Hacker News new | past | comments | ask | show | jobs | submit login
It's Getting Harder to Tell Banks from Tech Companies (bloomberg.com)
322 points by severine on Feb 15, 2018 | hide | past | web | favorite | 252 comments

I work for a large healthcare system, and the "we're not a tech company" line is often heard. At many levels, I agree—we're in the business of delivering healthy outcomes to our communities, so we need to make sure that efforts across the enterprise support that—but it is often applied in ways that stifle us from delivering what we feel would be the most impactful and cost-effective solutions.

I like to respond that Netflix isn't a tech company, either—they're in the business of delivering entertainment. Their position and reputation allows them to build a stronger engineering culture, but their technology still is subservient to their non-tech business goals. While "software is eating the world" has been quoted to the point of being a cliché, the organizations that recognize software as a competitive differentiator—rather than merely a cost center—definitely have an edge against their tech-averse competitors.

I work in technology for an investment bank and i hear “we are not a technology company” every quarter. Ironically, the tech we develop is a major factor, if not the only factor, for revenue. My motivation is instantly shattered when i hear that.

I used to work in another financial-related company and "IT is a cost not a revenue center" was thrown around as often as possible. Raise, promotion, formation, a computer made in this decade, anything really would get us that answer. Not only from my boss, but from the IT's top director in various meetings and one-on-one. I used to find it highly frustrating and unfair given that the company would cease to exists without our software, and that all new customers were acquired through new development in our software.

Now that I'm older and not working there anymore, I understand that it's a culture problem and that management just didn't know any better, but I've come to accept the fact that it was a great way to cut costs and get away with it. After all, they still have many employees and manage to make more profit year after year. It just made for a very demotivating environment.

The old "We are a tech company, we just happen to do retail/advertising/communication/movies/hospitality/transportation/payments better than you idiots" seems to be working pretty well for Amazon/Google/Facebook/Netflix/AirBnB/Uber/Lyft/Stripe.

This makes me both sad and happy. Sad, because that’s such a shortsighted view; happy, because I run a small financial services company and we compete with them.

We’re prudent, we don’t like to waste money, and we analyze investments from every angle we can think of.

But we look at Tech (we don’t call it IT) as a profit center. We don’t invest in anything unless it can drive down costs, increase revenues, or is needed for regulatory or security reasons.

And when we pull the trigger, I want to make sure that we’ve got the best stuff on the market. Everyone uses either Macs or PCs (we’re agnostic), and run as many monitors as they need to feel productive. Everyone has an iPhone and an iPad. When the hurricane hit Houston, our office building was out of commission for over a week, but we kept running the business.

I want to make sure that Tech allows us to multiply our capabilities. We have a paperless office where any paper is scanned at hi-res, put into our systems and linked to client or regulatory records. Everything is shredded afterwards.

Long ago, I worked in IT, and was dumbfounded by the shortsightedness I saw inside companies who don’t know how to utilize IT. I love competing against those idiots now, because my capabilities are way higher and our cost structure is lower. If we didn’t have all our paper copies scanned in when the hurricane hit, we would have been dead in the water (sorry for the pun).

So, I really feel sorry for all those people who work for morons like the Parent describes, but here’s my recommendation: go start a company and run with your vision. You may fail a couple times (I did), but once you win - and it only takes once - you will never look back.

THIS. This is the kind of company we should strive to work for.

Appreciating something just because it's novel is bullshit. Appreciating something that solves an actual problem effectively.. that's where I want to be.

The IT-Business Maturity model fits here. Considering IT a "cost center" that takes orders is a Level 1 organization. Level 2 is about optimizing IT processes and Level 3 is the partnership of IT and Business on strategic goals. If a financial firm is treating IT as a cost instead of partner, they have some maturing to do.


It’s funny that in the .gov space, I’ve seen management in agencies ranging from corrections to taxation to social services envision themselves as software companies, because it made sense. Every dollar invested in strategic tech produces $3-10 in outcome.

Ironically, the orgs responsible for technology “get it” the least, as they are in the business of ITIL bullshit and order filling. IT bureaucracy will spend $10,000 to save $0.10.

>I used to work in another financial-related company and "IT is a cost not a revenue center" was thrown around as often as possible

This has literally always been the case with IT; they were always seen as cost-centers, not profit.... which is largely due to IT departments historically falling under the CFO....

CEO: "Keep these motherfuckers from spending money"

IT: if "You want this [project] to meet your requirements of [XYZ]?, then == $2000000

CFO: "Fuck you, lets do it for half that and drop security and accessibility as a result"

We actually dropped the CFO and life is beautiful.

It's a common concept that many managements see IT/infrastructure as cost center, even their tech to be used to generate revenue. It boils down to minimize the systems to maximize shareholders value but requires high uptime.

Funny, I go the other way, somewhat. Last time I was interviewing for jobs, one line I heard from a whole lot of financial companies was, "We like to think of ourselves as more of a tech company than a bank."

It's an instant turn-off for me when I hear something like that. These are companies that should be thinking of tech as a means to an end, and not an end in an of itself. Even when I worked at a major software vendor, the tech was still just a means to an end.

If (good) tech companies use tech as a means to an end, and a bank wants to think of themselves as a tech company, that means that a bank thinking of themselves as a tech company doesn't necessary think of tech as an end in of itself.

Companies too invested in tech tend to spend themselves into bankruptcy. Companies not invested enough in tech tend to be overcome by companies with better tech more slowly.

There could be multiple reasons for this, two of which come to my mind:

1. The person who said it may be fancying his own role as tech-heavy and might relate to his/her imagination of what tech-pure companies do. Saying, we are like Google, appeases the person saying it, and satisfies their ego.

2. Often candidates like listening to this. It may not have worked for you, but for the candidates who like hearing this, this works wonderfully. If I were to wager, this line works for more candidates than it pisses off.

> It's an instant turn-off for me when I hear something like that.

While I can see why you find it grating, I think I would forgive them. Whatever else they are, interviews are also those awkward social moments when new acquaintances are just trying to push beyond small talk while trying to please each other. At this stage, it's normal for people to try a little too hard.

You're right that it's could just be a recruiter being over-eager and/or missing the mark on what would attract me personally. But that doesn't really change my stance on the subject.

Interviews aren't really like a casual social interaction. Progressing past the initial phone call typically involves spending from several hours up to to a whole weekend on some cute homework project, shirking on one's current job for more phone calls, burning PTO for on-site interviews and whatnot. I've got precious little time I'm willing to spend on that sort of stuff, so it's in my interest to be supremely petty.

Its kinda fashionable to say such things, I'd think its natural that they do that especially when trying to get good tech people in (I presuming here). Bigger question i8s how far beyond the interview that sentiment truly goes.

Maybe you can fix the problem where I want to download all my statements and transactions for the last tax year. It should be a one button push from their web site. But nope, it's always 15 minutes to a half hour tediously downloading things statement by statement, form by form, argghh.

I am sorry you have to spend half an hour downloading your statements. But what you’re complaining about comes under commercial banking tech, which is generally outsourced to IT consulting firms. Our tech platform is generally for institutional clients, who have systems on their end, and our systems interact with theirs to carry out business.

Also work in tech for an investment bank but I hear that we are a tech company every quarter. And it's pure BS, no one here cares about development experience :)

Come work at MUFG, largest bank in free world, $2.7T, which decided we have to be a tech company.

Our head of technology strategy for Americas is on hacker news at quarter to midnight. And hiring, not just for blockchain. (See my “about” text.)

Worth a chat if you’re in NYC or San Diego.

If you don't mind me hijacking your comment, I'm actually currently trying to solve a problem through which I am trying to connect to people such as you. Essentially, Kiva has served the world's unbanked and underbanked through crowd-funding to some success, but ultimately to reach all the underserved we need to get the data we and our partners have to large regional and global banks so they can price the risk of these potential borrowers and bring in larger amounts of capital.

It would seem if we can get the data to banks such as MUFG in a usable format, then it presents a large opportunity, however how would one even begin to interact with such a large entity to explore it?

This sounds great.

You’d troll their LinkedIn looking for a title that esonated, read the job description to be sure, and try connecting, exchange enough convo to move over to their bank email, and then share the idea.

In our case, I’m the right guy. You can email me (and connect on LinkedIn if you like) and I’ll reply with my bank email. Elaborate on the idea then.

I’ve got 250 leaders thinking about how they could use new kinds of data, we’ll connect you to right groups. Even I have a team of PhDs looking for interesting things to tie together to better serve communities.

Note though, getting capital isn’t our problem, we’re interested in more ways to give it out, use it, put it to work.

I find this statement funny, if only because in the context of Silicon Valley culture, the statement "[x] is on hacker news at quarter to midnight" normally just means "[x] isn't doing any work at quarter to midnight."

(I understand what you're going for and appreciate it, it's just interesting how social signals are interpreted given different industries / contexts.)

(... and yes, me posting this comment does, in fact, mean that I'm not doing any work while I post it.)

On the contrary, if your job depends on understanding trends in a field, read more:


Being aware of HN is part of my work.

In my case, we get four key dimensions of value:

1. Seeing trends early is in the job description, and a small slice of zeitgeist is here. I check here, several other self-selecting subgroup sites, a broad RSS list, white paper subscription sites, etc., and typically also read one subject matter book a day. With enough dots of info across a broad array of sources, you can see trends earlier.

2. Finding emerging tech likely able to go ‘enterprise’ is also part of my JD. That often depends on leadership. Some of those leaders are here. You can tell a lot about a company from its founders. I’ve been an extremely early customer of startups found here, years later they lead their segment and we’re already on board. Examples like Hashicorp or CoreOS come to mind.

3. Emerging debates that do matter are often here early, where threads are high quality enough to get exposure to well thought, good logic, alternative views. Being able to absorb more of those improves low data decisions.

4. You can even find talent early. Many met here have gone on to release awesome tools, join or found cool companies, get acquired by BigCos... it’s great to see talented and motivated trajectories and learn it’s not all just luck.

I work for a bank... and I hear that, 'we will be a tech company or not a company.' All the time...

Conversely, I work for a very small company where the owner likes to fashion himself running a startup even though he's been in business over a decade. Technology features heavily in what we do, but I cringe every time he says "really, we're a technology company".

How could you say that the tech you develop might be the only factor in revenue calculations? That makes no sense to me in investment banking.

If tech really is the differentiator, then hearing that means tech staff won’t get commensurately compensated for delivering that decisive factor to the revenue mix.

The rules of old business still applies in most organizations: the closer you are to the transactional point where money changes hands, the more you make. This many times means sales, then management, then finance, then marketing, then engineering, then back office.

Yeah I get all that. I guess what I mean is that I just don't believe that the technology at an investment bank is responsible for all of the revenue at the bank. There's no way that's true. There are people who actually go out and do advisory deals and such which are actual people making things happen. Sure, they are enabled by technology, but those kinds of deals have been done for decades.

Investment banks also do huge volumes of transactional business - dealing in shares, bonds, fx and derivatives of those. And all of those need hundreds of risk reports either for management or regulators.

The deal-makers can get by with a spreadsheet and a BlackBerry, but the rest needs serious amounts of technology.

Ideally advisory business is balanced with dealing and financing, but different houses will be slanted one way or another.

Tech is obviously not the only factor, but it certainly is an important one. One of the main factors in finacial services tech industry is time to market, efficienct algortihms, being compliant, handling huge bursts of volume and accuracy. Our systems ensure smooth functioning of the business across most exchanges in the world. I think the business is practically impossible without robust technology.

It looks like Goldman has no tech people among their top executives. I would have expected at least a CTO or CIO. I don't see this working out for them.


Marty Chavez was, but got promoted:

> Prior to assuming his current role, Mr. Chavez was Chief Information Officer, responsible for the Technology Division

Profile on him from a few years back: https://www.nytimes.com/2016/04/03/business/dealbook/goldman...

Well, that's something at least. They have a CIO. And someone with a tech background can make it to the inner circle. Still, the CIO is not counted among the top executives.

Why are you assuming that someone with the title "CIO" has a tech background?

I worked at a place where the CIO was a lawyer who knew the business really well, but no real tech in his background.

Because it is in accord with my experience and my general sense that people should know what they are doing in the areas they are are given responsibility over. That generally requires experience, and at high level probably extensive experience. It's hard to see how a lawyer with no real tech experience can make good choices in the area of technology.

He wasn't great but he wasn't terrible. He had a very logical mind and tried to choose wisely as far as whose advice he took.

His only knock was that he believed the Cyber Security FUD too much.

Other than that, solid guy and I'd work for him avain.

Then who do engineers report to?

Presumably a VP of engineering

Probably not - you'd be surprised how low the the title of "Vice President" is in finance companies.

In my opinion, not unreasonably for these two reasons:

Banks generally delegate responsibility for huge sums of money to quite a low level in the organisation, and their clients are usually much smaller than the bank is. There needs to be some kind of parity in the apparent seniority between client and bank. The CEO of an SME doesn't want to be dealing with someone who sounds junior when discussing deals regarding the total value of their company.

Banks are so large and complex that it's usually not far off the bottom wrung of the org chart that you're responsible for huge sums of money/risks/contracts. People in smaller organisations with similar monetary responsibilities (in absolute terms) are usually the senior ones with senior sounding titles.

For example, I was a lowly infrastructure engineer in a tier 1 global investment bank working in the platform engineering team not far off the bottom of the ladder, yet I was responsible for infrastructure spending decisions in the scale of 10-100 million dollars. A drop in the ocean for the IT budget of $5Bn dollars, but a larger monetary responsibility than the CEO of a mid sized company has. So yes, I was called a VP.

VP of Engineering isn't as devalued as VP of Sales even in companies where a couple of years of experience gets you VP. Subject matters.

I know people who are less than 5 years into their first jobs from graduation who are VPs in Engineering, in financial firms in London (City) and New York (Wall Street). Organisation/culture/industry matters more than subject, I think.

In my experience it's pretty rare to get the VP title in less than 5 years. In fact in my graduate intake nobody did it in that timescale. Good people would get it in 5-7 years.

2-3 years as an analyst, 2-3 years as an associate, boom, you're a VP.

I like to ask people what's the biggest advertising company in the world.

You'd be surprised how many people don't know :)

You know who else would be surprised, is 1998 Larry Page & 1998 Sergey Brin. Especially the part about hearing their company called an "advertising company," much less the largest one.

Which then makes any ad-paid company an advertising company. Which is a completely useless definition.

Don't mix source of revenue with core business, especially in multi-sided markets.

But their core business is selling advertising.

Don't complain to me, I didn't say it.

I say the same about largest media company in the world.

It's shocking that people don't know it's Facebook.

And in their case not just in the world, but the largest media company in history by a very wide margin. People don't see any connection between William Randolph Hearst and Mark Zuckerberg; Facebook really has the best PR team.

Well the difference is that Facebook is an aggregator and doesn't actually publish or write any of their own content. They're a media company, sure, but they're not directly comparable to newspapers or cable news outlets.

Google doesn't make the ads either, they just publish them on their platform

In a world in which any piece of content you want to exist does exist because someone is written it (and we're a lot closer to that world now than in Hearst's day), power lies not in production of content but in control over which content actually gets seen.

From a consumer point of view, Facebook is pretty much a news source. For some even the only news source.

Facebook is an editor(like at a newspaper), except in this case editorial control is determine by billions of users and a completely opaque "News Feed" algorithm.

Remember, people don't just share articles, they share articles with their commentary.

Channel matters FAR more than content.

("Content is king" was always a distraction. And that came from Sumner Redstone (Viacom), not Bill Gates, who only quoted it.)

They do though, Facebook Watch is them funding original content.

Exactly what do newspapers and cable news have in common with each other that Facebook doesn't?

They pay reporters.

Quite many of them don't, many (most?) not-top newspapers re-publish news gathered by others; many (most?) cable companies don't produce any news content themselves but just provide access to a bunch of news channels. Those still are media companies.

> many (most?) not-top newspapers re-publish news gathered by others

If you're referring to AP and other news services, they pay quite a lot for that, so the reporters are paid. Also, they contribute their own stories from their paid reporters to the service for other members to share.

Facebook employs moderators.

You don't see a difference between a moderator and a reporter in terms of content creation?

In other words: Facebook doesn't employ journalists, certainly not to do journalism.

> I say the same about largest media company in the world.

Indeed. I was just reading about the local broadcast rules (specifically: Sinclair broadcasting wants to buy Tribune media, and the FCC appears to be saying 'go ahead".) . Some people are concerned that this violates rules against concentration of ownership in broadcast stations.

But doesn't Facebook operate in essentially 100% of US markets? Should we even care who owns the TV broadcasts?

The rule was made because radio waves (broadcast) is a finite resource, and so there are a limited number of players who can compete.

The internet can, theoretically, have an unlimited number of competitors, so it's not regulated the same way.

Sure, that mattered in the 1930s but does it really matter any more?

Are there suddenly unlimited over-the-air frequencies available to broadcasters?

No, but there are so many channels of info these days OTA broadcast is a minor part of it. In addition the economics of the business lead to increasing amounds of network content rather than expensive locally generated content on the airwaves. And in fact OTA is much less common than watching local channels mixed in with otehr channels on cable, so again the local channels aer drowned out.

Looking up the stats a couple of weeks ago during a different HN discussion I was shocked how popular TV still is across all demographics, but local content was a plummeting percentage of the total.

Might it not be Google?

Who is the biggest advertising company? Google?

Unless something changed recently, I believe that $95bn of advertising revenue makes them the biggest advertising company in the world :)

They are the medium, not the advertising company, the same way a tv channel is the medium, not the advertising company. WPP, Omnicom, Dentsu, those are advertising companies. You are conflating here two different things.

they are distribution, just like the agencies. Only Google drives distribution costs down

Probably WPP or Omnicom.

Omnicom: $15B revenue annually, $17B market cap WPP: $18B revenue, $23B market cap Google: $110B revenue, $750B market cap

Seems like a good argument that being tech-forward provides a multiplier on value versus doing the same old thing, if you believe that these are all just advertising companies.

> Google: $110B revenue

Where did you get 110 from? I'm seeing 90 for last year.

That's all revenue, including non-advertising revenue.

True, the ad revenue was $95B.

Yeah, I screwed up and was looking at the wrong number.

Hardly a fair comparison, considering Google's extensive non-advertising product offerings

Those "other revenues" only account for about 13% of Google's revenue. So its a fair comparison. Google is almost 90% an advertising company.


Yes because all those apps exist... to funnel people towards Google's extensive advertising network, so the money shows up as adbiz revenue instead of with products.

So, by pursuing other technological innovations, they are able to expand their advertising business?

Extensive non-advertising product offerings that somehow don't translate into extensive non-advertising revenue...

"if you believe that these are all just advertising companies."

I'm not the one calling Google an all-advertising company, I'm just saying that if it is, it's by far the biggest.


By a factor of 30

30? More like 5. Still a good point.

Revenue vs market cap

A major reason why upstarts like Netflix could successfully disrupt old guards, while banks and healthcare industries are still largely immuned despite huge inefficiencies in the way they operate, is the trust factor.

People don’t mind trying out a new entertainment service by a small startup especially they get a free trial month. If it breaks, so what? The same cannot be said about places they choose to manage their finances or health.

Perhaps it might be more promising for a new partnership by big-names in adjacent industries, such as the Amazon-Berkshire Hathaway-JP Morgan healthcare initiative, to disrupt these trust-based industries.

If there is a good way to accelerate the process of consumer trust building, it might allow more newcomers a chance of disruptions. It should result in lower costs and economic surpluses for the consumers. ‘How?’ is the 64-million-dollar question.

100% Agree. It’s like saying Coca-Cola is not a marketing company, they sell beverages. Or Apple is not a “design” company, they sell hardware. sigh

> It’s like saying Coca-Cola is not a marketing company, they sell beverages.

They don't even sell beverages. Their (mostly) independently-owned bottlers do!

>Coca-Cola is not a marketing company,

I never agreed with this mentality. Advertisement is Coke's cost center; it's not what they sell. They sell licensing rights to soft drinks: that's what makes them money.

Advertisement is a cost of having a customer facing brand. Should P&G and Unilever be considered advertisement companies?

> Should P&G and Unilever be considered advertisement companies?

Absolutely. The value of their businesses are entirely based on the brands that they have built, and those brands are built by advertising. Making food to stuff in the boxes is just a cost of building a successful brand.

But the brand itself doesn’t make them money. That doesn’t mean it isn’t valuable or even that it isn’t their biggest asset. What makes them money is people buying their product.

Apple is a design company.

Foxconn, Qualcomm, and Samsung make their hardware.

Cognizan and InfoSys develop some of their software too.

You can fairly say that Coca cola is a packaging company.

Can't you play this definition game with any company though?

I kind of think that if an organization views tech as a competitive advantage and core competency, they are a tech company. E.g. Netflix may be in the business of entertainment, but they are also a tech company. Google may be in the business of selling ads, but they are also a tech company.

Yes you can, and that's kind of the point.

Slowly but surely, more and more companies are realising that their digital efforts are just as integral to their business as whatever else it is that they're doing. Every company needs to be a tech company now.

In Australia, Commbank realised this at a very early stage: there's actually very little differentiation between the financial products from bank to bank. Sure, interest rates change here and there but on the whole, everyone is the same. Commbank realised that they can make their internet banking and mobile apps just as much as their core product and their differentiator. People aren't going to stay with their bank because of a debit card, but they will stay the one with the better app with more features. They rolled out a bunch of firsts - peer to peer payments by bumping phones together in 2010, transferring money to others using just their phone number, or 'Cardless Cash' where overnight they updated their network of ATMs to allow withdrawals using their phone app.

Qantas is another example of a company that invests (for better or worse) heavily into their digital products all throughout the customer journey.

Every company needs to be apt at digital because, in one way or another, every company is a tech company. It's the successful ones that realise this and use it to their advantage.

CommBank certainly did lead the pack, but they have become arrogant and complacent, IMHO. Two things that have driven us to leave them: rejecting Apple Pay, and cutting off feeds to accounting software from certain credit cards. We're leaving for a combination of a local credit union and an 'online-only' everyday banking account.

> Every company needs to be a tech company now.

I've heard this mantra for some years now (in various forms, eg "all companies are tech companies, or dying ones"), I wonder where it stems from.

The way I read it, there is a management position of shunning new tech by saying "we are not a technology company". That statement can also be interpreted as "we are not an innovation company" which conveys an aversion to change.

I think the point of the GP is to say that Netflix "isn't a technology company", but they don't parade around saying that. Netflix uses technology and innovation to the fullest as the means to accomplish the ends of delivering great entertainment value at a low cost.

I don't think it's as much about the "definition game" as about the message that is conveyed.

I work at a financial services company.

You simply cannot exist if you are not a tech company due to the complexity of the regulations.

>but their technology still is subservient to their non-tech business goals. While "software is eating the world" has been quoted to the point of being a cliché,

Better to say "software ate the world, but netflix has eaten the software" by almost singlehandedly wiping all its upstart competitors from "indigenous tech companies."

To some extend, Netflix deserve some credit for tech. A non-dotcom running tier one CDN is not a small deal.

Yet, they are still a small fish "in the bigger picture of things" though.

My employer uses the line to justify paying developers below market wages.

You'd think this would pay more, to make up for lack of supportive infrastructure and career opportunity.

>"I like to respond that Netflix isn't a tech company, either—they're in the business of delivering entertainment."

knowing multiple people from netflix; this is literally a very-uninformed comment...

Much of what you even KNOW about cloud-computing, spot markets, streaming, CDNs, etc was pioneered by netflix.

They are secretive, but innovative as heck. So, your comment is a big NOPE.

Right, but the purpose behind all of that was delivering entertainment. Hence, entertainment company.

Fine. Ill concede that point.

That was my entire point. Netflix has introduced numerous innovations—many of which they've shared publicly—all while not making a dime (to my knowledge) from selling the software itself. Contrast that with a company like Microsoft, where software (either shrink-wrap, or as-a-service) is their core product.

> Much of what you even KNOW about cloud-computing, spot markets, streaming, CDNs, etc was pioneered by netflix.

Let's not overstate the case. Much of this was around before (sometimes long before) Netflix's digital rise.

Wow, I think you missed his point entirely.

That was the entire premise. Netflix isn't a tech company in the same way banks aren't - technically. But obviously both functionally are since their entire businesses depend on their technology

I'm of the opinion that all banks should be technology companies, in that pretty much all of their operations rely on computing, so the better they are at developing and operating computer systems they better they'll be able to do business.

Unfortunately, many banks (at least in the UK market) seem to see IT as some kind of overhead, to be minimized in cost and/or outsourced (e.g. Lloyds outsourcing to IBM https://www.theregister.co.uk/2017/06/06/lloyds_confirms_ibm...)

Of course there are ranges of new banks (the UK has a range of challenger banks most of whom take a more tech-focused approach) to challenge the "traditional" approach, and it'll be interesting to see how that shakes out...

100% agree. My wife works in private banking, and her descriptions to me of the IT/software environment they have to deal with are absurd.

Recently, the integration of a new loan management system was badly botched and several clients have been hit by it. Bills not going out, therefore not being paid, therefore money being taken forcibly from their accounts without notice. And these are customers with millions each in the bank. This has been ongoing for a couple of months now. The client managers themselves are irate--they've been cultivating these relationships for years, and it's all been jeopardized.

How can so many banks be so blind to the fact that everything they do depends on getting the technology right?

My theory has always been that the people in senior positions at these companies tend to be older and have come up through the ranks before IT was quite as integral to their operations.

So they don't understand technology and therefore tend to not take it as seriously as they should.

That said, that argument starts to hold less water these days. I had my first job in a web-focused bank 18 years ago now!

One of the consequences I've seen (and I'm guessing its even more prevalent these days with easy access to cloud computing resources) is that bad IT in banks leads to the rise of Shadow IT where departments take it on themselves to work around the main corporate IT function.

I studied business school in the mid 00s where most textbooks/classes still insisted that IT systems could be procured as standardized off the shelf products, despite the fact that all serious studies show that process management* and IT systems form a symbiotic relationship that makes it almost impossible to change one without the other.

*Business management is process management for most organizations.

I'm positive this is it. You see this all the time with medicine and government too - lots of old, successful people who don't like having to adapt to something new and difficult to learn

Why is it always the "notifying" part that breaks with shitty software like this, and not the "money being taken forcefully" part, which always seems to work flawlessly? Doesn't that seem a little suspicious?

If they were uniformly bad at writing software, the critical money handling part should fail at roughly the same rate. If they can get that part to work reliably, failure of other parts of the system seem particularly inexcusable.

Always my favorite. You can take money out of my account instantly. Refund? "You should see that in your account in 3-4 weeks". Just brazen, at times.

The product owners most likely decided to budget their time/resources accordingly so if something did fail it wasn't critical money handling part.

It's possible the withdrawals do break sometimes, but undercharging people results in far fewer irate phone calls.

I worked for a bank in the UK that's a US bank that has no retail presence in the UK (that probably narrows it down a lot) but much of the IT operations were being run and managed from the UK. The headcount in the global IT organisation was about the same as the headcount of Apple at the time. I've always known banks has been relying on computers for decades and probably longer than most industries and from my experience this bank was particularly progressive when it comes to technology and happy to innovate and experiment. When working in other parts of the financial industry in the UK it wasn't quite the same as this. I think there's certainly a stronger engineering culture in the US and the benefits of technology delivering productivity at scale is appreciated. The UK also has a legacy of a lot of well paying functions in the financial industry that wasn't relying on technology with people who don't have an aptitude for technology.

JPMC always seemed quite progressive - for instance they embraced open source and open standards, with AMPQ and the Apache Qpid message broker which was part of their tech stack and I worked on for a while there, commiting the code I wrote at the bank to the ASF repository. this is very different from a decade or so ago, when any hint of open source was seen as a possible risk...!

The problem is legacy culture & systems. Theres almost no chance of a transformation unless a paradigm shift in senior management thinking takes place.

Maintaing their mess costs massively and large projects carry risks and a bad track record no one on top can support.

Too bad the fintech revolution lost some of its momentum challenging the profitable areas of banking

I'd agree that a paradigm shift is definitely needed to address the problem. Banks need to see that constant investment not "put it in and let it run for 10-15 years" is what's needed.

And yes the point about risk of new projects... who wants to lead a project to replace a mainframe system that's been running successfully for the last 30-40 years and if it goes wrong you're guaranteed to be on the front page of the papers...

I wonder if a pattern will emerge where big banks let the challengers develop the tech, then they buy them and move across to it, once the kinks are all ironed out.

>The problem is legacy culture & systems.

I'd argue regulatory oversight also makes innovation and risk taking less viable because every failure will be noticed and require explanation, which puts a damper on "fail fast." You can fail fast internally during your SDLC but doing so in front of an external customer is an extraordinarily bad idea.

Most of what you say is true. But the reason they see IT as overhead is because their budget accounts for building the technology and find that supporting it may cost just as much. So they try to get rid of the overhead of Engineering the system to make room for Supporting the system they built. Now, they seem to quickly forget that you still need Engineering to build your next project. And that is why a lot of these companies treat IT Engineering as a disposable asset and then heavily out source the building of these projects to disposable IT Engineering consultants.

Not only in the UK. France, at least, is also like that.

On the other hand, it would be great for consumers and those wanting to offer software in the banking ecosystem if there were more standardization in the banking industry. I'd be surprised if anything a bank does is so unique that it requires all their software to be bespoke. Perhaps we could get better security and usability if there were some large providers that offered key software to almost everyone in the industry.

In the UK at least, we have a ray of hope on that front as mandatory Open banking APIs are on the way (https://www.openbanking.org.uk/) which should make it a lot easier for software providers to write applications on top of existing accounts and providers.

One of the large trends in financial technology development that I've seen is that banks seem to heavily prefer that their staff leave and form third party software service providing companies than hire and allocate resources to internal teams. This means transaction costs are driven up massively, but the leading industry players typically deploy innovations in tandem with their competitors.

My banker friends have big problems trying to modernize sections of their operations on the basis that too much political capital is needed to liquidate entire floors of staff - but that restriction only lasts until someone has an off-the-shelf solution offering 25M in labour savings per quarter.

These inefficiencies and cultural issues hamper execution leading to poorer client outcomes, lower margins, and weaker strategic positioning.

I'm glad that some banks are pushing forward to gobble up that easy ROI, but I am concerned that the net effect of this type of automation is systemic risk, as the industry begins to bifurcate between low and high margin banks, and the latter are afforded accelerating capability to consolidate the former.

Less players. More capacity. More reliance. Less redundancy. Not necessarily bad for John Q. Everyman, but if he's ultimately carrying that risk while pocketing none of the efficiency gains...

This sounds like marketing baloney. I would bet most of that "engineering" headcount is actually just analytical headcount. not a bad thing, but it's not what we used to call engineering.

This is part of an ongoing trend to describe everything as some form of "engineering."

Honestly, if I'm going to Goldman for M&A - I want a human relationship. That relationship may be buttressed by research and analysis using "big data" and "machine learning" - but that's not engineering, that's just good research using the latest useful tools.

I'm only addressing the actual investment banking component of GS. Sure, the consumer banking stuff requires technology and "engineering" but there's absolutely nothing new about this, it's just that consumer banking has lagged horribly behind in consumer-friendly technology.

Not much to see here.

Goldman makes about half its revenue from market execution, which is basically an entirely engineering driven business. Only about 20% of it's revenue come from the "investment banking" sort of equity and debt underwriting type of activity.

To be honest Software Engineering looks like child play when compared to actual engineering work.

Imagine the left-pad incident in a different engineering domain.

It's a bit disingenuous to draw parallels between 24x7 electronic trading operations and javascript web cowboys.

My imagination is failing me. What would the analog be?

Investment banking is not just advice. They also finance deals (i.e. come up with a few billion to make it happen) and execute - buy all the outstanding shares, or sell new ones in your IPO.

And since those kinds of deals don’t come along every day, they maintain the capability by executing as broker and dealer all the rest of the time.

(I know, I’m simplifying a lot. Hope it helps get the idea across)

The idea that companies can be defined as a binary "tech company" or "not a tech company" is a trope created by tech VCs/entrepreneurs as a means to signal that they are "special" from businesses that we created more than 15 years ago.

Most of my clients (mainly PE investors) typically define companies as either "Tech enabled", "tech lead", "nothing". This means usually two things to them: (1) the more tech enabled the higher the margins (EBITDA) I can achieve and (2) the more tech enabled the higher the multiple I can achieve.

And that's just it. Everything else is marketing.

That being said I wouldn't say Goldman is your typical bank. Other banks are much slower at embracing automation and software development is more often than not considered as a non strategic commodity which budgets and headcount fluctuates up and down based on how much is available for discretionary spending. And the statements from its senior management are also too often aspirational and disconnected from the reality of the organisation.

As someone who works in the regtech / fintech industry I totally agree. Large portions of even the biggest banks are severely underserved technologically. There is a lot of wide open space here for technology that improves efficiency and accuracy related to compliance and general operating activities.

That's interesting to hear! I'm going through "Flash Boys" right now, and it (to me at least) paints GS in a pretty negative light wrt their engineering culture.

Granted the book mostly focuses on HFT firms, so I could easily see that GS is way better than most large banks but a bit lacking behind, say, Citadel.

> That's interesting to hear! I'm going through "Flash Boys" right now, and it (to me at least) paints GS in a pretty negative light wrt their engineering culture.

As an aside, if you’re not planning to already, please read Flash Boys: Not So Fast when you’re done.

Is there a way to express/model technology so that it isn't considered this accessory but something that is core to their values/mission?

Have we not done a good job of proving that investment in technology is a force/profit multiplier?

The truth is tech is not a force multiplier in finance, except for HFT. A few firms do pay a ton for engineers (higher than google level comp), but there are probably less than 5000 of those engineers in NYC.

No. Many smaller financial firms derive their advantage from technology. There are many areas of finance aside from high frequency trading and they all benefit from significant force multiplication from technology. Even in the lowly area of operations automation there are firms that achieve significant scale with a small number of personnel through technological advantage. Granted this is sometimes achieved with a combination of in house and commercial technology packages, and sometimes the overall system design is done by non-technical people who just manage grunts to do all of the work, to say that these companies are not benefiting from force multiplying technology is foolish. There are very few financial professionals left who are not technology people in some sense of the word.

Many firms benefit from better data. All that data needs to be acquired and managed. Many firms benefit from better analytics and models. Those models need to be tested and implemented.

Your number is low.

Source is I work in this field. There are positions with big data and data management that pay well. But the ones paying 350k+ are way harder to get than working at a top tech firm. Theres a quant trading firm called engineers gate that has a bunch of ivy league programmers, the top of the top. They havent made a nickel off their technology. They have a had a huge outflow of talent to Google and FB. People do get paid 500k+ to manage a firms data, but thats because the firm makes so much that the money splashes around, not because its a huge component of their profitability.

Even at Two Sigma, the engineers are second class to the quants.

If your cutoff is $350k, I'd definitely say your NYC 5k headcout is way too low. By at least an order of magnitude. Even the smallest trading firm will have 5 tech people (devs/data scientist/quants) making that much after bonus.

Edit: maybe $350k is at the border for that 5 each estimate. But i still think you're of by an order of magnitude in nyc in general.

If "data scientists" working at Google are considered tech roles then finance quants should be as well, right? I agree that there's a distinction between data engineers and quants but to me they're all tech roles essentially.

The difference is that quants make way more than regular engineers in finance. At google, an engineer working on the next generation database can make a ton of money, his work is extremely valuable to the firm. Engineering innovation is central. In finance, engineers in HFT did make a ton, but HFT is no longer very profitable.

Quant is not a tech role. Quant is about maths and statistics.

This is absolutely false, tech is a force multiplier in nearly every business even if it doesn't require onsite software engineers to get that productivity bump. Look at the Bloomberg terminal in finance, Salesforce/CRM in sales, Epic in healthcare, Westlaw in legal work, the list goes on.

Now a lot of the time, firms can get a better product by purchasing one than trying to build something in house (where anyone producing HFT tech won't sell that product because it's an immensely valuable trade secret). This said, I'm sure there are forms of intelligence gathering that is also built in house by firms that allow them to implement market beating strategies.

I am only familiar with these three "Bloomberg terminal in finance, Salesforce/CRM in sales, Epic in healthcare" and these products are not what I would call a good tech

Really? Sounds like the author has never worked in either, and just reads news sites. Having worked in several of both 'MegaBank LLC' and 'A Small Tech Company', they're utterly different beasts. Just because both mention blockchain doesn't mean they're the same.

Culturally, they're worlds apart. Banks offer a career path up and away from technology so you get unmotivated developers joining who are just looking for the next leg-up the career ladder.

Banks are only driven to make money, and the "worst" of the employees are utterly driven by the end-of-year bonus, and to hell with anyone that gets in their way.

From the article:

> (Disclosure: I used to work at Goldman designing derivatives, not apps. Also I have a Marcus savings account.)

I wouldn't say those things are what makes a tech company a tech company.

"Banks are only driven to make money"

If you don't think tech companies are the same way, you're sorely mistaken.

While I do agree with your sentiment, switch "Banks" with "VCs" and "end-of-year bonus" with "IPO" and many people would consider your last paragraph an accurate description of the tech startup scene.

Odd, I was just talking to someone yesterday about this because there was a lamentation of not enough "tech companies" in our area to work for.

I pointed out that in the DevOps Handbook one of the lines I highlighted was that in 2013, HSBC (a very large bank) employed more software engineers than Google.

So I went to search other tidbits like that and found that in 2015 Goldman Sachs employed more software engineers than Facebook.

Banks aren't companies that have tech, they're tech companies that do banking.

This isnt really true. Banks dont pay anywhere near tech salaries. Its all lip service.

The flow of good engineers is always banks -> tech companies and never tech companies -> banks.

They don't pay anywhere near top tier tech salaries. They are competitive below that (the Oracle/Cisco/IBM level).

I make significantly more at GS than I did at google. The difference is that the majority of my compensation comes from the yearly bonus - a quirk in finance.

As a regular engineer? Or is there a math heavy component?

My understanding is that I could get about 200k base at goldman and maybe a 20% bonus, this is for working in a high in demand field. Google/FB/Other job I can get 300k+ while working less hours, paid for lunch, better treatment, more respect.

Google / Facebook / Goldman Sachs and JP Morgan have equivalent grades and conditions. Employees regularly goes back and forth between them.

More important than the company name, you should make sure to join the right role for you in the right group.

That's a realistic compensation package for a normal Sr engineer role at a big bank ime. No, it's not Goog/FB pay but a lot of the jobs are available in LCOL areas like Delaware and the Triangle fwiw.

That's the biggest thing some people don't take into account when it comes to salaries: cost of living. $300k is not always more money than $200k.

Interesting. I work for a direct GS competitor and compensation is nowhere near FANG level, not even with bonus accounted for

This would imply you're at the Director level or above at GS. I expect you were L5 or less at Google, so this would make sense, but I'd expect Director at a big bank to be equivalent to an L6+ role at a Google or FB, or am I totally off base somewhere. (that is, you aren't really expected to make director)

I've seen non director (pres) level offer packages up to $300k including guaranteed bonus minimum.

Those are well within what an L5 could get at Google. Unless you mean 300K base and a significant (ie 30-50% bonus).

That's non-director level, upper end of the vp level.

I'm not quite sure what your point is. Mine was that their bonus being > salary implied a role/level at a bank that was higher than what they likely had at Google, so a commensurate salary increase would make sense.

Could you explain what you're getting at?

Just that pay is high for basically a sr engineer at the right location (close to the money).

The fact that bonus is 50+% of base is normal for those roles, more normal than faang comp. Compensation is just structured differently, so it isn't useful to use it as a measure of seniority between the two industries.

Oh I misunderstood the first time reading this comment. Bonus > salary pretty much never ever happens in Tech, until you're in an executive role.

It happens earlier in banks, but for most banks, it still happens at the level of Director (as opposed to Partner), not VP. I wasn't comparing comp structure to compare seniority. I was using comp structure to guess at seniority within the ladder of BB Banks and then comparing roles.

It depends. In a back office role, you will have a more traditional salary plus 20% bonus. Infrastructure at a bank might be have a higher bonus component even at pres level.

On the trading desk, your bonus is often often open ended, or if you are doing strategy work then a percent of pnl.

Right, and my point is 300K all in is not more than senior comp at a FAANG.

I think it depends on the company. From my personal experience it's something like

crappy software company << bank << startup

So depending where you are in the software world, you can still be making a ton of coin at a bank.

Sometimes in the early stage of their careers people make mistakes (including myself) to move from tech company to a bank. Never doing this mistake again. I've never seen so much legacy code, poor engineering decision and code standards in my career so far. Of course this could be limited experience to my department, but given the comments here it feels like this is a standard situation across the "tech" in the financial world.

The salaries are good if you are banker.

Do they?

Have they dropped "corporate" solutions that barely work, sold in golf courses to people that are not affected in the least by their decisions?

Are they still using XML (or even worse, CORBA) as duct tape tying everything together?

Have they dropped waterfall?

Do they allow developers to have their say on architectures and solutions to use? Can they experiment with, let's say, Scala, Elixir or are they stuck with some old Java version and don't even think of adding a library besides the approved ones?

Disclaimer: I work as a data scientist at a large financial institution based in the US; to be fair it is one that has a reputation for being on the "cutting edge of tech", FWIW.

The data teams I work with don't know SAS and some members aren't even that facile with Excel, opting to use Apache Spark (mostly Python, but also Scala bindings) or pandas instead. Almost no one does serious data work on their local machines, and there is a big push to store all of our data off-prem.

The dev teams I work with actively experiment with different cloud-based architectures, devops automation tools, database solutions, etc.

From a product perspective, lots of teams use agile workflows, but each team is allowed to (and encouraged to) choose their own style of getting work done.

This is not meant to imply that banks have largely moved to this model; I think that we are the exception rather than the rule. This is _also_ not meant to imply that top-down corporate "solutions" don't affect us, and that greed has been completely factored out of the equation. But I've been pleasantly surprised by how much leverage and freedom we have as the "tech department".

Capital One?

XML gets a bad wrap. For a protocol you can easily use over RESTful APIs and define a strong schema around, plus with all of the well-optimized libraries in major programming languages and IDE support, I don't really mind working with it at all.

The biggest problem with XML is verbosity. Besides the excess complexity it also add bits over the wire to transmit all that extra crap.

JSON is king.

Are you intended to read this XMLs? Having strongly typed protocol is a very cool thing in large multi-entity frameworks. Have you seen any successful JSON schema? I have not.

I wouldn't say any of that would make them not a tech company.

Just a tech company that most of us wouldn't want to work at.

Those are absolutely the right questions to be asking, and in my experience the answers are mostly "yes".

Although from what I've heard, their version of moving from "waterfall" to "agile" could best be described as "iterative waterfall"

I'd take that over the "agile means we can yell at developers and just tell them to give us exactly the same things following exactly the same processes, just FASTER".

Tech and finance are two very broad brushes that both paint a large swath of the business landscape.

When speaking of finance there is everything from regional banks or loan originators to investment banks and trading firms. They are often very far apart in terms of culture, abilities, and pay.

On the higher end in tech you get front desk trading, and they are most definitely a tech subsidy inside a larger institution. Some of the best tech teams I've ever been exposured to. I would take the best team GS over the best team from Google any day of the week. Even a mediocre front desk team are far far above average for the industry because they focus on getting the best talent.

Even companies like Bloomberg that you wouldn't really think of beyond their news service is a massive tech company on the back end. They also have trading and low latency data infrastructures. One of the best teams i ever met was the core bond team at Bloomberg. I was blown away at how good they were.

All these places know they are tech companies and pay appropriately. On the other side you have regional banks or even old school trading firms where they aren't really tech companies, but are mostly admin work with average pay.

I've worked at tech companies and finance (hfts, algo trading, dark pool, banks), one thing i really like about finance is how difficult and interdisciplinary it can be. The tech problems I've encountered at finance firms have been so much more difficult than any other place. Google scales by throwing more machines at it, as most other web tech firms. Many times you don't have that luxury in finance. You have to be smart about it.

Remember, Google gives it developers Go because it considers them above average cogs. Finance firms gives it developers APL/A+/KDB :) I jest a little, but I think there is definitely something telling in what is expected of you.

You can go into a big bank and be a 9 to 5er in an auxiliary role. Or you can go in and workn your tail off and move up. But be prepared to work a lot for those bonuses.

> Even companies like Bloomberg that you wouldn't really think of beyond their news service is a massive tech company on the back end

The company's entire founding premise was to computerize tasks still accomplished by humans and paper in 1981. The News subsidiary was founded nearly a decade later in 1990. So it's more a case of the tech company considering News the "News-end" rather than tech being considered the "back end" :)

I was at an APL(ish) lecture a year ago that veered off into the early years of Bloomberg. I thought it was awesome.

I wish banks would learn what a "transaction" is. I have repeated problems with electronic banking where one party to the transaction shows it as a "success" and the other party shows it as a "failure". These problems tend to be tough for me to resolve as both parties insist on their end.

I'm almost ready to go back to writing checks by hand. Waving a cancelled check around has always worked well to resolve transaction disputes. There's no paper trail with electronic banking.

I think the headline should be "It's Getting Harder to Tell Goldman Sachs from Tech Companies"...

Insinuating that all banks are becoming tech companies is being a bit too generous.

EDIT: I am pretty sure Bloomberg does A/B testing on their headlines to see which get more clicks. The URL of the article indicates that the headline may have been different to start out with. I've also often observed the same article on the web and the Bloomberg Terminal with different headlines.

If by "tech companies" the OP means established technology companies like IBM, Cisco, and Oracle, as opposed to Silicon Valley upstarts, then I agree.

Otherwise, I disagree. Most banks are large, slow-moving, bureaucratic organizations saddled with decades-old technology.

Using the latest hyped thing does not make one a tech company.

I think another interesting paradigm is breaking up companies by:

1. embracing technology early 2. chasing technology 3. last movers

Recent examples of last movers that still have been extremely successful have been TJMaxx and Costco, both functioning with very little e-commerce presence. Why does this make some intuitive sense? The first mover has to break ground and if they are successful in applying innovative technologies, they reap the benefits. Meanwhile, several companies chase this innovator, building out similar systems, but these companies are unable to catch the leader, yet still spend a ton in R&D to develop the technology or spend too much acquiring it. Eventually, third-party software is built, commoditizing much of the associated costs. At this point, being the last mover is actually beneficial since the last mover can make the same technological shift without the associated costs that the chasers faced. In 2018, every company pretty much needs cloud instances, notifications, websites, etc. just like companies in the 1970s needed switchboard operators, fax machines, etc. I think its more a question of what companies quickly adapt fundamental new ways to think about an industry than whether the company employs a bunch of developers.

Is this an entire article based on a subtitle on the summary page of the CEO's presentation? Not a bad advertisement for GS but come on, this is incredibly light on content, by which I mean there is absolutely zero content.

I'd like to see a tech company where the time to process a simple transaction like a balance transfer is measured in days.

Banks are tech companies. When other countries say that they trust the US dollar, what do you think that means? It means they trust that our IT systems which manage the dollar are correct. They trust that when we say Warren Buffet has eleventy billion dollars, it actually means something tangible.

And when they discover that trust is utterly and entirely misplaced, that our IT systems have been run as cost centers and endlessly rushed and slapped together and subjected to every form of intellectual rot that modern business practice could infuse into it.... well then things will get interesting. What if Walmart goes to pay off a supplier and the supplier asks for proof that the numbers Walmart is sending over the wire are actually backed by something and not just computational funny money? Well, we do have a form of money that's actually verifiably real and non-counterfeitable and doesn't just rely on some garbage IT....

it is extremely easy to tell a bank from a tech company: tell if they hash their damn passwords. i'm not even talking salt here, just hashing

if you're thinking of some bank's propensity to ask for specific characters of passwords (rather than the whole thing) meaning they can't have hashed it, it's worth noting that many protect the passwords with symmetric encryption and then store the keys in an HSM to mitigate the risk of unauthorised access.

Unless the HSM is also doing the decryption, the key(s) can still be stolen while in RAM..

That's the whole point of an HSM... A somewhat tamper resistant device that executes all encryption operations. You give it the hash and a reference to the the key, and it returns the signed data.

indeed it is. With bank HSMs the encrypted key and characters from the user's password are passed into the HSM and the check is done within the secure hardware, the main system does not get to see the decrypted password.

I was just at a dev mixer and ran into an engineer at Bank of America. He told me they have 5000 Python developers on staff. I almost fell out of my chair - never would have guessed it. I know there's competition for top talent in NYC, but that's still a surprising number.

There's 100k IT workers at BofA.

The new software engineering stack is JIRA, bitbucket, Ansible, Jenkins, Artifactory, Sonar, etc. Every team is being trained on SAFe. Aka the executives choice for "agile" etc

Also the working areas are being transformed from full height or quarter height cubes to open offices where employees no longer have a laptop and use a thinclient attached to standing desks. This is a dream for information security team but personally I hate it as I spent 2-3 hours working from home every day and the rest in the office. I don't want to develop though a windows GUI in the browser :/

Bofa has an abnormally large python group because of their back office stack. Afaik, the head of team responsible for that decision left a few years ago when it was having some major problems. I didn't realize it was still their bread and butter.

As witnessed by Bank of America's aggressive pursuit of blockchain technology & patents:

"[Bank of America] has applied for or received at least 43 patents for blockchain, the ledger technology used for verifying and recording transactions that’s at the heart of virtual currencies. It is the largest number among major banks and technology companies"


I feel like some of the reasons for this are political rather than technical.

It seems like banks are realizing that they need to become tech companies faster than tech companies become banks.

No, it's that it is much easier for a bank to become a tech company than for a tech company to become a bank. The barriers to entry for becoming a tech company are merely technological. They are not trivial to surmount, but they are surmountable. By way of contrast, the barriers to entry to becoming a bank are regulatory, and nearly impossible to surmount, at least in the U.S. When was the last time you heard of a new bank being founded here?

https://monzo.com/ - not in the US, but in the UK some start ups have started overcoming the regulatory hurdles.

I doubt a startup could become a bank. But if Google wanted to start a banking subsidiary? They could probably throw whatever money it takes at the problem and get it done.

The real question is, would they care to do it?

There are many, many small banks in the US; you can just buy one that already has a license. Certainly not insurmountable.

None of those small banks are subject to the regulation that large BHC are. The potential barrier is off by orders of magnitude. The development cycle on those models is multi year. It would take a massive effort to recruit and build the quant teams alone, much less the finance, compliance, and legal personnel with the expertise.

Really pretty much every company is a tech company in 2018 at some level.

"Software is eating the world", came out 7 years ago and is still as true as ever.

It would be helpful to use the full term Informatation Technology to understand these trends.

Financial institutions and many other administrative orgs have always been about storing, distributing and processing information. They did this at scale even before the advent of computers. And they were also early adopters of computers. Their core business is providing service by using IT, whether that technology is computers, paper or human brains.

Back in the long ago we didn't call them IT companies, because that designated a supplier of IT equipment (and we used to think of software as equipment). But now in the age of SaaS, we think of IT companies providing services using IT, just as the banks et. al. have always done.

It's the same kind of business, evolved through two different paths, each with strengths and weaknesses. It seems inevitable that the two paths will merge.

I used to work for a bank for a long time. IMHO, this new fascination with technology is kind of a fad and more of a lip service for marketing purposes. In practice, we still use MS Excel as per what we have been doing for the past twenty years.

Any sufficiently large company is a tech company, and any sufficiently large tech company is a bank.

I'm pretty sure I originally heard that here on HN, I can't seem to track it down though. Seems to only have gotten more and more true as time goes on.

For some reason, I thought this article would go into a different direction - like about how tech companies are borrowing money at record rates because of dirt cheap Fed interest, which to me is a sign of a new financial bubble.

Banks are better described as tech companies than the media / advertising behemoths. Information security is their bread and butter.

As far as I can tell, the only thing that qualifies a company as "tech" nowadays is a technical tradition of building on open source from the ground up (eg. LAMP and its successors), while banks have traditionally been dependent on a constellation of vendors.

The open source approach has proven a more effective model and it seems that this realization has finally spread outside of silicon valley. It's about time.

I don't have that feeling when I see online banking websites...

Just Banks? No, “tech” - which stands for information technology- is just about everything, every industry and job is affected and will change. Information is like electricity...

Haha, i heard lines like that at so many places i interviewed with. 'we're not a finance company, we're a tech company that does finance' etc

I work currently at one of the larger banks, and they actually like to describe themselves as "a tech company who happens to focus on financial things". It's as true as you are willing to believe it, but modern global money movement is really just tech these days, and has been so well before distributed ledger tech came along.

Now, whether any of the banks are any good as tech companies is a different question.

(some grammar fixed)

what is a tech company, really?

it's like saying something is a "concrete company" because that's what the corporate buildings are made of, or a "people company" because they only employ humans

(terms that actually make sense: SaaS product (Salesforce, Dropbox), advertising marketplace (Google, Facebook)... and so on. Not "tech")

Anyone else think that the finance folks are just realizing what the tech sector gets away with by being incomprehensible and too fast for regulators? They're just doubling down and digging into tech to get away with more, for longer...

Finance have been doing this for a long time.

See J.K. Galbraith's The Great Crash: 1929.

isn't it everything ? everything is techno engineered these days

uber the tech taxi

amazon the tech store

... it's the engineered era

Amazon is the way to put your product on the internet. AWS is their biggest profit center and a lot of Amazon's product sales are from merchants who happen to sell their product on Amazon (and use Amazon's fulfillment centers).

Sanford Weill, while chairman of Citigroup, once remarked that their main asset was information, not money (I haven't been able to find the quote but i distinctly remember it from my b-school lectures).

Technology has always driven the financial markets--because better access to data is what allows new instruments and/or the active trading of new instruments.

Very true (and actually has been that way for a while now). Big banks are largely tech companies that happen to do banking.

It's still easy to tell banks from tech companies. Banks probably won't offer you 2FA or allow you to use secure passwords and will cheerfully bring down their website for maintenance for twelve hours at a stretch. They'll also charge you mysterious transaction fees.

heh yea, but no. All our clients are banks. I spent a good part of the day working on a project to convince them to upgrade their browsers to something that supports TLS1.2.

Isn't every company becoming a tech company?


meh. Software is a new form of literacy. No one goes around arguing if we have literate companies any more. It's like fish noticing water. Give it one more generation.

I work in a bank and man - shits cool af

Goodbye Wells Fargo, Bank of America, Citi, and the rest of the credit cartel! You will never be missed.

Hello Wealthfront, Robinhood, Simple, and the rest of the movers and shakers!

> Simple, and the rest of the movers and shakers

Yeah, a bank that announce that 1% of their customers had their account involuntarily closed with 30 days notice is sure to build trust in the new world order.


Simple engineer here, not representing the company's views necessarily but relaying my own opinions.

That situation was indeed a load of bullshit, but the other side of that coin is that we were doing something that has never had to happen in the history of modern banking, probably because it's insane.

You see, normally when a bank is acquired by another bank, the acquirer gets the acquiree's FED terminals, ACH numbers, deals with Visa, etc. that make the transition more-or-less seamless for customers. So customers keep their cards, their direct deposit still works, their checks still work, and all they need to do is log in to a different website to manage their accounts.

What we were forced to do was to essentially close hundreds of thousands of bank accounts at our former banking partner, open new accounts at our new partner, transfer funds, and issue new cards. And we had a contractual deadline before which the transition had to be 100% complete. We explored literally every other option available, including having BBVA acquire our ACH and Visa prefixes, but that was a no-go for reasons I'm not aware of.

So we got the vast majority of our customers to make this painful switch, which was a miracle in itself, but we could not make it all the way down the list, and that sucked. New bank means new account-holder rules, customers changed email addresses/phones, customers moved, we underestimated the complexity/time requirements, all sorts of reasons applied; but the bottom line is that it was an inevitable situation for this kind of transition, and while we did what we could, we let some customers down (0.7% to be exact).

At least I can say that we won't have to go through this again, and we're finally out of the woods with all of these platform transitions, and we're back to building products again. Feel free to AMA.

Are fed terminals still are actual physical vt100 like contraptions? Or it is simply a metaphor for a kind of API access over the internet?

I believe the vt100 terminals have been phased out for FedACH/FedWire, which are the "modernized" APIs (read: the same weirdo fixed-width NACHA format but sent over HTTP/SFTP).

I do appreciate your response, and it certainly sounds like you were between a rock and a hard place.

The end user experience, not so great.

Registration is open for Startup School 2019. Classes start July 22nd.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact