--- title: Planning answers in minutes, not months slug: farm-planning summary: For NSW DPI, turned the worst part of a 6-to-12-month farm development application, working out which rules apply to your land, into something a farmer could do alone in a single sitting. Grounded in 28+ customer interviews; 18 of 20 pilot users rated it highly desirable. Grew into myFarmPlanner, a 2025 planning-award finalist. url: https://charlieholland.com/#work/farm-planning --- # Planning answers in minutes, not months **Client:** NSW DPI · **Role:** Lead Product Designer · **Year:** 2021 to 2022. Led product and service design for the NSW Department of Primary Industries so farmers could work out their planning requirements without hiring a consultant. Grounded the design in 28+ customer interviews across rural and regional communities, then designed and validated an MVP with the farmers who would use it. The MVP grew into myFarmPlanner, now live to the public and a 2025 planning-award finalist. Headline outcomes: under 10 minutes to work out what applies to your land (tracked in product analytics), down from a 6-to-12-month pre-application process. Measured: 18 of 20 pilot users rated the MVP highly desirable. Projected: consultant reliance down 40%; staff efficiency up 50%; a $3.9m five-year benefit. ## Overview: a decision engine for land you're actually allowed to develop A farmer wanting to develop their land could wait 6 to 12 months and pay a consultant just to find out which rules applied. I led product and service design for the NSW Department of Primary Industries, turning that into a self-assessment tool that does the same job in a single sitting. An intensive agricultural development, a feedlot or a dairy barn say, sounds just as complicated as it is. To get one approved, a farmer has to make sense of local legislation, environmental concerns, and consultant jargon. It is tedious and easy to get wrong. One farmer described fees reaching $34,000 in an extreme case. We built a decision engine that holds the legislation and planning policy as code. I designed the way a farmer uses it. They draw the development they want, and the rules answer against their own land. These are the people who are the backbone of our country, and frankly they deserve better. ## Challenge: the paperwork, not the farming, was the barrier Compliance falls hardest on small businesses, and across Australia's SMEs it ran to around $80 billion a year at last count (Australia-wide SME compliance-cost figure, citing the Taskforce on Reducing Regulatory Burdens on Business, the Centre for Labour Market Research, and Deloitte's "Get out of your own way", 2014). For a farmer, the development application was the sharp end of it: 6 to 12 months of work, consultant fees, and a decent chance of reaching the end still unsure whether the project was even viable. As a result, NSW was losing agricultural investment to other regions. The department's first instinct was to improve the form and its logic. The problem reached much further than that. Conversations with farmers, industry experts, consultants, and the consenting authorities who approve applications proved far more telling. This was an ecosystem problem. Farmers described misleading and conflicting information, and not knowing who to trust. In their words: "Come on, we're family farmers, not a corporation." "Consenting authorities don't understand what's required to make the rural area run properly." "Farmers tend to have a lack of knowledge and awareness with the process." The consenting authorities, the council and agency staff themselves, described farmers arriving with no real sense of what the process required. The rules weren't just hard. They were scattered, written for experts, and impossible to apply to one specific property without paying someone who spoke the language. ## Approach: from field research to a tool a farmer could use alone ### Four personas, four definitions of "help" The work opened with 28+ customer interviews across rural and regional Australia, run fully remote under COVID. What came back wasn't one user with one need. It was four, and each one pulled the design in a different direction. - "Give it a go" Garry didn't trust the jargon. So plain language wasn't a nicety, it was the product. Every question in the tool had to explain itself. - "Generational" Jane needed an answer tied to her specific block and its soil, not a generic checklist. That is why a map of her actual land became the spine of the tool. - "Urbanite" Stuart wanted a straight, accurate read of what would happen, with no surprises later. - The consenting authority on the other side of the counter needed a complete, low-risk application landing in their inbox. ### The map is the territory Real context was the missing piece. A checklist can tell you the rules, but it can't tell you the rules that apply to your specific land. Buffers, separation distances, and environmental zoning decide whether a development works, and none of them can be answered off a form. The moment a development area is drawn, the feedback is live. Thirty codified rules from legislative acts, planning guidelines, and local environmental plans resolve against that patch of land, in plain language. For the first time, the farmer and the council are reading the same answer. ### One output, two readers The hardest decision was not a screen, it was who the screen served. The farmer's question is "can I build this on my land?" The authority's is "is this application complete and low-risk?" Rather than build two tools, I made the case for one self-assessment that produced both: for the farmer, a draft application and the contacts to take it forward; for the authority, that same submission arriving with the development intentions and a risk profile already attached. The map is where those two opposed needs finally met. ## Results: validated with the people who would use it The old pre-application slog ran 6 to 12 months. The tool takes its worst part (working out what applies to your land) down to a single sitting. Product analytics on the MVP clocked it at under 10 minutes; pilot users called it "quick to complete". I put the MVP in front of 20 farmers and industry officers. 18 of 20 rated it highly desirable. The client's business case projects the rest: reliance on paid consultants down 40%, staff efficiency in customer-facing roles and planning determinations up 50%, and a $3.9m benefit over five years as staff spend less time shepherding applications (based on the client's own economic analysis; modelled projections, not measured outcomes). I've regrouped the five-year impact case by who the impact lands on rather than the five reporting lenses the client's deck filed it under. The $3.9m benefit and the 50% staff efficiency gain both land with the department; the 40% consultant reduction is the farmer's. Beyond the numbers: for the farmer, simpler processes and faster delivery; for the region, economic growth and social cohesion by making it likelier the right development happens in the right place; for the department, a greater ability to meet customer expectations with fewer application rejections and delays. In their words: "The biggest achievement of the project is that it actually provides a tool that serves a real purpose for intensive livestock producers." (Group Director, Livestock Systems). "I was impressed by the simple functionality, advanced mapping function and quick to complete. Certainly, a useful and successful tool in my opinion." (Primary Producer, pilot). "myFarmPlanner will be a very useful tool for the dairy industry as a lot of farmers are land locked, and this will be further intensified by building barns and feedpads etc." (Dairy farmer, Northern Rivers NSW). ### Recognition The MVP became the foundation for myFarmPlanner, the farm-planning platform the department (by then renamed NSW DPIRD) later launched to the public. In 2025 it was named a finalist in the Planning Institute of Australia (NSW) Awards for Planning Excellence, in the Technology and Digital Innovation category. Delivering it meant working across government, alongside Digital NSW, the NSW EPA, and GovCMS, whose Rules as Code approach sits directly behind turning legislation into software. ## What was hard: simple was the expensive part Codifying rules written for experts. Legislation and planning guidance are not structured data. Turning 30 rules into something a farmer could trust meant sitting with subject-matter experts and making real calls about which edge cases to handle in the interface and how far to simplify without being wrong. Serving two readers with one artefact. Everything the farmer found reassuring, the authority found insufficient, and the reverse. The map survived because it was the one surface that could speak to both at once. Research you could not do in person. 28+ interviews with rural stakeholders, run entirely remotely under COVID, about land none of us could visit. Reaching people who do not live online, about property you can only see on a satellite tile, was its own design problem.