Concept visual: Wild Bite Club.
From Tesco’s five-store trial to Ireland’s first in-store scanner, grocers are turning hidden ripeness into a visible buying signal. The real product is not the machine. It is certainty about dinner.
At SPAR KCR in Dublin, choosing an avocado now involves a machine. A shopper places the fruit beneath a scanner and receives a reading about its ripeness before paying. The device sits among a much larger package of retail technology at the newly opened flagship store—basket-recognising checkout, digital deli ordering and energy-saving refrigeration—but the avocado scanner is the feature that changes the most ordinary behaviour. Instead of squeezing, guessing and hoping, the customer asks the shelf for an answer.
SPAR’s owner, BWG Foods, describes the installation as Ireland’s first. It follows trials in Britain, Australia and continental Europe, and turns a familiar produce problem into a small but revealing retail experiment. Supermarkets have spent decades making packaged food more predictable through dates, labels, portioning and barcodes. Fresh produce still resists that logic. An avocado can look identical to its neighbour while being useless tonight and overripe tomorrow. The scanner attempts to sell not merely fruit, but timing.
The machine replaces the squeeze
Observed: Tesco began testing OneThird avocado scanners in five UK stores in September 2025. The machines use near-infrared sensing to assess the fruit without cutting it open. Tesco simplified the output into two kitchen intentions: one state suited to slicing, another ready for smashing. That language matters. It translates a technical measurement into the meal a shopper is trying to make.
The trial addressed two linked forms of damage. At home, a badly judged avocado can become waste because its usable window does not match the planned meal. In store, repeated squeezing can bruise fruit that another customer might otherwise buy. Tesco reported that it had sold nearly 15 million more avocados in the preceding year, so even a tiny improvement in selection can touch a large volume of delicate stock.
The concept has since travelled. In March 2026, an industry-backed trial at Trim’s Fresh in Merrylands, western Sydney, tested what was presented as Australia’s first in-store avocado ripeness scanner. The output there used several stages, from very firm to very ripe, rather than Tesco’s two culinary prompts. In February, trade publication FreshPlaza reported that OneThird’s updated scanner was present in more than 350 stores globally, from Chile to Finland. SPAR KCR’s September opening adds Ireland to that map.
These figures and rollout claims come partly from the companies and retailers involved, so they should not be confused with an independent verdict on waste reduction. What is independently visible is the repetition of the same intervention across different retail formats and countries. A produce scanner has moved from trade-show novelty to a device that ordinary shoppers can encounter beside the fruit.
An avocado is a scheduling problem
The avocado is almost designed to frustrate conventional grocery retail. Its commercially valuable quality is hidden beneath thick skin. Ripening continues after harvest. The interval between resistant flesh and brown softness can be short, while shoppers buy for different moments: breakfast now, dinner tomorrow, guacamole at the weekend. A single “ripe” label therefore conceals several jobs.
Retailers have already tried to organise that uncertainty with colour-coded stickers, “ripe and ready” packs, loose fruit at different stages and staff rotation. Those tools classify batches. A scanner promises to classify the individual object at the moment of choice. It changes the unit of trust from the crate or label to this avocado, for this meal, on this day.
That is the product autopsy. The machine is not principally selling advanced optics. OneThird’s broader business uses AI-assisted scanners and quality data across the fresh-produce supply chain, where growers, distributors and retailers need to estimate shelf life and route stock. On the shop floor, all that complexity is compressed into a binary or four-step answer. The consumer does not buy data; the consumer buys relief from a small gamble.
Freshness becomes an interface
Packaged-food retail is full of interfaces: nutrition panels, expiry dates, QR codes, preparation instructions and reviews inside shopping apps. Loose produce has remained unusually analogue. Shoppers lift, smell, tap and squeeze. Expertise is embodied and uneven. The avocado scanner inserts a diagnostic interface between hand and fruit.
That can broaden confidence. A frequent avocado buyer may regard the device as unnecessary, but a less experienced shopper can use it as permission to buy. The commercial opportunity is therefore not limited to reducing shrink. If uncertainty suppresses a purchase, reliable guidance can expand the category. OneThird and retailers have said early deployments improved sales and reduced waste, although publicly available independent, store-level results remain limited. The direction is plausible; the magnitude is not yet settled.
The more interesting change is what the interface can do to merchandising. A supermarket normally groups avocados by size, variety, origin or pack count. Ripeness data permits grouping by use occasion. “Tonight” and “later this week” can become more valuable distinctions than small and large. A digital shopping list could reserve fruit for the intended day. Online fulfilment could select maturity according to the delivery window rather than treating every acceptable avocado as interchangeable.
The promise has a weak point
A scanner can also create new friction. The produce aisle works because it is fast. If every shopper must queue at a device, repeat scans or interpret an ambiguous result, the solution can become slower than the squeeze it replaces. A machine that is poorly calibrated for variety, temperature or seasonal variation could transfer distrust from the fruit to the retailer. Maintenance matters too: a dark screen beside a full avocado display advertises a promise that the store cannot keep.
There is also a behavioural complication. Shoppers do not all want the same answer. One person’s perfect slicing avocado may be another person’s disappointing guacamole. The strongest systems therefore avoid claiming an absolute truth and instead connect firmness to intended use. Tesco’s “slice” versus “smash” framing is commercially sharper than a technical score because it acknowledges that ripeness is partly a job to be done.
Nor should a consumer-facing gadget distract from larger waste decisions upstream. Forecasting, cold-chain control, supplier specifications, stock rotation and discount timing influence far more fruit than a single shopper scan. OneThird itself sells tools for these supply-chain decisions. The in-store machine is most useful when it is the visible end of the same data system, not a sustainability prop isolated from buying and replenishment.
Possible next: the aisle learns the dinner date
Possible next step, based on the evidence: ripeness will become a selectable product attribute. The first scanners ask shoppers to bring fruit to a fixed station. A later version could place sensing above the display, sort replenishment by maturity or feed verified states into online inventory. A customer ordering on Monday could choose “ready tonight” or “ready Friday,” much as they choose pack size now. These are WBC projections, not announced features of the SPAR or Tesco systems.
The format will not transfer equally to every fruit. It is best suited to products with hidden quality, meaningful value, a narrow eating window and enough volume to justify the hardware. Mangoes, kiwifruit and pears fit parts of that profile. A cheap, visually legible apple may not. The commercial test is not whether a scanner can measure produce, but whether its answer changes enough purchases, complaints or discarded units to pay for itself.
There is a risk that stores turn the aisle into a row of devices, each solving a tiny anxiety. The better model is selective: deploy measurement where uncertainty is genuinely expensive, then make the result immediate and culinary. The SPAR scanner is compelling precisely because everyone understands the problem before the machine explains itself.