Interesting, but seems to just look at the US landscape. I think the author would be pleasantly surprised if he broadened his field of vision not just to other developed nations but in particular some African countries to see signs of true consumer finance innovation.
I can't think of anything in software lagging as much as personal finance. I use GNUCash, which is a glorified balance sheet, and my online bank accounts. This sucks, but only slightly less than Mint which sucks slightly less than Quicken.
Making personal finance decisions should take seconds, not hours, and yet data collection by hand remains the only reliable method to make such decisions. This is unacceptable in 2012.
Here in the UK none of the banks seem to let you pull out your personal finance data, bank statements etc as a raw data format and use it that way.
For example, I do 99% of my grocery shopping at about 3 shops. So if the data was available it would be trivial to write a script which totals the amount I have spent that month on groceries and graph that against previous months.
The only way I can do this is to go into the banks website and manually add these figures together every month. This just seems like a terrible inefficient way to do it.
It would be great to be able to just go through my bank statement and say "oh , Steam. That's an entertainment expense, so in future put that into a bucket called 'entertainment' so I can see what that is as a percentage and track how it changes". I could also see blips in things like quarterly expenses (utility bills etc) and plan for them accordingly.
What would be truly awesome would be to apply some stats and ML to it and say "If I make large purchase X for £Y in August what is the increase in probability that I will fall into arrears come Christmas based on my historical spending patterns"
If I was feeling cynical I might think that banks don't really want me to plan my personal finances too well. They would rather I mess it up now and again and have to apply for credit.
I actually solved the "get the data out from the homebanking system" via 100 lines of javascript executed in phantomjs to screenscrape the site.
It isn't pretty, it might break, and I haven't looked into how to store the credentials "safely" (so it executes just when I call it and enter the u/p, and not in the background), but it beats the hell out of logging in and extracting the data by hand.
You might get some mileage of this, if you want to automate the data extraction and have an homebanking account.
I actually did consider this option and technically it's probably not too difficult.
On the other hand I'd hate to get used to using this and then have them do some obfuscation deliberately to stop me.
I'd also be somewhat paranoid that if my bank figured it out then they'd come down on me like a ton of bricks and use some convoluted law to freeze my account or send me to prison for "hacking the bank" or somesuch.
Just want to put a shout out to a company run by some good freinds of mine, Pocketsmith [1]. They've got a really good product that is by far the best that I've used (the ones I've been able to use - Heaps, Xero Personal and a couple of bank ones here in NZ all suck). They're linked up to yodeley as well, so they get bankfeeds from all over the world.
Sure but whats the killer app? What can the internet improve significantly? What is really useful to an average consumer? Not someone who is just looking for ways to spend time on a computer.
10 comments
[ 3.2 ms ] story [ 38.7 ms ] threadMaking personal finance decisions should take seconds, not hours, and yet data collection by hand remains the only reliable method to make such decisions. This is unacceptable in 2012.
For example, I do 99% of my grocery shopping at about 3 shops. So if the data was available it would be trivial to write a script which totals the amount I have spent that month on groceries and graph that against previous months.
The only way I can do this is to go into the banks website and manually add these figures together every month. This just seems like a terrible inefficient way to do it.
It would be great to be able to just go through my bank statement and say "oh , Steam. That's an entertainment expense, so in future put that into a bucket called 'entertainment' so I can see what that is as a percentage and track how it changes". I could also see blips in things like quarterly expenses (utility bills etc) and plan for them accordingly.
What would be truly awesome would be to apply some stats and ML to it and say "If I make large purchase X for £Y in August what is the increase in probability that I will fall into arrears come Christmas based on my historical spending patterns"
If I was feeling cynical I might think that banks don't really want me to plan my personal finances too well. They would rather I mess it up now and again and have to apply for credit.
It isn't pretty, it might break, and I haven't looked into how to store the credentials "safely" (so it executes just when I call it and enter the u/p, and not in the background), but it beats the hell out of logging in and extracting the data by hand.
You might get some mileage of this, if you want to automate the data extraction and have an homebanking account.
On the other hand I'd hate to get used to using this and then have them do some obfuscation deliberately to stop me.
I'd also be somewhat paranoid that if my bank figured it out then they'd come down on me like a ton of bricks and use some convoluted law to freeze my account or send me to prison for "hacking the bank" or somesuch.
[1]http://pocketsmith.com