[eins1234]

A lot of comments here saying that this is the death knell for Plaid, but I wouldn't be so sure. Stripe has launched competitors to existing fintech products and failed to gain traction plenty of times before.

Example: Remember the Stripe Corporate Card? It was "launched" as invite-only 3 years ago and is still invite only today, with little signs of traction. Meanwhile both Brex and Ramp have become decacorns through crazy revenue growth.

It's not really just a Stripe thing either. I think people are way under-estimating how hard it is for a large company to launch and succeed with a product outside of their core offering, especially when the competition is a small nimble startup with a singlular focus on that one product.

Startups die all the time, for all sorts of reasons, but "large company started competing with it" is a rather rare one.

[mbesto]

> A lot of comments here saying that this is the death knell for Plaid, but I wouldn't be so sure.

The fact that no one has mentioned Yodlee once makes me think very few people in this thread actually know this market.

Not to mention there are probably quite literally thousands of Stripe competitors.

The world is a big place and the SV bubblethink of "only one person wins" is highly myopic.

Plaid will be fine. They just might not be $10B valuation fine. That's all.

EDIT: Doh, saw Yodlee mentioned once much lower in thread now.

[ericlewis]

As someone intimately familiar with the space Yodlee is often not mentioned for a few reasons. But their ability to handle the long tail of accounts is the biggest reason it’s often not considered. When you’re trying to maximize the ability for people to access your service (and require a bank account or related information to do so) there is really no comparison to plaid. Further, plaid long had a better developer experience than yodlee and for some amount of time was much cheaper (thanks to enterprise discounts in exchange for prominence in leading apps)in a few cases.

People likely aren’t familiar with yodlee and maybe it’s an indication they’re not aware of the space. But for those who are - I’d not have mentioned it either, they barely touch plaid on many facets.

Edit: I know in one particular instance we “considered” yodlee at what is now considered a big consumer banking unicorn. But in no way was actually considered. It was used as a negotiation tool.

[mbesto]

> But their ability to handle the long tail of accounts is the biggest reason it’s often not considered.

According to a few google searches, this is simply not true:

Yodlee supports 14k banks and Plaid supports 11k banks

I personally have used Kubera as a consumer who use both Plaid and Yodlee to connect to banks. I have ~3 bank accounts (that represent great candidates for long tail) that simply do not work with EITHER service provider. Having spoken to the support staff from Kubera (who are great btw) they prefer to use Plaid for some of the reasons you mention, but not because its all encompassing.

Categorizing Yodlee as "they barely touch plaid" is simply disingenuous. Much of the functionality is comparable, it's just MUCH better on Plaid for both the developer and the consumer in terms of overall experience.

[eins1234]

Re: Plaid itself, having used some of their competitors, specifically Finicity, which is one of Stripe's partners for this, I wouldn't be too worried if I were them.

IME, Plaid is lightyears ahead in terms of reliability and overall developer and end-user experience.

[sangnoir]

> IME, Plaid is lightyears ahead in terms of reliability and overall developer and end-user experience.

I regret handing over my bank login details to Plaid, since they scraped my bank statements without stating so upfront and offered that information to 3rd parties (indirectly, as a score of some sort, IIRC, but that's scraping very personal information).

When I mentioned it on HN a year or 2 ago, someone who works there denied the practice - might have been a cofounder. A few months ago I was contacted regarding a settlement for a Plaid class action suit regarding the very actions that had been denied on HN. Plaid - never again.

[SpikedCola]

This is something that absolutely blows my mind. The bank's terms of service say you can't hand over your password to anyone, and here is Plaid asking users for their login information! I can't understand how they can ask users to breach terms of service, and have gotten so far.

[sangnoir]

I was dubious while doing it - and I gave Plaid access to my settling account. I consoled myself that they wouldn't have by income, but then realized they had a copy of my financial information on spending, which is equally bad, or worse than just income. I ought to have been more vigilant - I unfortunately fell into the trap of thinking that banks are slow-moving and reluctant to develop API access.

I will never sign up for a service that requires Plaid.

[chrisseaton]

Why isn't it done through a proper API that you grant them a token for?

And how do Plaid bypass your bank's per-login 2FA if they're logging in as is they were a user?

[pbreit]

Because not many banks have such an API.

Plaid just relays the 2FA question.

I think websites would have a tough time preventing users from sharing username/password. It would certainly be acceptable in a power of attorney situation.

[jjeaff]

Not only do they relay the login and 2fa information, they even show the login as a window with company colors and branding as if you were logging in to your bank directly.

[milesskorpen]

The class action was settled — which just means it was cheaper to settle than to fight in court, not that there was wrong doing. This kind of thing is very very common, and the settlement amount was modest in the scheme of class action settlements.

Plaid is pretty clear in their privacy policy that they DO NOT repackage and resell data (they do sell data - as in, when you use Plaid to give your banking info to a mortgage broker, the broker is paying Plaid for your data, but it is at your explicit request).

If banks didn't want Plaid to do screen scraping, they could build APIs. Some are now. But they've been VERY VERY reluctant to do so, because they want to hold customers (us!) hostage to their services and make it painful to go anywhere else to get financial services. I appreciate that Plaid figured out how to break their stranglehold, which has directly enabled the current blossoming of FinTech apps ... even if they had to do so in a way I don't love.

[sangnoir]

> If banks didn't want Plaid to do screen scraping, they could build APIs.

Even if we take this as a given - what if customers don't want Plaid to scrape their data? I only used Plaid to verify that I own the bank account - but they went out of their way to scrape my transaction information, just because they could, and that data is valuable - that is messed up. I'm sure if my bank had an API, Plaid would still have hoovered up my transaction information, so the "API access vs Scraping" debate is a sideshow.

If they hadn't scraped my transaction information, I wouldn't have been part of the class, but they chose to maximize data collection. If it had been Facebook or Google that harvested financial info the way Plaid did, no one would be saying "Their TOS is clear about it". Additionally, any big tech company can purchase Plaid and get that data (I can't remember if the settlement has a provision for deletion of that data).

[geraldwhen]

Not really. APIs cost money to develop, and people, and time. And many large banks don’t have robust engineering teams that can tackle new challenges or support a public api.

Small projects as you might imagine are measured in years, not days or months.

An API like you describe could take half a decade to build, at the cost of hundreds of millions of dollars. These are not fake numbers or estimates. This is what it would cost.

When you think big finance, think government.

[jjeaff]

I think you will rarely see a class action settlement for a complaint with zero merit. Otherwise, you would be getting a tiny class action check in your mailbox every week for every company that prefers to settle rather than fight.

Do you also believe that Bill O'Reilly and Fox News paid out $30m+ even though he didn't do anything? After all, they admitted no wrong doing in the settlements.

[mikeodds]

I got the same answer from them on HN here 10 months ago

https://news.ycombinator.com/item?id=27467797#27476452

Me: Always been curious - do you (Plaid) use the transaction data or any other data obtained from customers logins for anything other than the reason the customer supply’s their credentials? I.e if I use plaid to link to my Robin Hood account, do you in any way sell/share/use my data apart from allowing me to fund my Robin Hood account?

Response: Good question! No, we don't. Our official statement on this is at https://plaid.com/how-we-handle-data/ "Plaid only shares your data with your consent. We don’t share your personal information without your permission, and we don’t sell or rent it to outside companies."

[Terretta]

That could easily be a lawyer-speak official statement.

They say “personal information”. That is consumer-facing language for something which in banking has a legally (regulation) defined term: “PII” or personally identifiable information:

https://www.investopedia.com/terms/p/personally-identifiable...

It can be argued that lists of money spent at stores cannot be reversed back to a person without other information. So they might not consider your transactions PII.

As for the consent, the TOS click wrap generally gets your consent, in the part where firms mumble about “our partners” for “legitimate uses” or etc. while bucketing various data brokers in that class.

[propogandist]

change your bank account password, if you've ever used Plaid or any of the comparable services to verify your account, to prevent the scraping. They are constantly hitting bank accounts where they have access.

[fishywang]

>IME, Plaid is lightyears ahead in terms of reliability and overall developer and end-user experience.

I don't know. Maybe their competitors are even worse? But as an end-user Plaid's experience is far from good.

In their current state I really don't have much control after I gave them my banking credentials. If they can actually implement an OAuth model for their paying customers (e.g. third parties using their API) to integrate, that would be much better. For example when I try some service first I can initially only give it access to one of my least important account, evaluate whether it's useful, before giving it more of my accounts, or revoke it access to all my accounts. I have none of those controls today.

I actually also interviewed with them several years ago, and mentioned the OAuth thing during my interview. But after all these years it's still not there. If I have to _guess_ why, I would guess they fear such a feature will make their paying customers (API users) less likely to use their product, e.g. their business model actually relies on the end users' lack of control.

[Aeolun]

The product maybe, but my interactions with the company have been dissapointing.

[mupuff1234]

> Meanwhile both Brex and Ramp have become decacorn

That was at peak bubble. Given that most public Fintech companies crashed around 70% from ATH I assume the same holds for Brex and Ramp.

[B1FF_PSUVM]

So now they're just tricorn?

(The things you can do to/with a language ... "Verbing weirds language" takes the cake, though.)

[webmaven]

> Startups die all the time, for all sorts of reasons, but "large company started competing with it" is a rather rare one.

Currently rare, but as with other aspects of the market for startups, it is irregularly cyclical. Not too long ago Google accidentally killed startups every so often when they launched a product or service (the examples that particularly stick in my mind are Kiko - killed by the launch of Google Calendar in 2006, and Autumn AI, killed by Tensorflow around 2016). Go back to the pre-www days and Microsoft killed startups regularly, seemingly as a competitive sport. A reasonably entertaining account of one such is Startup: A Silicon Valley Adventure, by Jerry Kaplan, founder of Go Corporation:

https://books.google.com/books/about/Startup.html?id=UK7qDwA...

[smugma]

Companies fail all the time when large companies start to compete with them. Microsoft was famous for this in the 90s. The biggest example that comes to mind is Netscape.

[pbreit]

There will be plenty of users (skewing large) who will want to get account validation and payment processing from different vendors.

I wonder if this opens up a Visa acquisition?