The flavour wall at a vape shop presents an odd experimental condition: almost every customer arrives intending to buy, and almost every customer leaves having chosen between fewer options than they expected. Walk into a well-stocked liquid flavour shop and you might see two hundred SKUs, but the decision that actually determines the purchase is usually binary — this bottle or that one, salt or freebase, the familiar or the new. What happens in that final comparison is where behavioural psychology gets interesting, because the way people weigh a two-option flavour choice is not symmetric. Losses loom larger than gains, and in a category built on sensory reward, that asymmetry produces some counterintuitive outcomes.
The Two-Option Problem in a Saturated Category
Classic choice research suggests that adding options increases the likelihood a consumer will buy something, up to a point, after which the cost of comparison starts to suppress commitment. The flavour aisle sits well past that point. Retailers compensate by narrowing the final decision themselves: staff recommendations, "staff pick" tags, a two-for promotion that pairs a known profile with an unfamiliar one. The customer experiences the narrowed set as a favour, but it also changes the structure of the decision from a search problem into a comparison problem.
Once a choice is framed as a comparison between exactly two items, a different set of mechanisms takes over. Amos Tversky and Daniel Kahneman's work on prospect theory established that people evaluate outcomes relative to a reference point rather than in absolute terms, and that losses relative to that reference point carry roughly twice the psychological weight of equivalent gains. In a two-option flavour decision, the reference point is rarely "no purchase." It is the flavour the customer already knows and likes — the one they came in for.
That reframing matters. If a regular customer's reference point is their usual blue raspberry, then the alternative bottle on the counter is not a potential gain. It is a potential loss of the experience they already know they enjoy. The unfamiliar flavour has to clear a higher bar than simple novelty-seeking would predict, because the downside — a bottle that tastes wrong, that sits unused, that wastes eight dollars and a week of anticipation — is weighted more heavily than the upside of discovering something better.
Why the Incumbent Flavour Has an Unfair Advantage
This is the mechanism behind a pattern that shop owners notice but rarely articulate: in a straight two-way comparison, the incumbent flavour wins more often than its objective quality justifies. The challenger isn't competing against the incumbent's actual taste profile. It's competing against the incumbent's taste profile plus the customer's aversion to losing a known-good outcome. That's a substantial handicap.
There's a second layer. Richard Thaler's work on the endowment effect showed that people assign higher value to things merely because they possess them. A customer who has bought the same liquid for six months has, in a small way, taken psychological ownership of that flavour. It is now part of their identity as a customer. Asking them to switch isn't asking them to try something new; it's asking them to give something up.
Variable Reward and the Flavour That Surprises You
Set against loss aversion is the opposing force that makes flavour retail genuinely strange: the reward structure of vaping is not fixed. Nicotine delivery, throat hit, and flavour perception all vary with device, wattage, coil age, and even what the person ate an hour earlier. A single bottle does not produce a single consistent experience. This is, functionally, a variable-ratio reward schedule — the reinforcement pattern B.F. Skinner identified as the most resistant to extinction, and the one that drives persistent, sometimes compulsive, behaviour.
Variable-ratio schedules are usually discussed in the context of problematic consumption, and that discussion is legitimate. But there's a less examined consequence for choice architecture. When the reward from any given bottle is uncertain, the variance of the experience becomes part of what the customer is evaluating. A flavour with a wide range — sometimes excellent, sometimes mediocre — carries both upside and downside. A flavour with a narrow, reliable range carries almost none.
Loss aversion predicts that customers will overweight the downside variance. They should gravitate toward consistency. In practice, they often don't, and the reason is that the reference point shifts with experience. Once a customer has had one genuinely exceptional session with an unfamiliar flavour, that session becomes the new reference point, and every subsequent bottle is measured against it. The unfamiliar flavour has now acquired its own loss-aversion problem — it can only disappoint relative to that memory.
The Two-Option Flip
This is where the flip happens. Consider a customer comparing their reliable daily flavour against a new one they tried once and loved. At the moment of the first comparison, the incumbent had the advantage: known-good versus unknown. At the moment of the second comparison, the new flavour has the advantage in one dimension and a severe disadvantage in another. It has demonstrated a higher ceiling. It has also demonstrated that it can fail to reach it.
The result is a decision that looks irrational from the outside. The customer buys the new flavour, dislikes the batch, and returns to the incumbent — then buys the new flavour again two weeks later. They are not being inconsistent. They are running a comparison between a certain moderate outcome and an uncertain outcome with a higher peak, and the weighting they apply to each flips depending on which recent experience is salient. A single disappointing session with the incumbent — a burnt coil, a muted batch, a flavour that just doesn't hit that day — is enough to make the incumbent's certainty feel like a liability rather than an asset.
What This Means for How Flavour Decisions Get Made
The practical implication is that two-option flavour comparisons are not stable. They are reference-point dependent, and the reference point moves with every session. A shop that presents a binary choice is not presenting a fixed decision; it is presenting a decision whose terms shift based on what the customer experienced most recently.
For anyone thinking about this intersection — retail design, behavioural research, or just the odd pleasure of watching people choose — the useful question is not "which flavour is better." It's "what is the customer measuring against right now." A customer whose last session was excellent is comparing against a high bar and will be loss-averse about leaving it. A customer whose last session was flat is comparing against a low bar and will be far more willing to take the uncertain option, because the certain option has already failed to deliver.
This also explains why sampling programs in flavour retail have effects that outlast the sample itself. A taste test doesn't just inform; it installs a new reference point. The customer now has a concrete memory to compare against, and that memory competes with the incumbent's accumulated history. Whether it wins depends on how vivid it is and how recently it was formed — recency and availability doing the work that product quality alone cannot.
The forward-looking version of this is that flavour retail is likely to move toward shorter comparison cycles and more deliberate reference-point management: smaller bottle sizes that lower the cost of an uncertain trial, rotating limited runs that reset the reference point on a schedule, and pairing structures that bundle a known quantity with an unknown one to make the loss feel smaller. None of this requires deceiving anyone. It requires understanding that the two-option comparison at the counter is never just about the two liquids. It is about what the customer is afraid of losing, what they are hoping to gain, and which of those two the last session made more salient.