Insight

Habito. Humanising Artificial Intelligence With Daniel Hegarty

Richard Moss, Principal, Technology & Central Operations, Holtby Turner Executive Search in conversation with Daniel Hegarty, Founder & CEO, Habito.

In a market worth £1.3 trillion, with just over 11 million mortgages, the UK mortgage industry is the largest in Europe, in terms of the amount lent each year, and the total value of outstanding loans. Yet despite the money in the market, buying a home, or re-mortgaging one, remains a headache from start to end no matter how it’s done – online or offline.

It was a headache that turned into a nightmare for Daniel Hegarty, Founder & CEO of PropTech start-up Habito. His vision is to make getting a mortgage easier, cheaper, faster, and more transparent – and in so doing, is disrupting a trillion-pound mortgage market. Now, at just eighteen months old, Habito raised a Series B in September – with a round of £18.5 million from top VC fund Atomico – and launched its first TV ad campaign on ITV, which is sure to drive a high volume of traffic.

Habito’s beauty is it sounds so simple a business to build. But building it, with such a tiny team in such a powerful industry, was no simple venture. Richard Moss, Holtby Turner Executive Search’s Principal for Technology & Central Operations asked Daniel to share the story behind Habito to learn how their use of artificial intelligence is innovating real estate.

With a house in London found, an offer accepted, deposit good to go, and mortgage application submitted, the worst should have been over. Until the mortgage application was declined due to the broker listing Daniel once and his partner twice on the application.

A week later they went through it again – application submitted, application declined – this time with Daniel missing, and his partner listed twice. In Daniel’s own words, “another week goes by. Application submitted, application referred!? The lender needs three years of P60s … Our seller starts to get impatient, threatens to pull out and wants to see bank statements and credit reports. Our dream home was fading from view while we’re buried in paperwork. In the end we made it, thanks to some bike couriers and sympathetic ears in the HR departments of old employers.”

“I couldn’t believe this was normal in the 21st century” he continues, “until I looked at what was going on in the market.” It was that personal point of customer pain that prompted Daniel to begin building was has become Habito.

Today, Habito’s use of automation and artificial intelligence simply doesn’t allow for the level of errors that Daniel experienced. It also makes you wonder how the original process got to be so hard and involve so many stakeholders– from buyers to sellers, regulatory bodies to lenders with surveyors and agents in-between.

How do you make AI driven conversational bots feel more human, and trustworthy, than another person?

One of the first things you notice with Habito is that it’s jargon-free and has a calming, uncluttered design. We decided at the outset, we would communicate with you – our customers – like a person, because the only way to win your trust is to build a business which puts your interests first.

Where did you begin the design process?

Before we started Habito, the only way to receive mortgage advice was over the phone or face-to-face. It was slow, paper-based and prone to human error, frankly it was unfit for the 21st century.

We knew technology could do this much better which is why we took this huge step to revolutionise how consumers apply for mortgages today. Our system brings robo-advice to the mortgage market for the very first time.

Habito’s Digital Mortgage Adviser (DMA) explores all the key components of the customer’s financial life (such as employment, salary and personal life plans), and explains the impact of these on the kind of mortgage that will work best for them. Then the algorithm uses real-time market rates to calculate an indicative monthly payment, based on their circumstances and preferences.

The design process involved conducting hundreds of customer interviews to allow us to understand what was important to the consumer and what forms the basis of great mortgage advice. Based on reiteration, testing, and reiteration again as our methodology, we got evolving customer insights on the product as we built and designed our DMA.

How have your product team managed to humanise what is essentially a tech process?

We humanise the technology by making it conversational in tone. When you click on the DMA, you are welcomed, the process is explained and the stages are simple to complete. We actually slightly slowed the pace of the original DMA because our customer research showed people prefered to get the information as they would a conversation, and giving all the information upfront was off-putting. Once they have done the DMA, customers can also talk to one our mortgage experts (who are real humans!) via live-chat to confirm they’re happy and ask any further questions they might have. We’ve found that these interactive features increase consumer understanding and improve decision making.

How did you set about accessing all the data you needed amid the myriad of stakeholders in the property chain?

Slowly! Three of us started in the kitchen of my house and began to research and plot all the lenders and mortgage products in the UK (there are over 70 lenders’ and over 2011,000 products). It was an arduous process though, it took over a year and we are still developing our technology. Given the complete lack of digital infrastructure in the mortgage industry, even trivial things take serious engineering effort. We continue to expand our developer team pretty rapidly to achieve our ambitions.

Technically, building the Habito platform must have been full of integration challenges. How did the mortgage sector’s IT infrastructure cope with your highly complex algorithms and use of AI and machine learning?

Mortgage lenders in the UK are notorious for having multiple, nebulous and often ancient technology systems that simply don’t have the ability to interact digitally with other systems, so getting our system to interact successfully with lender’s systems, is still a challenge.

One of the more surprising challenges was not with the building of the product (we always knew that was going to be hard) but with gaining consumer trust.

Even on a good day, it’s difficult to navigate the complex array of financial products we all have accumulated over our lifetime. It’s daunting and unnecessarily repetitive, yet we accept the norm and never fight for a better experience.

We felt that an artificially-intelligent financial adviser product, which is not sensitive to advertising and can do the sums on all the incentives to see through to their real value (or lack of) on your mortgage rate, would be easy to sell to consumers. After all, few consumers relish getting into the weeds with even the most basic financial service, so what consumer would turn away from knowing that their financial life is a solved problem?

But, people can be complicated and there are big trust issues with machines, even now. With the rise of products like Amazon’s Alexa people are getting more used to the idea of conversational robots, but still one of our challenges is showing that done right, AI, machine learning and algorithms can give consumers the ultimate power of convenience over their finances at a fraction of the traditional cost.

When building Habito, how did you manage a continuous mode of innovating when still in the process of establishing product-market fit?

Like a lot of start-ups, we use a lean methodology when it comes to innovating. It means that we are continually tweaking, adding extra features, seeing if we can include other products. But as the site is live and functioning, with the numbers of customers using it growing every day, we know we have market-fit.

Launching Habito with a minimum viable product allowed us to test if and how customers use it, when, for how long. Do they drop out of the process and why? This is critical information for seeing if you have a viable business or not. It also leads to a continuous mode of feedback – some of our best product updates have, and continue to come from customer suggestions.

Where do you hope Habito will be in 10 years, and how will it have disrupted the mortgage industry?

Our mission is to make the mortgage industry fairer, more accessible and more transparent. In 10 years’ time, we would love for Habito to have helped prompt a cultural shift in the way people think about mortgages – from them being an expensive scary hassle – to being something that is as easy to switch as your broadband or car insurance from the comfort of your home, and something you do in moments, not weeks.

Read more about PropTech & Innovation in our Insight Report here, A Brave New World? Innovating Real Estate