All Forum Posts by: Steven S.

I made a community-based algo sharing site, and someone already got over 96% accuracy on the properties this week, in 53 lines of JavaScript!

If anyone else is interested, you can test, refine, and get your home-value algorithms ranked weekly (for free) at: runcomps.dev

Lowers the barrier to entry to just coming up with an algo rather than building the entire db and back-end for it all to work on :)

lmao good read! Hope you all can resolve it

Quote from @Steven S.:

The % of ARV varies by market/area. I'm in Los Angeles, and the most I can pay on flips is typically 60-65%.

You want to get recently sold comps to value the property, it's the best way to be sure of the value. I usually use Redfin, but Canada isn't supported. Checked Zillow, you are right, only a few places have results in Canada. Perhaps there is a realtor association that has this information? Out here we call it the MLS, Multiple Listing Service. I'd try to call a few agents and see if they can run sold comps.

Once you have access to sold homes, there's nothing to get caught up on. Just go through them, select the renovated/closely matching ones, and enter them into a model.

Here's my fix & flip model, you can see the comps entered and some quick stats below them:

Then I use that comp data to adjust the Sale Price in the model to what I estimate it will be (double check the resale $/sf is in-line with the comps), then adjust the purchase price until I hit my desired return level. 

Good luck!

This property I previously modeled is now active for $1.789M, if you check the model I was at $1.75M. Models are the way to go!

https://www.redfin.com/CA/Los-Angeles/8112-Airlane-Ave-90045...

I'm interested! Will it work for properties in Southern California? It has to beat app.runcomps.org though, that's what I'm using now.

Let me know what ARV estimates you get for the following properties:

And that ARV estimate above isn't a one-off either, varying the radius shows different values for this just-sold renovated home:

This is cool! Although what if you will add ADUs/Units to the property? Or add 500sf, etc? What if I fix & flip instead of rent, can it pull actual sales comps and group by renovated and estimate the ARV?

Seems very locked-down to Zillow listings in their current as-is layout. We only buy SFRs to rent them if we can build it into a 3-4 unit property, so the main house is just 1 out of 4 units to comp out. Plus you still have to actually underwrite the entire deal (purchase, construction, carry, etc to develop accurate CF/profit). The main house might be upside-down if rented, but not if you build 3 more units, etc. I always say 'I don't care what cap rate I buy at, I care what cap rate I can turn it into'

I built Market Analysis and Sales Comps tools for my local SoCal market that will let you analyze the current or terminal property & units, from wherever you get your deals from:

Sales Comps & ARV estimate:

Sales market analysis:

Rental market analysis:

Would love to discuss how to make your extension that much better!

Have you chosen a tool yet? I'd say forget the AI buzzword hype and go back to programming something useful based on real transaction-level data like below (that can help you with any deal you find, from whatever source you have).

Sales Comps & ARV estimate:

Market Stats:

AI won't be able to provide you anything close to this, nor does the RedFin/Zillow market data drill down into the market as deep as I do here.

Additionally, if you understand how the leading LLM's work based on a black-box floating point matrix, you just can't trust their conclusions like you can with your own program. Go ask ChatGPT or Claude or Perplexity what the current and historic quarterly $/sf in a City or Zip code is. Then compare it to my data-fed program. You'll quickly see why AI is not the solution here... not only will it give you a random number from a site that posted it, but it will hallucinate and draw quirky assumptions into the mix like oh based on this % appreciation and applying it forward/backward, etc. Totally useless for making go/no-go decisions, just like the average-only market stats from RedFin & Zillow.

Stay away from AI, demographic data, and average-only market data/stats if you want to "make a data driven decision towards purchase", they are all too broad/meaningless to actually help make that go/no-go decision. You will have to do something unique to make it effective & useful to you... because it's always no pain, no gain!

Would be happy to hear your thoughts/feedback, especially if you've been trying out different tools.

Quote from @Adam Jones:

Hello Bigger Pockets world,

I am new here but been investing in RE for a few years now. 

I am looking to switch CRMs and was wondering if anyone has REALLY DIALED IN their workflows in REI REPLY that they would be willing to share. I am sure someone has worked out the best workflows, triggers, messaging etc.... It's just a question if they are willing to share them.

Sure I could build these out, but I am much better at buying houses than I am setting up CRMs and am hoping someone out there would be up to helping another investor out. 

Thank you for even considering helping a fellow investor!   

First, don't use REI Reply, it's just a white-labelled/private-agency version of GoHighLevel (the actual provider of the CRM) that they just mark up and charge you extra for everything you do (every text, email, etc is 2-10x more expensive than just linking twilio/mailgun to your GHL account, etc).

With that out of the way, you can get to making workflows. Either you or someone you pay has to build them, no one is going to send you their workflows/automations/marketing campaigns that work for them because that's their competitive advantage. It takes actual thought and a programmer's mind to be able to build a CRM into something that's actually effective (sure a basic person could setup a phased text message or email automation, but it's tying everything together with opportunities, automated campaign re-entry/follow-up, social media integrations/automations, etc that is the real value). Building a Phase 1-5 messaging campaign is the easy part.

I'd recommend you build this yourself, so you can see what's working and what's not, and improve it over time with added features, automations, and complexity. No one will ever be as motivated to get results out of it than you are!

If you want to hire to have this built out and are willing to either pay good money for it or agree to 10% profit of anything that comes through it, I'd be happy to help build it out. The latter is what a typical marketing agency would charge, and they would provide and build you the CRM for free.

Quote from @Bruce Lynn:

@Liam Alvarez  I've never seen an app or a website that analyzes all markets and spits out the top 3 you should invest in for the highest returns.  If that is what you are looking for.

You have seen it now:

It takes a good second to understand what's plotted in the scatter plot, but this is my version of that. Has every market (by city or by zip) in SoCal, which is where we fix & flip & build rentals.

I'd recommend more drilled-down market stats like the bottom graph for validating your underwritings and making go-no go decisions, but it's still interesting to visualize & pick-out those super-performant markets.

To be fair, I haven't seen something like this either, that's why I built it!

I do fix & flips and build 3-4 unit rentals in the SoCal area. It's super competitive here and we have a large flow of off-mkt homes coming our way, so we had to develop tools to speed up acquisitions/underwriting.

I built something that allows VA's or very green staff members the ability to properly distinguish opportunity as if they had years of expert experience & local market knowledge.

It will run sales comps & estimate ARV for you:

And has drilled-down market stats that are actually useful! The top chart below quickly shows you what the renovated, avg and distressed homes sell for in that area on a $/sf basis over time, so if you are estimating way over the top line, you instantly know something is off, maybe you have an outlier comp on your hands, etc.

It doesn't have STR data, because any STR we build must also pencil as a LTR to avoid being upside down if it doesn't workout, so we use normal residential lease data when underwriting our rentals.

As for forecasts on the market, I don't do any aside from the current Q trending % at the footer of the charts, and it's really just the current Q's average data/performance held constant and applied to the rest of the quarter (since not fully done with Q yet). This yields the % up or down we are in the current Q relative to the last fully closed Q, so as the transactions roll in each day I get an idea of the current Q's direction.

I'll always advocate for building your own tools so you can build them to suit whatever your development schemes require for the best experience, efficiency, and confidence. This one only has SoCal data, but if you end up looking at deals out here, use it! Hope this gives you a good idea of how these tools help us. Would be glad to hear suggestions/feedback on it as well if you use it!