Most fintech teams answer flat retention by adding features. A product method called Next Move Theory says the growth usually lives somewhere else.
Every fintech roadmap I see has the same shape. Round-ups. A budgeting tab. A slicker card. A referral loop. The team ships all of it, and three quarters later retention is flat and no one can say which feature was supposed to move it.
The problem is rarely the features. It’s that the team keeps working on one small part of what the customer is actually trying to get done, adding more of it, while the growth move sits somewhere they never looked.
This product method has a specific way to find that place. It starts by drawing the customer’s job graph.
A customer never opens a banking app to “use a banking app.” They open it mid-chain: they got paid, they want to know rent will clear, they want to dodge the overdraft, they want to set something aside, they want to stop feeling anxious on payday. Those jobs sit in a hierarchy — small mechanical ones (check the balance) under bigger ones (stay in control of my money) under bigger ones still (feel secure about the future). The job your product fully performs is its core job. Everything above carries the motivation; everything below is where daily experience lives. Once you can see the whole graph, growth stops being “what feature next” and becomes “which move on the graph.” Here are five that fintech teams routinely miss.
1. Kill a job outright.
One of the highest-leverage moves isn’t adding something — it’s deleting a job the customer used to do. The brain values a removed job at roughly twice an equivalent feature. Apple Pay killed “reach for my wallet.” Auto-pay killed “remember to send the card payment by the 12th.” Square’s QR tipping killed “find cash to leave a tip.” Ask which job you could make disappear, not which screen you could add.
2. Take the job off the customer.
Sometimes the job must happen, but nothing says the customer has to do it. Wealthfront doesn’t help you rebalance your portfolio; it rebalances it for you. The job stays; the work moves to the product. In fintech, where every manual step is also a chance to make an expensive mistake, quietly performing the job the customer dreads beats any dashboard that helps them do it themselves.
3. Close the time gap.
Wait-time between jobs is felt as cost, and below certain thresholds it changes the job entirely. Stripe shipped instant payouts where banks took two to three days — same money, a different product to the nervous system. Find the longest gap in the chain (settlement, approval, verification) and treat compressing it as a feature in its own right.
4. Design out the negative emotion.
Money is the most anxious category most people touch, and emotion is the brain’s read on how a job is going. Removing a bad feeling registers at about twice the value of adding a capability. TurboTax removed “am I doing this wrong” with an accuracy guarantee. Insurance sells peace of mind, not fewer claims. Wealthfront sells “you’ll be OK in retirement,” not “diversified asset allocation.” Find the point in your flow that produces doubt or shame and remove the feeling, not just the friction.
5. Move up a level.
The most powerful move, when it’s available, is to climb: turn the big job above your core job into your new core job. It only works once you perform your current job completely — but when you do, the motivation lives one level up. A product that helps people avoid overdrafts can climb to helping them feel in control of their money. The second is a bigger job, and a bigger business.
None of these are features. Each is a different kind of move, and the reason teams miss them is structural: when a product stalls, the reflex is to tune the core job — polish it, add to it — because that’s the part you can see. The step-changes almost always live somewhere else on the graph, in a job you haven’t drawn yet.
So before the next planning cycle, map the slice of the job graph around your product and ask which move — kill, take-off, close-the-gap, design-out-the-emotion, climb — the team hasn’t considered. The answer is usually not “another feature.”
