I have bought media for Visa, PayPal, Nasdaq, JPMorgan Chase, and Blackstone at different points across two decades, and the most useful thing I can tell you about financial services marketing is that it is not one discipline. The consumer side is a performance machine wearing a compliance harness. The institutional side is a trust operation where the media plan mostly exists to make ten thousand very specific people slightly more confident before a meeting. Companies that run both under one strategy usually run both badly.
The failure has a predictable shape, because one side always wins the internal argument and imposes its physics on the other. When the performance culture wins, the institutional program gets graded on click-throughs and form fills, discovers that treasurers do not fill out forms, and gets defunded for missing targets it should never have carried. When the brand culture wins, the consumer program runs beautiful awareness work while competitors with activation-optimized funnels quietly take the accounts. Same company, same quarter, both sides underperforming, and the diagnosis in the QBR is always execution when the actual problem was a single strategy asked to govern two different businesses.
The B2C side: performance in a harness
Consumer finance is one of the most measurable categories in advertising. Applications, funded accounts, transaction activation: real events with real economics, which makes it tempting to run the standard growth playbook at full speed. Two things stop you. The first is compliance review, which turns creative iteration from a daily loop into a weekly one; the practical answer is building a pre-approved modular system, claims, disclosures, frames, cleared once, recombined endlessly, rather than submitting one-off ads and waiting. The teams that industrialize compliance ship five times the variations at the same legal cost, and in a category where creative volume is now the targeting, that throughput difference is a structural advantage no bidding strategy can match. Compliance is not the enemy of iteration. Unindustrialized compliance is.
The second is that acquisition quality varies more than acquisition cost, and in banking the variance is extreme. A funded account from a rate-chaser who leaves in six months is not the same asset as one from a customer who consolidates, the same cohort math that governs every subscription business; the first destroys value after acquisition cost, the second compounds for a decade. That is the CPL problem with a balance sheet attached, and the fix is the same: optimize the auction to a downstream event, activation, deposit thresholds, ninety-day retention, even if the platform howls about signal volume. Cheap accounts are the most expensive thing a bank buys, and a bidding algorithm fed on application volume will fill the book with them at industrial efficiency.
The B2B side: reach the room, then earn it
Institutional and B2B financial marketing inverts everything. The audience is small enough to name. Sales cycles run quarters or years. Nobody converts from an ad, and the ad was never supposed to convert them; it was supposed to make the brand a safe answer inside a committee you will never see. That phrase is the entire strategy, because institutional purchases are decided by career risk: the treasurer shortlisting providers is optimizing for the choice nobody gets fired for, and sustained, credible presence is how a brand becomes that choice. Media here is precision reach into professional contexts, LinkedIn done properly, the trade and financial press, the events calendar, sustained at a frequency that reads as institutional stability rather than a campaign. Going dark between pushes costs more than it saves, for reach-mechanics reasons I have covered before, and in this category the dark period itself is a signal; an institution that vanishes from its own trade press for two quarters gets noticed vanishing.
Content carries most of the B2B weight, and it needs to be written for the second reader: the analyst who forwards it to the decision maker. Which means specificity over polish, a point of view over coverage, the chart worth screenshotting into someone else's memo. And increasingly it needs to be structured for machine retrieval, because when a treasurer asks an AI model to summarize providers in a category, the sources it cites become the shortlist. That is GEO applied to the most valuable queries in the world, and financial B2B is the category with the most to win there, because the query volume is tiny and the deal sizes are not. A handful of citations in the right synthesized answers is worth more than a rankings dashboard full of green, and almost nobody in institutional finance is competing for them yet.
The one discipline both plans share
The two plans share one thing: measurement honesty, and each side fails it in its own dialect. B2C has enough data to fool itself with attribution, platform-reported returns crediting accounts that were coming anyway; B2B has so little that people fill the gap with anecdotes, the deal that closed after the conference becoming proof the sponsorship worked. Both are cured by incrementality thinking, deciding in advance what evidence would prove the media worked, then building the test rather than the story. On the consumer side that means holdouts against the activation events that matter. On the institutional side it means honest pipeline penetration of the named audience, tracked over quarters, against the list of accounts the program exists to move. Different instruments, same discipline, and the discipline is the only thing the two media plans should ever be forced to share.