and customers within six months. This was important because the real test of any application is once it hits the market and is used by real people.

So, we had to identify what was the smallest version of HungryApp that we could build and successfully ship to customers. Building and shipping an MVP is all about compromises. There will also be plenty of things that you want to include in a product, and being able to negotiate with lots of stakeholders and being able to stand firm and saying "no" is an important skill to curate.

Some notable missing features on launch where even basics like order history. While this is available in absolutely all food ordering application, we realized that it was actually not a show stopper when you are launching a new app. After all, most customers won't have lots and lots of orders right away and so this is something that can be built later once we have a larger user base with a longer history of order.

That being said, we still needed to build with the idea in mind that this application was eventually going to scale to thousands of orders and customers, and even across several different markets in the region. With that in mind, we had to ensure to build a scalable technology infrastructure from the start, and this is why we decided to use an open source eCommerce system called WooCommerce to power the backend of the application. This enabled us to focus on areas that were critical and specific to HungryApp without having to reinvent the wheel for every single requirement.

Additionally, we also were very strict in ensuring that all APIs were fully documented so that we had an API-first approach. This means that we had standalone documentation with code examples, and that all communication between the backend and mobile teams were primarily done via documentation, ensuring that this was kept up to date and stayed relevant.

Launching the MVP.

Working with a defined catchment area, we needed to ensure that the messaging of HungryApp™ was reaching the right demographic. This can pose challenges to advertisers as social media advertising channels often have restrictions on areas, although it is hoped that in the near future we'll be able to draw boundaries as is possible in Snapchat Geolocation Filters.

Working purely with radius locations can prove challenging for an exact catchment area.

For Facebook, the minimum radius for advertising is 1km around a particular map pin, we creatively used multiple pins to recreate the catchment area and maximize our messaging.  

Getting the word out about HungryApp™ was our first goal, then encouraging downloads and orders the ultimate aim. To ensure our adverts were efficient, we ran extensive testing throughout the soft launch period, creating highly optimized adverts as an end result.  

Further to this, we were able to monitor competitor adverts through Facebook's transparency tools and Google Search Ad auction results. It's important to know the competition, firstly to compete but secondly to spot new untapped opportunities.

Our Dynamic Display Ads allowed for a more immersive experience in social media advertising.

Starting with Facebook, we utilized various styles of adverts to maximize impact, alongside the norm, we launched an 'Instant Experience' advert- reaching over 1,000,000 impressions with a cost of less than a cent per result.

Testing Matters.

To ensure our adverts ran effectively, we tested multiple creative visuals and copy, and continued to optimize until the engagement was high and the cost was low.


To aid our advertising campaigns, we created a range of unique coupon codes and associated them with our various ads. The differentiation allowed us to further track our ad effectiveness, following the usage amounts of coupons and attributing the success of a campaign to each unique code.

When launching HungryApp™ we wanted to build momentum, and therefore offering discounts as incentives added to the appeal of the new food delivery application.

QR Codes.

Aside from digital campaigns, we also created various printed assets with printed coupon codes to further assist our awareness campaign.  Placing QR codes directly on printed materials acted as a catalyst for increasing downloads at speed, and ease.

A5 Flyer with QR Code.

Growth Stage.

Once we launched the  MVP  successfully, our product development priorities switched from trying to get out of "MVP" mode as quickly as possible, ensuring that the application offered all the features that modern savvy customers would expect. Via user research, we identified the following gaps in the product:

  • Order History. The ability to review the order history is critical to ensure that customers can re-order favorite orders, which is expected customer behavior.
  • Product Search. Because all the brands in HungryApp are real-world brands that customers know and love, they often already know which products they like, and so providing a search function is an obvious way to ensure that customers can quickly and easily find the products they know and love.
  • Favorite Products. Linked to the above point, giving customers the ability to create a list of favorite products that they can quickly re-order is an integral part of ensuring that we promote repeat orders as much as possible.
  • Rich Notifications. These are critical to ensuring that we can have many repeat customers, as we can remind users that HungryApp has various offers for them. Our MVP notifications just opened the app to the home page, while the new notifications.
  • Additional Promotions. Initially, we launched with only the ability to have fixed discount coupons, but of course, we wanted more flexibility in our marketing promotions. We identified the key ones were BOGO (Buy One Get One) that is especially popular for pizza products, often boosting sales by 2x-3x on the days that this promotion was running. The % discount is also important, as it enables RMA Group staff discounts and being a favorite with consumers.
  • KPI API. All great companies use live KPIs to ensure that they are on track with their stated objectives. Having a specific API (Application Programming Interface) that exposes critical data for KPIs was vital. This ensured that the client could have all the necessary data to run their dashboards in PowerBI.

Digitizing Your Business Growth

Work with our expert team to transform your business and exceed objectives. 
Together we can Make It Happen.™

Start a Project
back to top icon