Thread by @patio11

An interesting consumer credit card offering: petalcard.com

The most unique thing about it is the schedule of cashback, which gets more lucrative the longer you have the card (to a cap), contingent on on-time payments.

(If you follow me on Twitter it is unlikely this is competitive with your best options, since it purports to target users who are not traditionally underwritable for credit, but it’s interesting how they must be modeling where the margin comes from.)

The consumer credit industry has middling to high sophistication with respect to customer segmentation by behavioral profiles, and so would look at what the product features of a card would accomplish over a portfolio of blended customer archetypes.

One reason why you see “reward you for your loyalty” sort of schemes less with respect to the top end of the market is that one *extremely lucrative* subgenre of credit card user a) spends a lot and b) will switch providers based on who has the best offer.

So if you are designing a product for credit card users who have options, rear-weighting the rewards for the card tends to make it non-competitive with other offerings which frontload them (via e.g. a signup bonus). You’d thus not do this if that customer archetype mattered.

Many smart geeks of my acquaintance are in this archetype and marvel about the persistent stupidity of bank marketing and product departments, believing the banks to continually lose money on their custom.

This... is not the smartest belief of my acquaintances.

There is always a spreadsheet.

There are a lot of startup founders of my acquaintance who are interested in e.g. banking the unbanked or otherwise increasing access via technology. This is a worthy aspiration.

It is tricky to build a business around, and people get surprised by how it is tricky.

If you offer a product for people in economically precarious situations with the idea that consumption of your product will improve their situation and yay all good all around, you may be building a transitional product.

Transitional products are tricky.

For sake of illustration, consider a toy model which has two subgroups of consumers: Very Desirable Customers and Everyone Else.

If you have a widely available credit card product, it will almost by definition not be competitive with options available to VDC.

There exist many providers who specialize in the needs of EE. You look at their pricing, policies, etc, and consider them odious. You’re going to do right by your customers, darn it, and make enough money to keep the lights on by durably improving lives.

Suppose hypothetically your company successfully executed on this mission for 20% of your customer base, and the margins from their accounts subsidize your services to everyone else.

Bad news: by successfully working with you, you’ve flagged that 20% as VDC.

“Flagged? Flagged how?”

Oh remember that part where you’d underwrite people without material credit history? You know what they have after using your card for a year? Credit history.

And thus the (extremely sophisticated and well resources) firms serving VDC will get in touch with them, plausibly using information (addresses, etc) that you reported to the credit bureaus, and say “Why not switch to our product? It is better for you in every way.”

And here is a hard realization: the big bank marketing department telling your best customers this? They are telling your customers God’s honest truth. They factually are offering better terms than you are, plausibly across the board.

There was a time in which financial institutions had relatively undifferentiated offerings and so the covert subsidy by your VDC of your EE was laundered through the bank. You can still find this in some markets. (Japan is an interesting example.)

And thus folks building products for underserved folks find that they generally need to make the economics work on the population which *does not* transition out from being underserved.

And then the mission... sounds a lot less fun.