Slashdot Mirror
Developer-Friendly Banks?
tyen writes "Any suggestions for a 'developer-friendly' bank for small businesses? The banking world is awash in data protocols that business customers who are/have coders would find useful, like BAI to extract all the raw data from an ACH or SWIFT transfer. Unfortunately, the ones I have spoken with about this access are still stuck in the Dark Ages of computing; they price the access like only big companies still have the skills to tap into these interfaces. For example, one of the four US banks with a perfect trading record this past quarter quoted us USD five figures for access to several of our accounts via BAI format. Per year. After waiving sign-up fees. Are there any banks out there that have a more progressive attitude about letting small, entrepreneurial developers work with their business accounts in a more modern, dare we say automated, way? With big businesses demanding EFT integration from small business vendors, and globalization rewarding premiums to nimble, lean businesses that automate wherever possible, automating the retrieval of this information (which is not available in consumer-oriented access like OFX) becomes an increasingly pressing issue for the small guys."
If you can access it with a browser, you can script it too.
Unfortunately businesses always ask more for not-so-common and premium features. This is also quite close access to their internal systems and since it's probably 0.01% of customers who need such access, they have to ask more from those who actually need it. Banks have to develop, secure and maintain the system too and it costs.
Sounds like an oxymoron to me.
This ain't rocket surgery.
Banking thrives on secrecy. That last thing they want are outsiders poking around with their data and protocols - lord know what you might find. You best bet would be to hit up the next bank president you come across on the golf course (private - not public). Barring that you could try dealing with one of the online banks like ING or perhaps a credit union?
Truth: If it's not one thing, it's another
... they're in it for the money.
Very few small clients will access their accounts in that way. As such, the service is priced with the large client in mind. They will not drop the fees for a small client because then the larger clients would insist that they drop the fees for them, too. Plus, it's a pain in the butt to administer the "small fish" they might pull in by lowering their fees.
Tell you what... You think this is a brilliant way to make money? Open your own bank. It's actually legal (and relatively easy) in this country. It's a good time to open a small bank, too - people are sick of the big banks. Find a backer and let them know that there's this underserved banking market out there. I'm sure you'll find plenty of takers if the idea is a good one.
But, in general, the main reason a banking service isn't offered to you is because the service isn't profitable enough to offer it to the market you're part of.
That is all.
Roll-out your own credit union for geeks. I'd be interested in a bank with the services you've described. I'm absolutely sick of big banks and their big fees for even minor infractions.
I do not know about stupid bank fees, but I recall that ACH is as you say extremely well documented. And there is a setup for a testing protocol built in. There is a spec book that I imagine is say $100 and freely available. If you were not particular about language or schedule or, what is the word, maybe track record, I figure you are talking $10k to do the client software from scratch. If you want it done this quarter in C++ by a name, say $200k. But no real cost analysis here. This was a long time ago, but I think I called my bank to see what sort of obstacles they would put up. As best I understood from a single conversation, there were no obstacles or fees. I suppose it might be relevant that there were personal relationships. For what it is worth this was a regional chain and if you want the name, email me.
I overheard this conversation between some programmers in my company about BAI, It was pretty heated. What I namely remember is:
Ok! Ok! I must have, I must have put a decimal point in the wrong place
or something. Shit. I always do that. I always mess up some mundane detail.
I can't remember what happened to them... It was around the time that we had a big fire at our main office.
many are free
http://www.fatwallet.com/forums/finance/714617/
Ironically in the security guide section it lists getting software updates as key to your computer's security.
If sharing a song makes you a pirate, what do I have to share to be a ninja?
The Future of Money: It’s Flexible, Frictionless and (Almost) Free/
Basically from what I can tell it sounds like you're going to have to go with a startup type bank/payment service like Paypal which has actually made an effort to open its platform up.
I suspect most traditional banks won't change significantly for at least a few decades. My bet would be on ING and its brethren to open up first, try talking with them.
Five digit sums for remote access, per year? Hell, it'll cost half of that just for the security software licences, let alone administration overheads, hardware, networks... Corporate data exchange is not cheap, it's not easy and it's not something you throw together in a hurry for a bank relationship.
Most banks offer corporate customers the ability to manage their own accounts. This includes web or fat client deployment, download of files/reports (including transaction histories and balances) and submission of payment mandates.
Use the interfaces already provided, and shop around for the cheapest bank. Nobody offering the interface you want at the price you want? Deal with it.
Or start your own bank. You too can put up with the horrendous regulatory framework, the stunning liabilities resulting from membership of various payment schemes and the complexity of managing multiple payment mechanisms for multiple customers in a timely and (above all) accurate manner at lower cost than all of your competitors.
After all, you think everybody else is clearly doing it wrong. Go for it, show us how to be better.
That would be true for HTML. Unfortunately, many banks have flash-only access.
Not really. That %$@%$# bubble will catch you.
Disclaimer: I write financial software for a living.
First, I don't see why OFX can't be used for that purpose. You could manage several hundred accounts, payroll, billpay, collections, wire transfers, funds management, etc. Not only that, it's two-way. It's not just displaying account data, it allows you to perform the actual transactions. I know of some payroll processing centers that use OFX for exactly this - either it goes to printer or it goes electronic.
Second, because there is no salable demand for individuals requesting the raw file formats for the backend transfers, those features don't exist. This is common sense - what motivation exists for a company to spend the time and effort providing a feature if there is no money attached to it.
Third, certification. There is quite a bit of hullabaloo in the banking industry about certification, and they're serious. See, there's not a lot of security in the banking world. They rely on hard connections, network separation, and effectively, trust. What they DO have though, is auditing trails. They might not be able to stop a fraudulent ATM transaction, but they can tell you every node, clearing house, third party processor or financial institution it went through. Certification is the thing that allows them to reasonably trust members in their transactional world - you can't just show up with homegrown software and hook right in.
Last, as you said, "The banking world is awash in data protocols". Lemme tell you something - the raw protocols are only about 1/20'th of what goes on. No protocol is perfect, and many systems have what you or I might consider 'undocumented features' that are handled by clever manipulation of the protocol (aka, hacks). The paper description, for example, of the ACH file format can be compressed into about 6 pages. There is a two volume set of books, each about 300 pages, of small print, so thin as to be nearly transparent pages that actually describes how those 8 pages work.
That's one protocol. There are dozens of major ones, and additional complexities when you add in feature specific cores or sinks (back end systems that banks use to store the actual data on, like those provided by MISER or Fiserv )- do you support overdraft protection, provide memo services (a hold against an account for an amount prior to it's actual processing), and if so, which of the dozen ways do you provide that information?
So in the end;
* no real financial benefit to providing that access (especially when you say you don't want to pay XD )
* no certification to provide that access
* actual software knowledge requires domain knowledge many magnitudes greater than just file formats, including per-FI non-public knowledge
last and not the least important items not discussed above;
* financial institutions are slow movers when it comes to adopting technology.
* early adopters are NEVER the small banks, and they always require a hefty ($$$$) reason to include it.
So it could be done - and probably will in the next 20 years or so - but not today. ...
as an aside,
The reason you'd be getting billed 5 figures for access is because they'd end up assigning someone to manually pull your ACH records out of each daily batch and save them to one side. Manually. They may even need to have someone actually type the new entries in (separation of networks, removable media would be disallowed - of course.)
Small progressive smart banking. I work for a small developer that makes software that connects their system to accountants across the country.
Try to talk to Eric King, If you have a good idea for a service they could be offering, he'd be the one to talk to.
http://www.rivercitybankky.com
my experience with US and Canadians banks has been awfull, old systems either trought manual input forms or absurd flatfiles. on the other hand my best experience has been with European banks, they all have some sort of webservice either soup or RESTful service where you can access all you can need to integrate with your systems. i hope in the future they adopt some kind of standart to work with their data, like the OpenTravel thats being used more and more this days and make integration with travel systems very straight fordware.
I stopped reading after like 5 insightful/interesting posts.
Note to the Article poster: SWIFT is a bank to bank network. You will never get access to it as a non bank, and if you hack into it ... prey that no one figures it.
To the others pointing out why the OP never will find what he wants:
You ofc. can ignore all my "for free" claims above, as that is not for free but covert by the base yearly fees.
To the OP: try Pay Pall, they have an oen API to access them, no idea how useful that is for you or which applications exist to use it reasonable.
Regards,
angel'o'sphere
Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
I'll state at a high level that I work for a Credit Union, and there are a lot of us that believe in a model such as the one you are describing. Can I take this discussion in a slightly different direction? Rather than "where can I get this today", how about "what would you want from a service like this"? Reply with a list of features and describe the problem you are trying to solve.
Do you want to only access your own account, or offer a service to multiple customers of the financial institution?
Are you thinking along the lines of web services?
What type of transactions would you want - realtime (i.e. what's my account balance now) or batch (show me all transactions for the last 6 months)?
Are you talking about wire transfers, ACH, checks, etc?
Are you thinking a pull model, where you query into the data or a push model, where you are alerted when things happen?
Don't get dragged down in any pricing or cost at this point - just tell me in more detail what you want.
Any suggestions for a 'developer-friendly' bank for small businesses?
Translates into honest english as
Any suggestions for a bank that will give me expensive features for free?
The cost of providing such access in a stable and secure way and all the support that goes into it has to be split up between the handful of customers that want this functionality.
Besides, I wonder what kind of information he want to extract that isn't provided in more regular formats. What interresting information does he expect to find?
FWIW, I have some experience with maintaining parsers/writers for the shitfest known as SWIFT.
Slashdot social media options: AIM, ICQ, Yahoo, Jabber and Mobile Text. Why no MySpace?
Yodlee is a service that back-ends online banking applications and payment systems. Maybe they would sell you a development license or something...?
Take off every 'sig' !!
So, the underlying systems that most banks use are extremely robust after having been tested for decades and - very possibly - being a lot simplet than similar systems would be if they were coded today? That sounds pretty ideal situation, to be honest.
It seems that there is an opportunity here.
Put up or shut up. Does anyone actually want to *make* the geek friendly bank?
According to the almighty internet, we would need (depending on the state) about $5mil starting capital. About 10-20% of that comes directly from the founders, the rest can come from shareholders.
We would need some founders who have cash, some who have the knowledge and ability to implement the system, some who have the ability to *run* the system, some with the ability to negotiate the legal and procedural, and some with the ability to deal with personal interaction.
If a reasonable project plan were available, I could be one of the cash founders of the bank.
Anyone else?
Disclaimer: I am a developer in one of America's largest banks.
Of course, I do not speak for them - just for me.
That said, think about what the OP is asking.
He wants unfettered access to funds transfer information.
Just to keep righteous with the Feds, we'd have to forcibly limit his access to accounts only his business has - it's not like we could open the tap and just let him run BAI queries on any account he can think of, the way our own internal users can.
But a web portal with customer-keyed access is already present, and isn't good enough for Mr. LookAtMeI'mABigBadCoder.
So we'd have to build him a distinct data transfer channel, test the hell out of it to make sure he can't break it and look at someone else's stuff or - God forbid - foul up our nightly batch cycle with stupid data requests outside of standard Banking hours. Then we'd have to test it with him involved to make sure it returns the data he wants.
All that is probably a several hundred hour project. Per customer. For something no one ever wants but this yahoo.
Of course, several hundred hours assumes the full banking software is Bank-owned. Most folks outsource this stuff, so we'd have to test cross-vendors with him, too, generating costs for us.....
You want it for free?
You're fucking clueless.
Let me throw this out as an initial proposal.
The purpose of the bank is not to make the maximum money, the purpose is to get the most customers.
The bank makes money in two ways:
1) Differential interest between Fed and Savings, and
2) Lending
Lending is done very conservatively.
We cater to the geek crowd:
1) Online access only (but administrative offices with real employees)
2) Good web presence and features
3) Comprehensive accounting and audit trail
4) Excellent security
5) Open API for access
6) No charge for electronic transfers/access
7) Electronic statements
8) Nominal/at cost charge for reading in physical checks
9) Nominal/at cost charge for cutting a check and sending it by USPS
10) Nominal/at cost charge for printed statement
11) Legally compliant in all ways
12) FDIC insured
We would need a couple of people with serious cash, and a couple of people with the wherewithal to implement (or manage the implementation) of each of the items above. ...and we would need a metric buttload of people who pledge to invest in such a bank, ...and another buttload of people who pledge to become a customer of the bank when it's up and running.
(And maybe some alpha and beta testers)
If this comes together in an appreciably believable way, I could be one of the cash founders. As in - getting the seed money is not as difficult as you might think.
Reviewing the responses above, many people profess knowledge of how to implement one or more of the specifics.
Any takers?
If you ask: "hey, can I get online access to BIA or SWIFT messages" they will come with very expensive solutions.
However, do you really need that? My company works in financial business and most of our SWIFT messages are send out a couple of times a day.
And bank statements (MT940) is retrieved once a day.
The moment you start asking things that have never been asked before banks will have some problem delivering.
I can give some horror stories from corporate experience about interfacing with some of the big names in the financial world....
But maybe you can explain the exact situation to them, why you need specific data.
And they might well have a better solution that would fit your needs at a reasonable price.
Or see a business opportunity.
I'm one of a number of people working on software for mutual credit systems (LETS: http://en.wikipedia.org/wiki/Local_Exchange_Trading_Systems ) and alternative money and local currency systems, especially within the context of transition towns: http://www.transitiontowns.org/
There's a list of software, including mine, Cclite: https://sourceforge.net/projects/cclite/, at the end of the wikipedia article.
The end point of this is to take back both banking and control of currency into a more responsible and democratic sphere. This will probably only be partially achieved but will give a healthier mix of monetary control and resilience than at present. Meanwhile, I'm hoping and working towards a stage where some of this can be used for, at the least, running Credit Unions. There is a lot of Credit Union software [at least inthe UK] but all of it is closed-source, expensive and tends to be Microsoft-oriented.
I marked this related because it isn't an answer to the OP, but it's an important related and more radical part of the subject.
On y va, qui mal y pense!
Bank of New York Mellon is huge in processing. A main focus of their business is doing processing for other banks or other financial institutions so they process many times their internal assets as a matter of normal business. Because of this they have a whole department that all they do is customize processing. I don't know if they have a minimum $$$ of business but they are setup to help you do anything that is possible. One of the surprising things is how enthusiastic they are about finding ways to customize processing to fit your needs.
What you are looking for is generally called 'treasury accounts' - it is expected for bank in that case to provide some kind of electronic access with full details (in/out wires, etc). Most big banks do offer that.
Try Chase - it's not cheap but not unreasonable either (something like 100$ per month plus some per-user fees etc). I have that and it's fine. Plus you get SecurID token to authenticate outgoing wires, etc.
Unfortunately, you kind of have to write a screen scraper to get to the CSV reports, but the data is there.
I'm the co-founder of Banksimple. We are still a few months away from launch, but our entire front end is built atop and API that we intend to expose to the public. I personally started working on this project when I came to the realization that banks in American fundamentally make money by keeping customers confused. Banks make more money from fees than interest margin. Because of this, banks have very little incentive to let their customers better understand their own finances. We think that sucks, and our API is just one of the ways that we are going about fixing this.
:wq ~ ~ ~ ~ ~
The primary cause of price swings in cotton is the health and abundance of the cotton crop. When the cotton crop is harmed by lack of rain, too much heat, flooding, or whatever, then the price goes up. When every farm works hard to increase the amount of cotton they grow because they price is high, the price falls.
I don't think you really know what you're talking about.
Options volatility is a self-fulfilling prophecy: because volatility happens people use more options, speculators see businesses using options and speculate around them, etc.
Okay I'm just going to say that you're confusing options volatility with volatility in the underlying asset, which is what we were talking about. The rest of your post makes the same mistake.