SmartAsset (YC S12) Helps With Tough Financial Decisions Like Homebuying | Hacker News Reader

SmartAsset (YC S12) Helps With Tough Financial Decisions Like Homebuying

techcrunch.com

For housing, how does this compare/compete with NY Times' Buy or Rent calculator?

http://www.nytimes.com/interactive/business/buy-rent-calcula...

The NY Times Buy or Rent calculator is a great tool (and probably one of the most commonly used).

However, it has certain flaws: for starters, it assumes you know exactly what tax you pay, what mortgage interest rates (and fees) you qualify for, what your property taxes will be, etc, etc.

One of the key features about SmartAsset.com is that we do all that for you, and then let you tweak the numbers if/as you wish.

We also pull in live mortgage data (along with a number of other datasets) to compare various buy-v-rent scenarios that actually apply to you.

This is also only the beginning of the product. Apart from comparing buy/rent, we want to ensure users understand how much they can really afford (and why), how taxes will change over time, how their credit-score affects their ability to borrow money, etc.

Beyond that, there's also managing the process that goes with any major financial decision (what we like to call the "interest, learn, shop, buy" steps.

It would be interesting if you could hook up with a site like redfin or whatnot to embed that on their site so when you are looking at a house it automatically fills in all the data it knows about.

@icefox: that is exactly on our roadmap. We want to (a) be able to show users what they can afford in their selected (and neighboring) regions; and (b) to get all the right information for specific properties.

We're coding this right now, and should have it released this summer.

Redfin already has property taxes and mortgage rates, it just doesn't bother to put taxes in the calculator, for unknown reasons. Doesn't have the credit score stuff, but that's not really relevant -- if you have poor credit, buying a house anywhere near your affordability boundary is likely a huge mistake anyway.

NYT does a good job relative to most calculators but it is still a calculator and consequently limited.

The big differences then are accuracy and ease of use. On accuracy, there is a difference both in how the numbers are calculated (we use our financial modeling back-end) and the quality of the assumptions used - we have local data sets for transaction expenses, real estate taxes, income taxes etc - NYT uses default values, like 28% for your marginal income tax. This means they end up overstating the tax advantage for lower incomes and understating it for higher incomes). We are also much easier to use, precisely because the user is not relied on for all of the important assumptions - we know your tax rates, we know which mortgages are available to you etc.

Hope this is helpful. Happy to answer any other questions.

Love how easy and straight forward it is to use. But I wouldn't have understood these advantages of using smartasset that you outlined above just from landing on your site. Maybe consider outlining this somewhere on your site before I think of it as the same as the NYT calculator and the zillow and the redfin offerings too?

Beautifully done for sure, looking forward to more from you guys!

There's something strange with the math in general, and going from 29 years out to 30 specifically. I entered:

Income: 100k, rent 1250, other expenses 1250.

In year 29, the expenses on the owning side are $1,017k. Go to year 30, and they drop to $405k.

Also, I can't figure out where Net is calculated from. It seems like it should be Income + Home Equity - Expenses - Tax - Home Payments, but that doesn't add up.

This plus the fact that you assume non-home investments do no better than inflation makes this pretty questionable to me.

@ianferrel -- thanks for pointing that out (I'll check what's happening in year 30 of the model).

As for the interest rates on non-home investments, right now we are making an assumption (that cannot be changed). However, that will be configurable (at a fixed rate, or against a curve) very shortly.

Please let me know if you have any other issues or comments/suggestions.

thanks, Phil.

Ian, we'll take a look at your specific financial profile. I can say, that in each year, calculating your net wealth is the sum of over 90 line items (ie not simply income + home equity - expenses etc). In the future you'll be able to change the performance of non-home investments to match your view.

It's nice that you try to be inclusive with "partnered", but then you throw it all away by not supporting the likely tax filing status of gay couples: joint for state, not married for federal.

Honestly, that's the first question on the page and it looks like it was designed by some committee trying to find a compromise that offends the fewest people, with no concern for whether you're using the most common terms or whether your model actually fits the real world.

@kevinpet -- you are spot on there. We have discussed this several times internally, and feel very strongly about including this. However, we simply tried to get the most done in the shortest period of time possible. More features (like this) are on the way shortly.

The issue with LGBT couples, or un-married partners, is simply complexity. LGBT couples, for instance, can file joint taxes in certain cities/states, but then separate federal taxes, applying or not applying various deductions, etc. It gets very messy!

Stay tuned for more... and feel free to send us any comments/suggestions you have!

Tax options of "not married" and "married state, not married federal" would probably capture all of that. I'd assume that state and local are always identical.

I was confused by "partnered". I took it to mean specifically domestic partners and would have been confused if I were trying to use the tool if I were unmarried but buying a house with my girlfriend or college buddy.

That stood out to me as a specific tweak. Generally, I'd like to see the ability to change some numbers. For example, you recommended that I can afford a house about 20% more than what I actually feel comfortable with, so I'd be more interested in comparing rent vs. buy with the lower house value I'd like to use. I'd also use a VA loan over a FHA due to better terms (no PMI), but there's no where for me to tweak those values.

Although, if you feel like you need to handle same sex couples properly, I should note that there are many times more couples eligible to use a VA loan than same sex couples.

At this point, I would not use the recommendations of your tool. There isn't enough information about what assumptions you make, like how you calculate property tax and whether you are bundling in insurance into the part you assume to be an income tax deduction.

I'd also like to see expected case home price changes, inflation, and investment returns, along with the combination of likely range of returns that would push rent vs. buy in different directions (i.e. good investment returns with low inflation makes renting better, high inflation makes buying better).

Awesome feedback, definitely feels like the direction we're heading (greater user flexibility) is one that resonates with you and others. The only note I'd make is that you can currently change the home value on the buy vs rent tab (it is beneath the other data entry tabs on the left of the page - clicking on that or on the button that says 'change' underneath the home value should get you going).

@kevinpet -- totally understand where you're coming from, and we've received several similar comments today. To date, we've been trying to "mask" a lot of the complexity as well as the numerous variables that go into the model.

However, clearly a number of users want to see more details, along with an explanation of assumptions being made, and the ability to tweak and change these assumptions.

We hope to add these to SmartAsset.com very shortly.

Between the 0% real rate of return on investments, 0% cost of selling your home, no HOAs, this seems to low-ball cost of ownership by at least 10%. Couldn't even come within 10 years of the NYTimes rent vs buy with my assumptions. Not to mention that this uses a 4% mortgage rate and then offers consumers 4.65% loans.

But when putting people in loans in the only way to monetize, it is going to be hard to give unbiased advice.

Hi, looks like we need to make our assumptions clearer. We assume an approximate 6% cost on selling a home (changes regionally). You will be able to flex your ROI shortly.

Not sure I understand the point about 4% vs 4.65%. If you see a rate of 4% on our site you can execute at that price. If you have problems with the bank, let us know and we'll go to bat for you but everything on our site is actionable.

From the buy/rent calculator:

    RETURN ON SAVINGS: We assume the money you have saved
    generates a return of 2%/year.
    INFLATION: We assume a general rate of inflation on 
    your income and expenses of 2%/year.

This seems unrealistic. What is the basis for these numbers? It seems to bias the buy/rent calculator pretty heavily towards buying.

Which part? Expected return on savings or expected inflation? Relevant to inflation expectations: http://krugman.blogs.nytimes.com/2012/07/14/is-our-economist...

we're working on giving users the flexibility to change these assumptions. Obviously viewpoints differ and we want you to be able to make the best decision based on your view. That said, all default or base case values are what we consider to be 'consensus' estimates.

It would also be good to be able to change the 'expenses' field for ownership.

For example, if you live in NYC there's pretty much no chance you'll find a place which doesn't have (at least) a few grand/month in "coop fees" or "condo fees".

Also, there's "how much will you spend on maintenance", which you don't need to worry about when renting. Replacing a roof, hot water heater, etc. I'm sure you built in some numbers for that, but it would be good to be able to set.

And, as other comments have said, the 30 year mark seems to just be broken. I filled in some random numbers (100k, 60k) and I get 339k expenses in year 29, and -125k in year 30.

I like it. One "bug":

I get a $50,000 reduction in the amount I can afford if I go from $0 debt to $500/month and $15,000 total outstanding debt.

Okay, maybe that makes sense since I'd have 2.5 years of reduced income to make payments with.

But, if I reduce the total outstanding to $500/month and $500 total outstanding, that $50,000 reduction doesn't change.

Edit: One more thing: on Topic 4, the interest rate and minimum down payment percentage do not seem to change when I change the FICO range. They are both 3.25% no matter which score, except for the lowest range (which states I couldn't qualify). Also, if you can find me someone who will give me a 3.25% loan with 3.25% down in California (I have near perfect credit and strong numbers), I will literally pay you $10k cash on the spot :) Maybe you should be pulling actual rates/amounts from what people have actually been able to get in certain areas? (You could pull this info straight off the recorded TDs).

Otherwise, really cool tool that's relatively easy to use and looks nice!

@liber8: thanks for pointing that out! Will get that fixed.

On your point about our credit score page, we are working on a better way to present this. There are circumstances when a lender is will to offer the same or similar terms across credit bands because it is insured by the FHA.

FHA mortgages start at 3.5% down payments. Our decision engine only shows actionable opportunities (so we've made sure your income is in the right range and that the loan is within the limits defined by the FHA for your zipcodes). So providing your data inputs are accurate, we can help you get that mortgage (and help you find a charity to donate your $10k to). info @ smartasset period com.

The decisions listed (buying a house, retiring, paying for college, etc) are all fairly infrequent, separated by many many years - so not much value from return users. I think getting $$$ from people for that is going to be very very tough unless they can quickly expand and become the "financial decision-making engine" for pretty much every $500+ purchase in your life.

If that's not in the cards, they must be thinking of (A) targeting intermediaries (e.g. mortgage people at banks to use with their clients), or maybe (B) planning something a-la-Mint (e.g. present you with relevant offers to whatever it is you're considering buying/leasing)... or maybe something else?

Curious to hear how you guys are planning to monetize...

the obvious one is about three comments down - a guy in the UK wants to join up and use the engine. I suspect that most mortgage providers would be happy to have you play with a "independent" site that shows you how much richer you will be in 20 years if you buy their mortgage, and price comparison sites would love to show you which mortgage will pay off most in 10 and 20 years.

That's probably worth following up - is there a way of putting multiple mortgage lines in there ? I assume its doable.

Edit spelling

This could well be problematic under UK financial regulations. You are giving financial advice, which is a regulated activity and as far as I can see hard to do with automated services under the existing rules, for personalised services like this (and also if you charge). Most financial advice is from journalists who are basically exempt from the rules, or are generic not personalised.

The situation in the US is fairly similar I thought, would be interested in the legal situation there, as I think this type of service is very useful for people.

we intend to build a platform for all personal finance decisions. the home purchase decision is just the tip of the iceberg and even here we are just getting started. In 18 months we intend to be able to provide personalized, interactive answers to over 200 questions, on topics from car buying to going back to school, to family and retirement planning.

Buy/rent calculators are definitely a great tool for homebuying assistance but we need more things like how what area to choose based on preferences (suggest me possible areas based on my preference?), should i really buy even if financially feasible (single vs. married couples etc), what is my possible job outlook or status in next 3-5 years, do i expect to move soon? if an app can take in all these factors along with financials, it could suggest something close enough.

some of the above is already available (eg we model taxes for single and married couples so you can flex that in the assumptions - also most of the data we use is local to the zipcode you are searching in) and some of it is on the way (like contextualizing our advice with the type of home you can afford and more qualitative neighborhood info).

Couple of questions

Is this Net Present value ?

Do you publish the calculations that got us there (I could not see it)

There seems to be some sensitivity to a few key variables (down payment?) - is there a write up of the algorithm (in plain english :-)

But its pretty good - I just chucked in my UK prices and guesstimated equivalent rental, but its quite a shock - I expected to see rental as a more expensive option much much earlier and by a wider margin. its 20 years before you start saying - well thats worth it.

Buy vs Rent chart is not net present valued... not sure if that was your question though?

It was one of several questions - sorry I tend to do that

if it is not net present valued then surely the calculations are off - won't that push the ownership beats renting even further out in time?

Also I did not see if it was interest only vs repayment (one of the biggest issues IMO)

Hope this helps. When Apple says their 2015 forecast for sales is $200B (made that up) that is not a number that is PV'd (because that is 2015 number). Similarly when we are modeling your income going forward those are not numbers that should be PV'd.

However, when we sum future expenses or cash flows (across years) and present them as a single number, generally in today's money, those values are all PV'd. For example, on our mortgage comparison page the total expenses over the amount of time you plan on holding the mortgage is a number we PV.

The design of this site is really impressive. If you're working on a web product you should be taking notes.

Probably the #1 problem for a brand new site offering financial advice is to establish credibility. Lots of past entrants did a lousy job of this, but these guys have nailed it on day 1. Their site looks a lot like the NYT's info features, and it helps them look authoritative.

Thanks Phil. It's been quite a tough job keeping things looking clean/simple, yet providing all the content and info needed. And I believe we've still got much to learn and improve on...

However, kudos to Michael Calleia -- our UI/UX designer.

Would love to collaborate on a UK version of your Rent or Buy calculator for our users at OpenRent. Saves us writing one from scratch when you already have something so functional!

Unless of course someone knows of a decent one pre-calibrated for UK taxes already?

Awesome. Lets speak. info@smartasset.com.

Looks like it will be good.

Seems to go very wrong when you can afford very high down payments.

Also tricky to use sliders on iPad.Would be good to allow direct numerical entry as option.

It should offer an existing home equity option with the available cash and gifts options.

I hit some kind of bug in the UI, after a while adjusting the rent/home appreciation percentages didn't update the display properly.

Can you hit us up on the Olark chat on the page? We'll see what happened and get it fixed asap.

Useless to someone with no income but substantial savings -- it refused to get started. A niche group, of course...

@abecedarius -- quite. We've tried targeting the general home-buyer (who usually put 3.5% - 20% down, and takes out a mortgage to cover the rest).

The decision mechanics for someone in your (or such a) situation are completely different, right?

I honestly don't know -- it's unpleasant to sort through all the info and advice from sources who all want to sell you something and I end up bookmarking a few things and putting off any decision. It sure could be lucrative to be the kind of person who gets as interested in finance as I do in, say, compilers.

Good luck with your startup.

Def seems to have a pretty deep level of analysis behind it but the interface is nice and simple; well done guys.

thanks for the feedback... glad you like it!

This looks great!! Can't wait to see it develop further. No doubt it will go far. Good luck and keep it up.

This looks great - I've always been curious about the financial impact of buying vs. renting.

Thanks! That's exactly one of our objectives. Plenty more to come over the next few months...

The only problem with a calculator like this (intelligent as it is), is that it doesn't model the most significant cost of home ownership - upkeep. Compound that with a complete inability to model the risk associated with "home price appreciation" and it's really only of academic interest.

In my experience this decision cannot be made empirically - home ownership will probably be more expensive, but will afford more freedom and stability (no one will sell the property out from under you and disrupt your family). In addition, like any investment, there is a possibility but not a guarantee that the asset will appreciate; however this is not a given - look closely at the economic future of your location and try to decide if there will be more or less demand to live there in twenty years - the Bay Area will probably be safe :)

There's also a significant opportunity cost attached to a down payment.

Don't get me wrong, I think SmartAsset has done a phenomenal job with this calculator! Just keep some of the intangibles in mind that are impossible to model this way before making the decision to rent/buy for yourself.

Just to be clear, opportunity cost and maintenance / upkeep are both modeled and incorporated in the values you see on the buy vs rent chart.

There are also several components to the decision. Even if you give greater weight to the emotional rather than empirical factors (as many do), it is still prudent to make sure you can afford your home.

And yet, like most (all?) calculators, there is no questioning about the houses age, condition, pending/expected costs, etc. Those are things I've always needed to tack onto the calculator results to get a better answer.

Edit: Sorry, I think my message's tone should be better. I enjoyed testing the site but while looking for a home last year it would have been nicer to have more control of how these calculations were done.

Interesting concept... Looking fwd to see you develop the product further.

This is a great idea. I hope you guys go far.

Thank you!

Astray: Will the be sued by Vitamin Water?

I'll give you a piece of advice (take it for whatever you want) that you're not likely to get elsewhere.

People aren't stupid.

The dominant model in many online ventures is that in some sense the "user is the product", which they're selling to companies. There will be a huge natural push on you to sell a part in your user's "decisions" - you need to resist this, and then you will be ahead of 99.97% of web sites in your sector.

The term "Your trusted advisor" is all but meaningless online. Make yours have meaning. Then, you are not competing with the rest of the Internet (well, maybe with 0.03% of your niche) but instead with real trusted people, like financial advisers.

There is a lot of money there, and the Internet isn't getting any of it, because whereas people (rightly) pay for people in real life that they trust, people (rightly) don't go online for the same thing.

Are you familiar with mint.com?

thanks for the thoughts. cannot stress enough how important it is to us to build a business and product that people have confidence and trust in.

Many people would love to get a second opinion on their finances, and they like to come into meetings already having a certain amount of information on the topic.

I am a heavy user of mint.com and I could absolutely see this site being very handy.

It never hurts to get a second opinion.

Reply on news.ycombinator.com