All Forum Posts by: Steven McCord

Looking forward to attending meetups!

Fortunately, there's good data out there, but it's not being made easily accessible. At least not without a lot of expense and time-consuming jumping around between sites. We solve the data problem by working with the industry's best data vendors that have applied a lot of post-processing and enrichment to the original government sources, or have collected their own proprietary data.

Not all of those vendors are equal in terms of accuracy, and we learned this the hard way through experience -- I come from the commercial real estate research world, and my co-founder from the retail site selection & revenue forecasting world. We aim to bring institutional-grade research data to the single family home investor. We add to this by considering "point" data nearby that are related to supply, demand, performance, and risk. These hyper-local measurements are what start to tease out bigger block-by-block differences (and changes over time), which are often significant.

I'm developing a software-as-a-service offering for real estate investors to find great locations -- down to the block level -- to invest in single family rentals. It is especially aimed at people looking for better yields further away from home. We take all kinds of location information, mix it together, and give you 1) a detailed mapping of risk & return 2) an early warning on locations that make sense for you.

Some questions for investors who have several SFRs or more:

1. What is the BIGGEST headache you face in trying to figure out "where" to invest?

2. Without giving away your secrets, do you do any sort of analysis to identify locations? Or focus just on whether each deal "works"? Do you want to be able to do better/faster/cleaner analysis?

3. What solutions have you tried and were they insufficient?

4. What professionals do you turn to, besides your agent, to get investment ideas. Your property manager? REI groups?

Feel free to DM me and we can arrange a deeper discussion. I'll give you a 25 dollar Amazon gift card for a completed interview!

I think it's a reasonable idea. There are lots of vacant lots in inner parts of Dallas and Fort Worth, it's a question of identifying which specific blocks are too risky vs ones that are OK. There are many factors to consider. I do this kind of analysis using many types of data.

@Malik M. I know what you mean by analysis paralysis. My approach has been to automate the early calculations using big data to save time, and allow me quickly focus on areas where the numbers make sense. Here in Dallas-Fort Worth, the areas with high yield at market prices have become thinner and thinner, but are not impossible to find.

@Nirbhik Modi I like to answer this question using data, and I have built a system to do automatically identify areas that have the highest rental returns at the lowest risk.

My team built a system to model rents using machine learning based on comps, location, and local economic strength. We take out holding costs specific to each location -- expenses for everything from taxes to insurance to vacancy and a repair allowance -- to get to a true net operating income yield.

In the example below, I filter out everything except "B" neighborhoods, which I have defined using my own risk criteria. It gives me a pretty good idea of that middle ground between too expensive and too risky. Some areas around Balch Springs and Pleasant Grove are good places to start. It does vary block to block, so the data that goes into this is at the block level. The areas the light up in orange tend to be the areas where risks are manageable but yields still high. However, some people prefer "A" areas or units that are newer, in which case the target areas would be different.

I also built a script to analyze active listings on the market and output yield estimates, risk scores and rankings within a few seconds. That allows a very fast way to narrow down the options worth a closer look, based on investment criteria.

I like to combine location selection criteria into a full analysis that will tell me where I can get the highest rental yields at the lowest possible risk as a passive investor. Rent can be modeled using machine learning based on comps, location, and local economic factors. From there I take out holding costs specific to each location -- expenses for everything from taxes to insurance to vacancy. I model this down to the block level and then filter out everything but "B" neighborhoods, which I have defined using my own risk criteria. Here's what I get for the Dallas-Fort Worth area. It gives me a pretty good idea of that middle ground between too expensive and too risky. The areas the light up in orange tend to be the areas where risks are manageable but yields still high.

A 2% deal -- or 24% gross rental yield -- may exist in distressed neighborhoods, where you have an acquisition price below 30k on an off-market basis and combined with some significant rehab work. The level of risk is high on several fronts 1) maintaining stable tenancy and minimizing evictions in a distressed area 2) maintaining what is likely to be a very old property 80-100+ years old, and 3) few, if any, exit strategies exist except to sell to another investor, which limits your ability to cash out. True net yield after all costs, repairs, vacancy, and capex may be a little higher, but at a great cost of time and effort and engagement with the community. I personally am more of a passive investor and look for a good balance of risk, return, time and effort using a data-driven approach.

I am considering developing a web-based mapping platform that will allow single family home investors -- and their service providers/advisors -- to quickly determine suitable locations in the US to focus their investment efforts (down to the block group or subdivision).

I wanted to get some quick feedback on whether you would use such a tool or think that someone would.

My thinking is the target audience is people with no geographic restriction and are open to anything -- the ultimate out-of-state-investor -- who wants to find areas worth focusing on. Of course, there are ways to make money in any market, but I think we're talking about the passive buy-and-hold investor.

The system would take demographic, economic, and sales/rent data to allow you to start from the whole U.S. and then filter down -- to very specific small areas -- that meet the requirements for your investment objective and merit a closer look. It would be done through a weighted scoring system, with weightings and variables pre-configured. And brilliantly visually displayed on a mapping platform.

For example, you are a buy and hold investor and you want an (estimated) average net yield above a certain level, with the lowest possible crime, highest possible school rating, lowest possible average holding costs, reasonable liquidity (transaction volume), and in a metro that hasn't already doubled since the last market trough (i.e. not that hot). The result is a set of locations worth a closer look.

That is something really hard to identify by going and looking at listings with today's listing-specific tools. My goal is to make the process a lot faster.

Is this something that might be useful?

I am considering developing a web-based mapping platform that will allow single family home investors -- and especially their service providers/advisors -- to quickly determine suitable locations in the US to focus their investment efforts (down to the block group or subdivision). 

I wanted to get some quick feedback on whether you would use such a tool or think that someone would.

My thinking is the target audience is people with no geographic restriction and are open to anything -- the ultimate out-of-state-investor -- who wants to find areas worth focusing on. Of course, there are ways to make money in any market, but I think we're talking about the passive buy-and-hold investor.

The system would take demographic, economic, and sales/rent data to allow you to start from the whole U.S. and then filter down -- to very specific small areas -- that meet the requirements for your investment objective and merit a closer look. It would be done through a weighted scoring system, with weightings and variables pre-configured. And brilliantly visually displayed on a mapping platform.

For example, you are a buy and hold investor and you want an (estimated) average net yield above a certain level, with the lowest possible crime, highest possible school rating, lowest possible average holding costs, reasonable liquidity (transaction volume), and in a metro that hasn't already doubled since the last market trough (i.e. not that hot).  The result is a set of locations worth a closer look, that might have the right risk-return balance.

That is something really hard to identify by going and looking at listings with today's listing-specific tools. My goal is to make the process a lot faster.

Is this something that might be useful?