Pruksa taps on Tableau to help customers home in on the right house


Increased sales by finding the right house for customers
Reduced time to analyze construction site progress
Improved construction processes by studying defects

With projects spanning across Thailand, India and the Maldives, Pruksa is one of the biggest real estate and construction companies in Asia, raising US$1.47 billion in total revenue in 2018.

The company, founded in 1993, stands out in an industry dominated by traditional practices for its forward-looking use of data, which has helped it edge ahead of the competition. Pruksa’s Head of Data Nassal Leehard said Tableau has boosted the reporting of construction progress and enhanced understanding of customer behavior – an “intuitive and fast software” that has helped translate big data and machine learning into tangible benefits for the business.

Before adopting Tableau, our system was quite slow, and the data we got was not up to date and inconsistent. The data was unusable. So we started to look for a tool to solve such issues – to make data consistent, more real-time and with a more analytic approach. We found it in Tableau.

Boosting sales by pairing customers with the perfect fit

Property developers often face one problem in sales: they pour in large sums into marketing efforts and build exquisite showrooms, only to find that showroom crowds do not correspond to sales figures.

“It has become a pain point when people visit our projects but leave without buying,” said Nassal. This may well be happening because customers are viewing an unsuitable property for their individual profile and needs, and are unable to commit to a purchase, as a result.

To better rake in sales, Pruksa now uses Tableau to organize customer information, to match the right project to the right client. The data is sorted according to key metrics that range from the loan amount that a customer can receive based on his income and his marital status to which departmental store he frequents.

In a market as competitive as real estate, peripheral information like these make up a small advantage that can add up to big wins.

“This customer behavior is recorded to develop a machine-learning model. We then analyze it so that we can find the best project for our clients,” Nassal said.

Minimizing manual labor through smart data use

For a large real estate developer like Pruksa, keeping tabs on multiple construction sites to ensure all projects are delivered on time is critical. After all, delays mean additional costs and are likely to put a dent in a good reputation that has been carefully honed over the years.

With Tableau, construction progress reporting at the company has become faster and neater, while precious manpower has been freed up from menial tasks to do higher-value work.

For instance, after collecting data by flying drones over construction sites, Pruksa can now analyze the data when it is inputted into Tableau. This allows it to distinguish between greenfield and brownfield sites, and to ascertain the overall progress of construction.

“This is instead of having someone survey the site, take notes and analyze in Excel, which will consume so much time,” said Nassal.

Tableau’s intuitive interface cranks up the speed further by allowing employees to extract data much faster than before.

“Before adopting Tableau, our system was quite slow, and the data we got was not up-to-date and inconsistent. The data was unusable,” Nassal said. “So we started to look for a tool to solve such issues – to make data consistent, more real-time and with a more analytic approach. We found it in Tableau.”

This customer behavior is recorded to develop a machine-learning model. We then analyze it so that we can find the best project for our clients.

Building safely by using AI to study defects

In the real estate business, safety and customer service is paramount.

As part of its after-sales service, Pruksa takes care to address home defects that have been reported and to make sure that the same mistakes are not made again. It is also typically a laborious process that requires an expert to attend to defects one at a time and manually categorize it, before writing a report.

Tableau has simplified and sped up this process by crunching data collected on defects.

Using photos of defects – such as decaying concrete, broken windows or damaged door knobs – Pruksa is able to use the software’s machine learning capabilities to analyze the cause of such defects, attribute it accordingly to the consumer or supplier, and categorize it based on an extensive database. This has improved the firm’s construction and delivery processes greatly.

Nassal said: “A study found that by 2030, AI could replace as much as 14 per cent of human labor. With the vision of our executives, Pruksa intends to focus on big data and machine learning to maximize the benefit for the company.”

A study found that by 2030, AI could replace as much as 14 per cent of human labor. With the vision of our executives, Pruksa intends to focus on big data and machine learning to maximise the benefit for the company.