Background

ACCENTRO Real Estate AG is the market leader in residential privatization in Germany. A large part of the customer base is made up of private investors who purchase real estate from ACCENTRO as an investment. Such investments are linked to important parameters. For example, properties are compared on the basis of location, market price, potential return and price development. Excel spreadsheets are often used for this purpose, but their maintenance and creation is laborious and error-prone. For this reason, there was a need for a more efficient software solution to this problem. A clear real estate comparison site for private investors was to be created. In cooperation with the agency Cheil, we put the requirements into an intuitive web application called “PropRate”.

The Sidestream Approach

Cheil first developed a concept for this. For us it meant:

Make the developed concept and user interface work as a web application. To do this, we first evaluated the feasibility. First, we had to identify basic features and functions in order to develop a first product (MVP) in coordination with our project partner. It was important to combine the implementation of the features with the user experience (UX) developed by Cheil in a meaningful way. This guaranteed a fast implementation with a focus on quality and also enables the use of state-of-the-art software technologies in the future.

The implementation: Focus on code quality and UX

Using the draft developed by Cheil, we delivered the MVP in five sprints. Within these five weeks we delivered the entire software architecture, developed the core features and already carried out test runs. The special thing about this is that we worked particularly closely with our project partner on this project. The daily exchange with Cheil throughout development enabled a direct culture of cooperation between UX and development specialists.

The product: efficiency and user-friendliness

Efficiency and user-friendliness are at the core of this project: “PropRate” combines both better data processing and a clear and intuitive user interface. Instead of tedious, manual work, the user is offered an automated solution. For example, the scraping of web pages enables the pre-filling of data such as the real estate purchase price. With just a few clicks, “PropRate” evaluates relevant properties and clearly displays the most important comparative values. Thus, nothing stands in the way of potential investments. In the future, the planned AI will completely take over the comparison process and thus find the right property without any effort on the part of the user.

Technology Deep Dive

Die Challenge bei der Umsetzung dieses Projekts war es, eine stabile, skalierbare Anwendung mit einem guten Look & Feel zu entwickeln und das gleichzeitig lean und in kurzer Zeit. Unsere Lösung: Der AWS Amplify Stack. Ausgerüstet mit diesem Set aus Tools und Services legten wir einen Schnellstart hin und das ohne Abstriche in der Qualität. Die Webanwendung ist eine moderne Mischung aus GraphQL, Nuxt und VueJS. Die komplette API wird dabei automatisch erzeugt. Dazu mussten wir lediglich ein GraphQL Schema schreiben. Alles Andere passiert von alleine. Denn AWS Amplify generiert eine AWS AppSync Anwendung inklusive der Zugriffsrechte und Authentifizierung off-the-shelf. Außerdem wird parallel eine DynamoDB hochgezogen.
‍
Auch das Deployment inkl. Ci/CD Pipeline steht quasi automatisch. Jeder Commit im Master wird automatisch gebaut, getestet und dann deployed. Diese Komposition ermöglicht einen hohen Entwicklungsspeed. Tailwind - unsere Lieblings-CSS-Bibliothek - sorgt währenddessen im Zusammenspiel mit VueJS für eine perfekte Modularität und flexible Wiederverwendbarkeit der einzelnen Komponenten. Außerdem wird so eine hundertprozentige Flexibilität und Individualität im Look & Feel garantiert. Dieses Fundament eröffnet eine schnelle und einfache Implementierung zukünftiger Features. Eines davon wird die Bewertung mittels Machine Learnings sein!

‍