Digital unit in focus: Lighthouse Project DairyNet

We spoke to Holger about GEA Digital’s Lighthouse Project DairyNet, the demands of SaaS, and what makes the dairy industry so much fun for software developers.

Why was DairyNet chosen as a GEA Digital lighthouse project?

Data is the new gold! At GEA we are working close with our customers to create new digital products that can unlock the power of data provided by our machines and systems. We’ve had a big head start here at GEA Farm Technologies since we’ve been in the software business for decades already. The predecessor of our DairyNet herd management software – DairyPlan, essentially the enterprise resource planning (ERP) system for our dairy farmers – actually dates back to the late 1980’s (think MS DOS). We launched DairyNet in mid-2021 as a brand new successor with a completely new design. It is more intuitive and responsive, uses mobile apps, and will become increasingly cloud-based. It is a software-as-a-service (SaaS) product, so an established example of a recurring revenue business model, which is another focus area for GEA and GEA Digital.

What is the key to successful software?

We’re an engineering company here at GEA, of course, but developing software, especially SaaS, requires a slightly different approach. We start by creating a solution that is recognized as versatile and state-of-the art in all relevant markets – that serves the biggest common denominator. Once we have achieved the minimum viable product (MVP), we get it out to our customers for real world application as soon as possible. SaaS customers expect regular updates and new features; at the same time, we rely on their feedback to guide our next steps. Without the MVP out there, we can’t learn about what is working well for customers and where there is friction.

What is your development process?

Our core software teams are structured in typical agile fashion, with a product owner (PO), scrum master, four software engineers and 1-2 testers per team. Ongoing feedback from pilot/prototype farms gets channeled through the PO and integrated into the list of priority tasks for the next sprints. During the initial development phase, our rhythm is a new software release roughly every 6 weeks. That’s pretty quick, but pilot/prototype customers expect this; they want to see that we’ve integrated their feedback. For our developers, as well as our sales and service teams, it’s a never-ending process. After the launch, the time between releases is a bit longer, but our subscription customers expect regular improvements and updates since they are paying for DairyNet monthly.

Often our team will return from a customer visit with feedback and additional requests that need attention right away. So our sprint planning needs to be flexible; we need to be able to accommodate issues that come up and resolve them quickly. It’s an ongoing, circular process of feedback, fixes and new features. There is no final product. It also means we have to continuously invest money in making the software more stable and intuitive with more functionality. If we stop doing this, we can raise the white flag. This is just a fact of life in the SaaS environment; we’re in perpetual start-up mode.

Why do developers like working on DairyNet?

One of the things that makes the DairyNet experience unique – and fun – for our software developers is the “real world” nature of the solutions they’re developing. They visit the farms regularly and see how their solutions are helping improve animal health, increase milk yield or reduce machine down time. If you’re a developer writing algorithms for a big financial or insurance company, you’re probably not getting such direct feedback on the impact of your work. In fact, all the developers we’ve hired came from other industries. They like the “hands on” nature of the work here.

Our developers are also contributing to more sustainable farms. DairyNet empowers farmers to reduce milk loss, increase feed efficiency and boost cow productivity; this means not only more sustainable diary operations, but also notable resource savings. Another one of our teams is working on GEA’s automated feeding system, which can help farmers significantly save time, reduce energy and feed losses, or even help them adjust feedstock ingredients, which can help to reduce animal methane emissions.

What are the next steps for DairyNet?

Farmers face a new challenge these days: more focused manure management. Due to fertilizer shortages since last year – as well as the Ukraine crisis and the prospect of further natural gas shortages – standard mineral fertilizer has become prohibitively expensive, in part because ammonia is made from natural gas. In response, farmers are reverting to some use of manure as organic fertilizer where possible and available, which is a very old practice of course, but also a good example of circular economy in action. As with feeding systems, our software can become a valuable tool to help farmers monitor manure production, composition and usage.

We are also moving more and more of our software functionality from the farm into the cloud. This means data collected on the farm is transmitted to the cloud for storage, organizing data, additional analytics, and business intelligence (BI) reporting. As the barrier between local and cloud functionalities blurs, DairyNet starts to function more like Office 365, for example.

All in all, dairy offers a more fast-paced, dynamic environment than many people realize. But that makes sense when you consider that we’re building software solutions for complex interrelationships between man, animal and machine – and helping guide this system towards the best, most sustainable outcome.