AI In Real Estate Use Case #3: Houzen

  • 11 July 2019
  • Sam Mire

This interview is part of our new AI in Real Estate series, where we interview the world's top thought leaders on the front lines of the intersections between AI and real estate.

In this interview, we speak with Saurabh Saxena, founder of Houzen, to understand how his company is using AI to transform real estate, and what the future of the real estate industry holds.


1. What’s the story behind Houzen? Why and how did you begin?

SS: My career spanned private equity and CxO advisory, where over the years I advised invest/don't invest decisions for buyout funds, and advised CxO's on corporate strategy, m&a, branding, pricing, etc. across twenty-plus industries and three continents. In one such deal for a large global PE fund, I advised against buying one of the largest real estate agency chains in Western Europe. That deal helped me recognize the grassroots-level issues in real estate, i.e. the highly fragmented nature of agency market, high commissions for low service levels, and most importantly, the very long time it takes to transact a buy/sell or a rental property — stretching to several months in some cases. What I also realized was that these issues were similar across the world whether transacting in UK, France, Germany, India, or Singapore.

Houzen's concept was born as a global exchange which is based on a Netflix-styled recommendation system to match properties and customers, i.e. tenants, buyers, and short let tenants globally. A pull system, versus the current push system, would hopefully help enable transactions in near real-time, similar to how we buy flights and hotel rooms today, and reduce costs by 50-60 percent across the value chain.

I set up an initial engineering team of three while still in my day job, and started experimenting with product features. I ran into a senior executive from Greystar at the Oxford real estate conference where we debated about the future of global real estate. To my pleasant surprise, my hypothesis seemed to fit perfectly with the needs of global real estate titans. Greystar became our first client, I quit my job and the rest, as they say, is history!

2. Please describe your use case and how Houzen uses AI:

SS: My pre-MBA experience was advanced statistics, which today goes by a fancier name called machine learning or AI. Essentially, it is all about creating an equation which is taught as being right or wrong as the dataset grows from a sample to a population. More data means more opportunity to be right or wrong and the equation learns from the outcomes constantly. Having been in the market for two-and-a-half years now, we have collected sufficient decision data on what tenants or buyers are renting or buying into.

For context, there are about 150,000 property units listed on portals live in London, and about one million tenants or buyers search them each month. We have built a data-backed recommendation equation, which matches property attributes with potential buyers and tenants based on a number of factors. Over time, since we've captured a lot of data points on both supply and demand sides, this equation is becoming more accurate, and hence the lead time to close deals is shortening due to the higher quality of matching. It's quite simple in theory, however extremely hard to execute, as this data and these transactions are locked in a highly fragmented and cagey industry of agents and brokers — unlocking this is a trillion dollar opportunity!

3. Could you share a specific customer/user that benefits from what you offer? What has Houzen done for them?

SS: Sure. We work with some of the largest global private equity and asset managers such as Greystar, Invesco, Kennedy Wilson, L&Q, etc., who traditionally used the big six agents such as Savills, CBRE, JLL, Foxtons, etc. We come in as a challenger brand with the pitch to be two to three times faster than any of these corporate agencies. In each development we've supported, we've delivered one to two million pounds ($1.25 million to $2.5 million US) in additional revenue and have been two to five times faster than the competition. We typically work on entire blocks of hundreds of flats and have proven that our service works beautifully across all customer types, i.e. tenants, short lets, buyers and property types e.g. houses, new builds, old stock.

4. What other AI use cases in real estate are you excited about?

SS: Since real estate is a non-standardized product, right valuation or pricing becomes a lengthy decision-making factor. Due to the lack of good valuation tools, the skill in real estate deals becomes less data led and more people and negotiation skills led. We see a big opportunity in using computer vision to correctly price assets based on geo-tagging, and to test the internal quality of assets through pictures and videos, street views, etc. especially in such a culturally and socioeconomically and diverse city as London. This is a great use case, however, it requires a lot of investment. Startups may find it difficult to find investment to scale this kind of an idea, however corporate tech players could do this if their shareholders permitted it.

Another exciting use case could be in architecture design. My sister is an architect from AA, a top three global architecture school, and works with AECOM, a top global firm. We regularly have a family debate about the future of architecture. What I feel is lacking in this discipline is the feedback loop from consumers.

Architects typically build with a budget and briefs in mind but rarely hard, predictive consumer data in mind. I have great respect for architects and feel that they could also start using customer feedback data to optimize their design not only on the aesthetics but also on the engineering design side using design optimization through the BIM/CAD softwares. Since most designs are stored on CAD/BIM, one could argue that the supply side could be aggregated. All you need (much easier said than done) is to receive historic and live customer feedback on how they use these spaces and building structures so that a predictive design model could be constructed in CAD, and thus building could be built for a specific purpose.

5. Where will  Houzen be in 5 years?

SS: Europe and Asia. Potentially also parts of the US. We are already recruiting for Singapore as we are seeing a lot of traction from Asia. We'll also be the trusted global ecosystem to close deals across borders and locally in real time, and also settle payments and handle other legal documentation around transactions.

About Sam Mire

Sam is a Market Research Analyst at Disruptor Daily. He's a trained journalist with experience in the field of disruptive technology. He’s versed in the impact that blockchain technology is having on industries of today, from healthcare to cannabis. He’s written extensively on the individuals and companies shaping the future of tech, working directly with many of them to advance their vision. Sam is known for writing work that brings value to industry professionals and the generally curious – as well as an occasional smile to the face.