Building a farmland valuation product that grew a lending business

Context

How do you reach a customer before they are ready to buy?

I led product strategy and growth for a relatively new agricultural lending business. The goal was to attract more farmers, turn them into financing customers and introduce them to a wider ecosystem of agricultural services.

One of the flagship initiatives was a farmland valuation product. It combined proprietary data science and machine learning models with a digital experience that showed farmers what a piece of land could be worth, based on terrain, wetness, slope, soil structure and potential yield.

I owned it end to end, from customer research and strategy to development, launch and commercial activation.

Challenge

A farmer buying land needs to look past the asking price. They need to understand the land’s characteristics and potential, judge its value and decide whether financing makes sense.

That was our opening. Instead of relying only on traditional marketing to promote loans, we could give farmers a useful tool for a real decision. If we helped them evaluate farmland, we would build a relationship with prospective borrowers early and give them a natural path into lending.

The hard part was turning complex machine learning valuation models and proprietary agricultural data into a product farmers could understand, trust and use.

Approach & Execution

Started with farmers, not the models

I spent a full month working directly with farmers to learn how they evaluate farmland, what information they need before a purchase and what they look for in a loan. I ran UX research and built user personas, and those insights shaped the product architecture and experience.

I put prototypes in front of farmers, gathered their feedback and built it into the requirements. The question was always what would help a farmer make a better decision, not what the models could do.

Translated data science into a product

I owned the product strategy and requirements. I worked with data scientists on the valuation models, and with designers and engineers to turn those capabilities into an experience farmers could use. The product gave farmers more data points to weigh before buying land and exploring financing.

Connected the product to sales and marketing

Building the product was only part of the work. Farmers and our commercial teams also needed to understand its value.

I partnered with marketing on the branding, service mark and launch campaigns. I also led discovery workshops that turned the valuation models into sales enablement collateral and marketing assets, so commercial teams could explain the data science in terms customers understood.

Fixed friction in the lending funnel

Alongside the valuation product, I dug into the lending funnel to find where prospective borrowers dropped off. I used cohort analysis and funnel diagnostics to understand their behavior, then ran structured experiments, including A/B tests, to improve conversion.

I also built a land loan calculator as an entry point on the website. Prospective borrowers could estimate financing costs at our rates and see what a loan with us could look like.

Outcome

• 50% month-over-month growth in qualified pipeline, as sales and marketing put the valuation product’s insights to work.
• Improved lead-to-customer conversion through funnel diagnostics, cohort analysis and structured experiments.
• Wide media coverage after the launch.

Takeaway

A product does not have to earn the revenue to grow the business. When you understand what customers need before they are ready to buy, you can build something useful that creates demand, earns trust and introduces them to the rest of what you offer.