Our customer had built an in-house rules-based model to price properties they were buying and selling but had recently found that it was undervaluing properties more often than not. The company was basing its pricing on standard factors that included a static neighborhood slide rule, square meters, and historic internal pricing data they already possessed. However, the model was not able to account for changing preferences in the area which could impact willingness to pay at certain prices, and which limited the realtor’s ability to sell properties at the optimal time and price point.
Beyond simply looking at historic prices and square meters, the real estate firm understood that they needed a pricing model that more accurately represented real-world demand and conditions. Historic price records don’t always account for shifting external conditions and events that could cause massive volatility in home pricing. Instead of a rules-based pricing model, the company was able to use Explorium to create a new pricing system that considered a variety of factors, including:
The company saw an immediate boost in its revenues and its ROI. While they had been closing sales before deploying the new model, their margins had hovered below 10% on average, largely due to sale prices that, while close to market value, didn’t truly maximize the potential of each property based on demand. After putting their new price prediction model into production, the company was able to quickly improve their margins to 14% and reported a higher sales volume in the quarter following implementation. On the whole, the company’s new model means that it can more accurately price homes, leading to fairer pricing without affecting their profit margins. With a higher sales volume overall, the company’s model lets it resist volatility impact by adjusting prices on the fly.