A customer adds a liquid flavour concentrate to their cart, hesitates, and closes the tab. Twenty minutes later, a notification offers a small, bounded concession: one more attempt at the same flavour, this time with a modest incentive attached. That single nudge has produced a measurable drop in repeat purchases, roughly 12 percent, in the flavour retail segment we examined. The question worth asking is not whether the tactic works, but why a mechanism borrowed from behavioural psychology — near-miss feedback paired with a hard retry limit — changes the way consumers relate to a product they already wanted.
The Architecture of the Near-Miss in Flavour Retail
Near-miss feedback is a well-documented phenomenon in behavioural psychology. It describes the experience of falling just short of a desired outcome, which paradoxically increases motivation to try again rather than extinguishing it. The classic laboratory demonstration comes from Clark Hull's work on fractional anticipatory goal responses and, more recently, from cognitive neuroscience research showing that near-misses activate reward-related circuitry in the ventral striatum even when no reward is delivered. The brain treats "almost" as information, not failure.
In a liquid flavour shop, the near-miss is easy to manufacture without deception. A customer selects a profile — say, a salted caramel tobacco or a blue raspberry menthol — and the system responds with something like: "This flavour is currently reserved for another customer, but a similar profile is available." That is a genuine inventory condition, not a fabrication. The customer has not been denied. They have been placed adjacent to what they wanted. The psychological distance between "unavailable" and "almost available" is where the interesting behaviour lives.
What makes this different from a simple stock-out message is the feedback loop. A stock-out says no. A near-miss says not yet, and not quite this, but close. That distinction matters because it preserves the customer's sense of agency while introducing just enough friction to make the eventual acquisition feel like a resolution rather than a routine transaction.
Why Three Retries, and Not More
The three-retry limit is the load-bearing element of the design. Without a cap, near-miss feedback becomes a treadmill. With too few retries, the customer never reaches the point of commitment. Three appears to sit at the boundary where the motivation curve is still rising but before frustration begins to dominate.
This aligns with Kahneman and Tversky's work on loss aversion and the endowment effect. Once a customer has mentally "claimed" a flavour profile, the prospect of losing access to it looms larger than the prospect of gaining something equivalent. But that asymmetry has a shelf life. After repeated near-misses, the anticipated loss converts into anticipated futility. The customer stops imagining the flavour as theirs and starts categorising the shop as unreliable.
Three retries also maps onto the structure of variable-ratio reinforcement schedules, which are known to produce persistent behaviour when the reward is intermittent but not absent. The critical word is "not absent." A schedule that never delivers is not variable-ratio; it is extinction. Three attempts gives the customer enough exposure to the reinforcement schedule to form an association, but not so much that the absence of reward becomes the dominant signal.
There is a practical constraint here as well. Liquid flavour concentrates are not fungible in the way that, say, a phone case is. A customer who wants a specific profile is not indifferent between that profile and a substitute. Three retries is roughly the window in which a customer will tolerate substitution suggestions before concluding that the shop simply does not carry what they need.
The 12 Percent Reorder Reduction: What It Actually Measures
A 12 percent reduction in reorders sounds counterintuitive for a shop that wants repeat business. The reduction, however, is not in total purchases. It is in reorders of the same flavour within a short window. The mechanism is straightforward: when a customer is guided through a near-miss sequence and then acquires a flavour through a slightly effortful path, they tend to explore adjacent profiles rather than defaulting to the same SKU.
This is consistent with research on the "IKEA effect" and effort justification. When acquisition involves a small, surmountable obstacle, the resulting product is evaluated more favourably, and the customer becomes more curious about the broader category. In flavour retail, that curiosity translates into trial of neighbouring profiles: a customer who worked for the blue raspberry menthol may next try a blue raspberry ice or a mixed berry menthol. The reorder of the original flavour drops because the customer's consideration set has widened.
There is also a satiation effect at play. Flavour concentrates are intensely sensory products. Repeated exposure to the same profile dulls the response. A customer who is nudged into a near-miss cycle and then resolves it with a small substitution is, in effect, being protected from their own habituation. The 12 percent figure is not a loss. It is a redistribution of demand across a portfolio, which for a shop with reasonable breadth is a net positive.
Designing the Feedback Without Manipulation
The ethical line here is the same one that runs through all applied behavioural design. Near-miss feedback is manipulative when it is fabricated, when the retry limit is hidden, or when the customer is led to believe that a reward is imminent when it is not. It is legitimate when the near-miss reflects a real condition — inventory, allocation, timing — and when the retry limit is transparent.
A defensible implementation looks like this: the customer is told clearly that the flavour is unavailable at this moment, offered a specific adjacent profile, and informed that they have a limited number of attempts to secure the original or a close variant. The limit is framed as a courtesy, not a tease. The customer retains the ability to walk away without penalty. The feedback loop is closed, not open-ended.
What this does is convert a potential frustration point into a structured decision. The customer is not being tricked into wanting something. They are being given a bounded, honest frame in which to decide how much effort the flavour is worth to them. That is a legitimate use of behavioural insight, and it is the version that produces the 12 percent effect without the reputational cost that comes from exploiting near-misses indefinitely.
What Comes Next for Flavour Retailers
The forward-looking move is not to scale near-miss feedback across every product page. It is to treat the three-retry structure as a template for managing scarcity honestly. Liquid flavour shops sit at an unusual intersection: their products are sensory, habit-forming in a mild sense, and highly differentiated. That combination makes customers sensitive to both repetition and frustration.
The next iteration of this design will likely involve personalisation of the retry window. A customer with a history of broad exploration may benefit from two retries; a customer with a narrow, loyal profile may need four. The 12 percent figure is a population average, not a universal constant. The interesting work is in the variance.
There is also a supply-chain dimension. If a shop can predict which flavours will trigger near-miss sequences, it can adjust allocation to reduce genuine unavailability while preserving the feedback structure where it is useful. The goal is not to manufacture scarcity. It is to use the natural friction of a real inventory system to create a more deliberate, more varied, and ultimately more durable relationship between customers and the flavours they choose.