Executing a Proof of Concept for an app to easily request new products
Just as consumers buy their products online, supermarkets also have to buy their products or request new articles for their assortment from a distance. For many this is done centrally. The chain is organized in such a way that every country orders their products at the supermarket headquarters. Mainly done by sending emails or filing request forms. This takes a lot of time and money and is also prone to mistakes.
A German supermarket holding recently asked LINKIT to look into this and to build a PoC (Proof of Concept) to validate a better way of requesting new products to be added to the product range. With a simple web app it is possible to directly order products from a catalogue, with most of the requested data automatically filled in. Doing so, every supermarket purchaser can choose from all possible products the chain has to offer. A supermarket for the supermarket.
Saving time and effort
Previously, processing numerous requests from all over the world was a full-time job for ten to 15 people at the supermarket’s office. In total, ten to twelve fields have to be filled in before a single request can be carried out. Next to this, when everything is requested by mail or standalone forms, there is a significant chance of missing vital information.
At some point, two managers came with a proposal to innovate this process and make life easier for everyone. LINKIT got the assignment to build a PoC of this app in low code. The purpose? To verify if an app would really be the simple solution the managers had in mind and if low code is the go-to solution.
The beginning of a PoC
Building a PoC is not just a technical challenge, explains LINKIT’s Solution Advisor Jimmy Iliohan. He was closely involved in this project. “It is also a challenge to make the right selection of functionalities to be implemented. Building a PoC only requires a short amount of time, ten days, in which we need to prove the concept. This means we need to focus on one or two of the core functionalities and leave out everything else. To figure this out we kicked off the PoC with a design workshop, together with the product owners from the supermarket chain.”
“To distinguish the nice from the necessary we divided all functionalities (or user stories) into categories”, Jimmy adds. “Those which were easy to implement and would have the greatest impact on the application were our main focus. Next to this, there were some criteria we had to decide upon: the minimum requirements we had to build to be able to confirm success. They also have to be included of course.”
Adapting to reality
The workshop led to a different outcome compared to our starting point. Jimmy explains: “Originally the idea was to build an Instagram like catalogue where you could not just order products, but also swipe through them and like them. For the PoC this was way too fancy. Therefore, we would focus on just the ordering mechanism eventually.”
Developing a solution in low-code
LINKIT’s OutSystems developer, Abeer El Assal, was also involved during the whole process. “Using low code for a PoC is a smart choice,” she says. “Unlike any other technology available, low code makes it possible to create an app amazingly fast. This is because it works with ready-made pieces of code you can drag and drop, instead of having to write everything from scratch. By pulling everything in the right logical order you can come up with a fast model to test the main functionalities. Stakeholders from the supermarket helped by providing some mock-ups for the design we could use as a start.”
Progressive Web Apps as an alternative
For Abeer, this made it easy to work. “We build the app as a Progressive Web App. Unlike a normal (native) app, you don’t download a PWA on your phone. It functions as a website on your browser, but with the extra functionality that you can add it as a direct link on your phone. It appears on your screen just like any normal app.”
“That doesn’t mean it’s simple to build,” she continues. “PWA’s have their own perks. One of them we noticed when we wanted to include push notifications. For example, when you have filed a request and got an approval or denial. For Android, we could work this out, but not for IOS. They have made it impossible for a PWA to send any messages directly into their system, so we had to find a workaround. Eventually we chose to send SMS-messages instead, which also does the trick.”
After fifteen days the PoC was finished. A little late one might think, but LINKIT decided to extend the ten days themselves to deliver more value. The next step was the final review in which the product owners give feedback on the result. For Jimmy this was one of the more exciting moments of building a PoC. “Luckily, we got particularly good feedback.”
What happens after a successful Proof of Concept?
The main reason to carry out a PoC is obviously to see if there is Proof for the Concept. This will decide what will happen next. According to Jimmy there are three possible scenarios: “It works and we will expand the PoC into a MVP, a Minimum Viable Product. Or, nr. 2, it doesn’t work, but we think we can make it work. Or, nr. 3, it completely fails and we stop working on the PoC.”
Jimmy continues: “Often a PoC is a success, but sometimes it doesn’t. That doesn’t mean you have thrown away your investment. Especially for bigger projects and IT investments it is already worthwhile if only one out of ten PoC’s is successful. It is a great way to explore new innovational concepts against limited investments. And to assure more business for your company.”
Are you also interested in building a PoC for one of your innovational concepts? We can help! Get in touch with one of our experts for more information.