am3101

I have for years synced all my financial data through aggregators (Mint for a long time until it was shut down, YNAB for a bit, Monarch Money now, and have experimented with a dozen others along the way). I probably have 20 accounts across credit cards, bank accounts, investment accounts (traditional and crypto), retirement accounts, and others (HSA, transit/commuter card).

Something I have anecdotally noticed is that the institutions that struggle the most (whether initial connection setup, reliability, or accuracy) are almost all "digital native" financial institutions.

Cash App is the worst offender. First, it requires consistent MFA re-authorization, so my account gets disconnected every few days. That's a security design decision not necessarily a bug, although it feels like you could solve this with app-specific authentication (an aggregator logging in to check a balance should probably get more grace on MFA than a user logging into the app). More importantly, the API seems to spit back a ton of garbage. Every time I spend money at a vendor with a Square loyalty program, the "reward" shows up as a $0 transaction. This may be a Monarch thing, but their support engineers claim that they are unable to distinguish loyalty transactions from others. Finally, and most blatantly, the actual process of doing MFA is crazy. Cash has historically texted an authentication URL to your phone, but that makes it hard for the aggregator to authenticate. Monarch (or rather, Plaid's) solution is to make the user copy the URL and paste it into the MFA box in Plaid (which is almost impossible on an iPhone since long-pressing on a link to copy it pops up a preview, which triggers the authentication, which obviously can only happen once).

Cash is the most egregious to me but there are a lot of other examples. I've had all three rounds of Bilt's card, and when it was on Evolve, aggregators consistently had issues connecting (haven't used in a couple years so maybe it's changed for the better). I likewise quit Marcus because, incredibly, Bank of America's decades-old tech stack somehow provided better connection reliability. My recollection is Coinbase used to be pretty bad, though in fairness to them they've gotten a lot better.

I've seen a similar thing this year with taxes. TurboTax has a feature where, via Plaid, you can import tax forms from various institutions. Robinhood (which actually works fine on Monarch, though it did not always) had a one-click option, but for some reason, the one-click import became active a week or so after my 1099 was emailed to me. Worse, when it finally became active, it failed to load multiple times and when it did finally load (days after promised) it imported it incorrectly. I ended up just typing in the 1099 manually.

It would be one thing if all my accounts had this issue, but Bank of America, UBS, Vanguard, HealthEquity, Alight, Morgan Stanley, Wells Fargo, Amex, Chase, and even the NY State 529 provider all sync flawlessly. The whole point of these "digital attackers" is that technology should be changing how people do financial services, but the incumbents have all this legacy tech debt that make it impossible for them to offer the same quality of service.

Why is this? Is it that there is just so much tech to build if you are a greenfield financial services firm that you don't have time to maintain a stable API? Is it a desire to build customer lock-in so that your customers come to you, not to an aggregator? Is it that underneath the shiny hood, all these firms' tech is actually quite bad?

(As a side note, shoutout to Venmo and PayPal who, for all their many issues, have seemed incredibly reliable in this area.)

Non-Googler here who is curious—news reports tend to say that tech talent is fleeing New York/SF and moving to Austin/Charlotte/etc. But I have a couple of Google friends who work in other offices and have had a really tough time trying to transfer to New York, and have said they anecdotally heard New York is the most popular office (allegedly something like 1,000 more people transferred into New York than transferred out).

This is anecdata... does anyone have any "real" data on this? Seems like Google would be an interesting bellweather for the future of legacy urban centers like New York, at least from a tech perspective.

I work for a corporate innovation group in a slow-moving industry (think similar to financial services). We've built a couple of software products that have been pretty impactful to our parent, but the parent is getting tired of spending a few million per year on software when it's not their core business.

What is the path for an innovation group getting the capital to become independent? I've heard venture firms generally don't like investing in corporate carveouts since it's pretty complex. Are there any firms that are more open to this? Any examples of successful spinouts that are pretty early stage (I think we'd be Series A/B if we were totally independent)?