04-15-22 | Tech in Real Estate News | Population Trends

Migration patterns in the United States have been drastic over the last two years with shifts from large cities to smaller metropolitans. In this week’s Tech in Real Estate we cover population trends, data visualization, and APIs.

Transcription

Ariel Herrera 0:00

Hey everyone. Happy Friday and welcome to the tech in real estate podcast news for this week, we have a wide range of topics, including population declines in major cities and how that can affect demographics in the future. As well, we're going to be looking at house price predictions. And we're going to be focusing on what particular data you could look at if you want to analyze for sale properties off market properties, as well as just the general market. My name is Ariel Herrera with the analytics area channel, we bridge the gap between real estate and technology. If you haven't already, please join the Facebook group, where we're starting to build a community to share information on data analytics and tack within real estate. This will help us as real estate professionals and entrepreneurs to get the latest insight to help grow our businesses. All right, let's get right to it. The first article that I landed on was City's lost population in 2021, leading to the slowest year of growth in US history. Now for many of us, this isn't new news, right? We've seen due to the pandemic and the height of remote work, that a lot of people have left very expensive cities, including myself to move to the Sunbelt area. So those that are south, something you may not be aware of is the impacts that it's had. So for example, one of the experts say that skyrocketing housing costs were to blame as well as American demographics that began before the pandemic, such as the steadily falling birth rate and steep drop and immigration. So these were also factors that contributed to people leaving cities. In addition, some of those cities that saw some of the largest drops were New York, Los Angeles, Chicago and San Francisco. Whereas according to the census bureau, cities that saw a big uptick were Phoenix, Arizona, Houston, Dallas, Austin and Atlanta, gaining more than a total of 300,000 residents. Large metros had a large decline, and then smaller, metros had a large push to increase. But in terms of rural areas, they did have an increase but not as impactful, right. So people are leaving large cities, but they're also going to cities, again, just going to maybe smaller cities. So some impacts that this has, is a population loss for these wealthy cities, has potentially less taxes, right, so less people living in it less, were able to pay taxes, and this can cause funding shortages for schools, health care facilities and other services. Because of this, I really do believe that the whole idea of we need to live or rent or buy in a neighborhood that has great schools is eventually going to go away. Right? So we're going to have the schools have less funding and what are some ways to compete, maybe the schools are able to go fully online and get students from impoverished areas or rural areas to join their network in their school. So I really do think that being local to a good school system is going to dissipate at some point in time. And one of the questions here was, did the cities have it coming and I think they did even before the pandemic. We have an example here of two people who moved to Atlanta from California. And it says that they spent nearly five years of living with roommates to save money on rent. Most recently, they had paid $1,500 a month for one bedroom and a three bedroom 1200 square foot apartment that they share with two friends. This is mind blowing. You're talking about people who went to college, have professional careers, and still after years cannot afford to live on their own because of these ridiculous house prices and ridiculous costs in these cities. So I really believe that even with a pandemic, wasn't the catalyst, it's something else would have happened to really showcase that this is ridiculous, you know, just to live in a city to have to pay these astronomical costs. Another piece that was really noteworthy from this article was that forecasts for the future there will be too few working age adults to support a growing population of aging baby boomers, jobs in nursing utilities and other fields will go unfilled regions where state and local governments do not make it easier to build affordable housing will face trouble. So for us real estate investors, I think it's really imperative for us to follow local laws understand,

do we see single family home zoned areas turning into multifamily zone? It If that's the case, we can build at us as Matt, the lumberjack landlord has said and has a lot of experience in this will allow us to possibly have higher return on our investment. So these are definitely things to look at. And I think even if we rebound completely from this pandemic, these lasting effects are going to be with us for some time. On to the next part a little more technical, real estate house price prediction using data science, this was an article that I found on the medium it was using out of United States. So other datasets, have housing information, and looking to use Python and machine learning to actually help calculate the price. So this particularly was using some Zillow data, and the area was Bengaluru house price data, you can actually download this data from Kaggle. If you want to give it a shot yourself to try to predict some of the prices, I'm not gonna go too deep into it just because the code is pretty self explanatory. But I highly suggest if you want to get started in data science to check this out. Now moving on to something that I found very, very interesting was how to build effective and useful dashboards. So whether you are in a field like you're a data analyst, or data scientist, or maybe you're a high level, you're an executive of some sort, it's important to be able to have dashboard that summarize information so that you're not going line by line rows in Excel and trying to figure Hey, do I need to pivot what I'm doing? Or should I make a move? So dashboards are very effective. But if they do not tell the story across the correct way, they are not effective. So this blog basically goes into how do you tell the story from a dashboard? How do you take all of this data, all this code that you have all of these charts and put it into place? Where anyone if an organization can quickly make decisions here, they go through several steps, step one, ask the right questions, this is completely important, you can't just create a dashboard out of nowhere, you need to make sure that's actually going to be useful for the consumers, then draw the target. So specify what metrics you want to track. So say in the case, if you're doing, you're an agent, and you want to increase the amount of conversions you have on leads, knowing the number of leads that you're getting in your funnel per month, and within each sector. So maybe you're getting some leads from Facebook. And if you're getting leads from newsletters, you want to track what those numbers are, then also track conversion rates at the top. So then the second part here, the level of aggregation, do you want to see it monthly, weekly, daily, hourly, our think hourly is way too little for a task like an agent. But you probably have enough data month over month to really make the case to look at those metrics. And then there's also a time horizon, should the data be compared to last year's figures last quarter, and say, if you're an agent, I think probably looking at last quarter would be effective, as well as year over year deal with seasonality, right. So say if you're seeing your conversion metrics tremendously dropped month over month, or maybe you're going from October to November. And in November, a lot of people are focused on the holidays or sales are going to dip, there's going to be less interest. And that's normal. But it may look like there's a catastrophe happening on your dashboard. So if you look year over year, you'll be compared to that same seasonality timeframe. Therefore, you wouldn't see likely that drastic shift, this article goes on to actually show an example of an effective dashboard, which I completely agree with. At the top there should be some quick highlight metrics. And then underneath that be some interactive charts that you can then utilize. Some ways that you can create these dashboards are using tools like Tableau, mode analytics. And if you're a little bit more savvy, you can actually build them out using Plotly dash or are shiny. Now going over to the last piece, which was how to extract data. I have time and time again, stated different API's to use if you want to programmatically get property details, say if you are searching all homes for sale in Tampa, Florida, and you want to calculate what your return on investment would be. Now if you don't know programming, it will take very long time to copy all for sale properties that are on the website, put them in a spreadsheet and then calculate something based off of it. Now a programming you can actually get that data because web scrapers have already scraped Zillow and provided that to be publicly available through API's. So zillow.com is probably my favorite. That's been up for about a year now. So I don't have any hesitation that it's going to be removed anytime soon. But they basically take property details that are often those pages and make it available. So if you're looking on YouTube, you could see in this video that for the property extended search, we could see all properties. In this case it is for Santa Monica, California, that are up for sale. And if we go to property details, the sample is that we could see things like the Zestimate how long it's been on Zillow, nearby homes page view count, how

often have people looked at this home, description of the home, and even more. And if you want to get details for a list of properties from this API, but you don't know how to code in Python, no worries, I've actually created a tool where you can upload an Excel file of just a list of property addresses. The tool itself pulls the data from this API, and then provides you an Excel spreadsheet with all of that information. So I hope this has been useful and very informative. Before I close out, I'd like to start sharing some of my weekend plans since that's one of the funnest part of it being Friday. This weekend hoping to hit the beach. That's one of my last speech days in Tampa since I'll be gone for four weekends in a row tending three weddings, which be a lot of fun, so can't wait and thanks so much. Hope you have a great weekend as well.