When Dairy Australia first approached us, they presented an interesting challenge. They had valuable data about bull genetics in their Good Bulls Guide, but farmers weren’t able to access and use this information as effectively as they could. As one farmer told us during our research, “I’ll talk to other people for their opinion, the fellow at Viking will question my selection… there’s too much theoretical data that makes bad bulls look good.”
We quickly realized this wasn’t just about digitizing a guide – it was about fundamentally improving how farmers make breeding decisions. Through our initial discussions with Dairy Australia, we identified several key challenges:
Many farmers were spending hours poring over printed catalogs and guides. One farmer shared, “With phone calls and reading catalogs, it’s 2-3 hours annually. If it’s raining outdoors I’ll do it over morning tea.” Others told us they spent “5-6 hours over a week or so” researching bulls.
The existing process was frustrating. As one advisor explained, “Farmers don’t feel that genetics are important… but you’d generally just try to not use negatives, rather than going for huge specific positives.” This indicated a need to make the value of genetic selection more apparent and accessible.
Trust was a major issue. We heard repeatedly that farmers were skeptical of resellers and wanted independent information. One farmer told us bluntly, “I’m skeptical of some of the salesmen in the area.”
Our research phase was intensive and eye-opening. We spent three days at Ellinbank and Warragal, conducting in-depth interviews with farmers, advisors, and industry stakeholders. Rather than sticking to a rigid script, we let conversations flow naturally, which led to some fascinating insights.
One of our most surprising findings was the diversity of approaches to bull selection. We learned quickly that there was no such thing as a typical farmer. As one advisor put it, “All farmers are different, and some are really interested in genetics.” Another told us, “80% generally like what they’ve been told to like. 20% know what they want.”
The interviews revealed some consistent patterns in how farmers approach bull selection:
Price Sensitivity: “Cashflow is important so price is really important. Often budget will be the starting point, and then you try to find something that works within the price range.”
Trust Issues: “Farmers don’t like to make mistakes. They’re looking for someone to blame!” This insight helped us understand why independent, reliable data was so crucial.
Connectivity Challenges: One farmer explained their situation: “We don’t get mobile phone reception at the moment. Any app that I use has to use wifi in the house. Data is really expensive.”
Taking our research insights, we began crafting a solution that would work for everyone from the most basic user to the most sophisticated. We knew we needed to create something that, as one farmer put it, would be “very practical” and have a “clear point of view on what it does.”
We designed a sophisticated search system with several key technical components:
As one farmer explained their ideal process: “Select the five traits you’re interested in first, then set the scales for them, and then view search results. Otherwise I’d get jack with it, and get frustrated and give up.”
One particularly insightful piece of feedback shaped this design: “I’d like to look at protein, milk - so when I go to the next screen, my picks are shown, rather than all the other traits that are available.”
Given the connectivity challenges, we implemented:
As one farmer noted: “Data is really expensive. $89/month gives us 3000 mb in the afternoon. Midnight to midday we get 12,000mb.” This directly influenced our data management approach.
We created a hierarchical information display system:
This was influenced by farmer feedback like: “Numbers are really the most useful. You can look down the list and see where they sit.”
Our September 2015 testing sessions revealed crucial insights that led to significant design iterations.
The first round of testing exposed several navigation problems:
As one farmer noted: “How do I know to pick one? Need a visual cue to select bull. Tip on screen - pick one - return to menu to select a different bull.”
We went through three major iterations of the search interface:
User feedback was clear: “I feel like it is asking me to fill out everything.” This led to our first major revision.
A farmer suggested: “Can we go through what the most important traits and select them first as filters, then it goes to a second screen that allows those traits to be set for the search.”
This was validated by user feedback: “Better to add in one trait at a time when doing search and applying filter, then you don’t risk missing out on a really good bull.”
We made several iterations on how to display bull information:
User feedback was direct: “The listings don’t need to be too big. The company isn’t what we’re picking. I want the best bull and his name/traits.”
As one farmer noted: “Numbers are really the most useful. Data listing with trait = X is best, you can look down the list and see where they sit.”
This was influenced by practical considerations: “Farmer fingers - big fingers, cold days” and “I don’t want to look at 50 bulls in one list all the time. I’d rather see the top bulls first.”
We developed several iterations of the comparison interface:
As one farmer explained: “Once you’ve got them side by side you can compare and then take off the ones you don’t want.”
[Previous sections 6-8 remain the same…]
We implemented a sophisticated data management system:
To ensure smooth operation in rural areas:
This enhanced technical implementation was directly influenced by our user research and testing, ensuring the app would work effectively in real-world farming conditions.
We developed several iterations of the comparison interface:
As one farmer explained: “Once you’ve got them side by side you can compare and then take off the ones you don’t want.”
We implemented a sophisticated data management system:
To ensure smooth operation in rural areas:
This enhanced technical implementation was directly influenced by our user research and testing, ensuring the app would work effectively in real-world farming conditions.