Loft Home Rental App

We had an idea for an app that matched people moving house with neighborhoods that matched their lifestyle.

Loft final UI

Design Problem

Most of us have had bad experiences with renting apartments or homes. Finding the right place is difficult, listings can be misleading, and information about the surrounding neighborhood usually requires a local’s knowledge – placing new arrivals at a significant disadvantage.

To begin with, we set out to validate that this was truly a problem via survey and group interviews. We found that:

  • 80% of respondents had had negative experiences finding rental homes
  • 70% of respondents expressed interest in a rental matching service
  • 65% of respondents viewed Craigslist negatively

Therefore, our initial personas became:

  • University students who tend to move frequently
  • Young people (18–30) interested in settling somewhere for the longer term but  unable to buy a place

This led to two basic user stories:

  • I’m a university student looking for the right neighborhood
  • We’re a young couple saving for our first apartment, but still want to be part of our community

User flow

Our initial thinking was that users would simply answer 4 qualitative preference questions (ranked on a scale of 1 to 10) from which our algorithm would recommend a short list of suitable neighborhoods. From there, users would begin their search, relatively confident they were looking in the right places.

Loft whiteboard

We condensed this...

Loft App Flow

Into this.


We went through many iterations to arrive at a user flow that made sense. We started with why, moved through the what, before presenting the results.

Loft first user flow iteration

Our initial wireframe presented users with a matching flow – starting with questions – designed to understand their rental needs but gave them no option to skip it entirely.

First Iteration

Our first iteration drew heavily on dating apps for inspiration. We began with a list of questions to gauge what the user was looking for. First we asked why in order to ensure a quantitative match, say near a university or work if the reasons for moving were listed as school or work respectively. We then asked users what they were looking for in a neighborhood. This was centered around qualitative lifestyle factors such as restaurants and nightlife, parks and cafes and so on.


After testing this with users, it became clear that users wanted to see rental listings immediately, and beginning with a wall of questions – even in the context of returning more highly personalized results – was a significant obstacle to that.

Loft second user flow iteration

Our revised user flow reversed the order, moving what was the results page to the start of the flow, and changing the matching flow to an optional process.

Second Iteration

In the second iteration we reversed this flow, placed the listings first, and made the matching process an optional feature the user could engage with if they wanted to.

Users responded to this change by engaging more with the app. However, the matching algorithm – the key feature of the app – was now buried, and without engaging with it, the app lost its core strength and became more like every other listing platform out there. We worked with a mathematician who worked in Las Vegas calculating the odds for new games being introduced on casino floors to build an accurate algorithm. To lose this core feature which worked exceptionally well, pushed us to revisit our original premise entirely: is there a market for a rental app that matches people with properties?

Unexpected Obstacles

The unexpected obstacle then was that while people said they would like a better rental matching experience than what currently exists in market, their behavior suggested that finding the right rental apartment just wasn’t something they were willing to invest time in. For example, we found our initial target users spent three weeks researching and planning vacations, while on average, only three days researching rental apartments. That is, it's seen as a chore not an experience, which would explain the enduring popularity of Craigslist’s seemingly outdated experience.

Our initial target market was overwhelmingly motivated by price and proximity to school or work. So much so that qualitative neighborhood characteristics became virtually irrelevant. There were exceptions however. We found that young families and corporate movers both cared deeply about neighborhood characteristics and would pay for such a service.

The final UI has a flow that users found more natural with an emphasis allowing users to move through the process quickly to reach their goal of viewing suitable properties.

Loft Final (mockup) UI

While we never spent much time on the final UI, we envisioned it to be clean, fun and intuitive. Here I've used Pablo Stanley's Humaaans illustrations and stock photography to give a sense of what the finished product may have looked like.


The UX research process worked well to invalidate our initial working assumptions and allow us to iterate to a solution that met user needs. We were able to reveal a gap between what users said and how they behaved and learned a lot about what works and what doesn’t in terms of onboarding and getting users to the content in the fastest way possible.

We also identified two groups that were not initially considered – young families and corporate relocators – to whom this service was valuable, and around whom a viable business could be built. However, this potential solution didn’t need to take the form of an app, so with a whole lot of learning under our belts, we wrapped up the project to focus on bigger and better things.


Annee Ngo – Canadian 30 under 30 Entrepreneur
Euphemia Wong – User Experience Designer, Relic Games