9 of the 10 largest issuers spent $6.96 billion winning customers in a
single quarter. Here is what each account cost, and the one number I did not expect.
By Anuj Shahani
There is a silver tray by my front door that was bought to hold keys. Lately it has been
holding credit card offers. This month I stopped tipping them into the recycling and started
counting instead. 11 credit card offers, 4 checking offers, and 1 very confident letter about
a personal loan I have never once considered. All of it addressed to me by name.
That tray is my day job arriving at my own front door, and it made me want to put a number
on it. So I built one. Across the 10 largest United States card issuers, acquisition marketing
reached $6.96 billion in the second quarter, up from $6.21 billion in the
first. That is 12% more spending in 90 days.
Then the interesting part. Accounts acquired grew 6%. So the cost of winning a new customer
rose about 5%, from $184 to $193.
Exhibit 1Spending against
accounts acquired
Spending does not always directly correlate to Accounts Acquired. And that Gap is the
story.
Each issuer, spending on the left and accounts won on the right. Change
from Q1 2026 to Q2 2026.
* Citi reports $283M of advertising and marketing. It carries the rest of
its acquisition cost inside revenue, where partner payments and new account acquisition costs
sit, so I count a documented share of that line and show Citi at $415M. Both quarters are
built the same way. The full derivation is in the report.
Capital One raised spending 11% and accounts 9%, so its cost per account barely moved.
Bank of America raised spending 38% against 9% more accounts across all products, and 13% more
card accounts, its strongest card quarter in 6.
Exhibit 2Where the
acquisition dollar goes
The number I did not expect.
I assumed marketing was most of what an account costs. It is not. Take the median card with
no annual fee and open up the full cost of putting it in someone's wallet. Marketing is 47
cents of the dollar. The sign up bonus is 39 cents. Identity checks and onboarding make up the
rest.
Every square is 1 cent of the median acquisition investment behind a card
with no annual fee.
The point
The offer costs almost as much as the media. Media gets reviewed every
quarter. The offer gets reviewed once a year.
So check the offer first. If your cost per account moved this quarter, the bonus is as
likely a cause as the media buy.
The quarter3 things worth
knowing
3 things worth knowing.
01
Credit got better while spending went up
Card loss rates improved at 7 of the 10 issuers. Capital One cut its rate
by 39 basis points. Chase lowered full-year guidance to about 3.2%. Paying more for accounts
you expect to be worth more is a very different story from paying more for the same
account.
02
2 issuers got cheaper
Wells Fargo and U.S. Bank both held spending flat while accounts kept
growing, so the cost of a customer came down. Both had stepped up earlier in the year. That
is what it looks like when a build starts paying.
03
Some accounts arrived without anyone winning them
3 books of accounts changed hands this quarter, including more than 2
million American Airlines accounts moving to Citi. Bought accounts carry no acquisition
marketing. I removed these from my calculations to keep the comparisons like-for-like.
04
Barclays US is still to come
Barclays publishes first half results on 28 July, so its second quarter
figures sit outside every number here. It stays in the panel roster and joins the aggregates
once the results are public.
Exhibit 3Cost against
value, every issuer
Where the annual fee earns its keep.
A card that charges an annual fee returns about $2 for every $1 it costs to
acquire. A card with no annual fee returns about $1.10. The fee lifts revenue
per account and lifts retention at the same time, and that is the whole gap.
Here are all issuers in a single chart. The X-axis is what an account costs. The Y-axis is
the Lifetime Value of the customer or cardmember over 5 years. The hollow circle is where each
issuer sat in the first quarter, and the arrow shows where it moved.
Acquisition investment against 5 year contribution value, per account.
Circle size is accounts acquired. Value assumptions are held steady throughout the year, so all
movement reflects cost.
Barclays US publishes on 28 July and sits outside the chart.
That is the case for the premium repricing running through Chase, American Express and Citi.
It is also the reason a card with no fee needs a second product behind it to be worth
opening.
For your leadership team3
questions
3 questions for your next meeting.
1
Your cost per account
moved this quarter. Can you say whether the cause sits in what you spent or in what
converted?
2
Marketing is less than
half of what an account costs you. When did your team last review the offer with the same
rigour as the media plan?
3
Cards with an annual fee
clear breakeven across the whole industry. What is your plan for the accounts that do not carry
one?
Acquisition Compass
Cost of an Account, Q2 2026
16 pages on what a new customer costs at the 10 largest issuers, what each
one is worth, and the 3 adjustments that make the comparison fair.
You are receiving this as part of a community of 17,000 financial services
leaders shaping the industry’s future. Reply anytime. I read every response.
