Amazon 4-star

» see amazon.com/4star as it lived, on the wayback machine

~ the eulogy ~

amazon 4-star sat in the part of the amazon empire that was us-only at the storefront layer. by 2018 i was 34, in ankara, buying almost nothing at amazon (amazon.com.tr did not exist yet, amazon.de shipped to turkey with import taxes that made it not worth it). a physical amazon store in a new york neighborhood was in the same category for me as a whole foods aisle: something i read about in retail-industry press.

the first 4-star store opened on 27 september 2018 at 72 spring street in soho manhattan, a block from the soho apple store. the concept was simple in a way most amazon experiments were not. only products rated four stars or higher on amazon.com made it onto the shelves. the mix shifted as the underlying review data shifted, so the same store could have a different assortment week to week. a lot of small kitchen gadgets, echo and fire products, board games, toys.

by early 2022 amazon had scaled the concept to about thirty stores across major us metros, alongside twenty-four amazon books locations and fourteen pop-up kiosks that had been running since 2014. sixty-eight physical spaces total, most of them small-footprint retail that used the same phone-checkout tech. on 2 march 2022 amazon announced it was closing all of them.

~ the dossier ~

~ the shelf-tag idea ~

the shelf tag at 4-star was the whole product. each item on display had a printed card with the amazon star rating, the review count, and a short pulled quote from a top review. shoppers were meant to feel that the store had already done the sorting for them, that this shelf was a real-world manifestation of the review database. it was the first mainstream retail concept to make an algorithmic curation model visible as physical merchandising.

the store had no cash register. checkout ran through the amazon app on your phone or through a staff handheld, the same model that amazon go would push much harder over the next few years. the price on the shelf was the prime price. non-prime shoppers paid a slightly higher amount at checkout, a small nudge that also showed up in amazon books stores of the same period.

~ the unraveling ~

4-star never claimed to be a profit center on its own. it operated as marketing spend and as a data source for the online business. the problem was that by early 2022 amazon had crossed an internal threshold on physical-retail patience. andy jassy had taken over as ceo in july 2021 and was cutting product lines that did not have a clear path to standalone economics. amazon glow got cut a few weeks earlier. amazon care and amazon halo would follow within the next eighteen months.

inside the 4-star chain itself, the head-count math worked against it in the same way it worked against amazon books. staffing thirty small stores across expensive us metros needed hourly workers and lease commitments that added up faster than the incremental prime signups the format generated. the curation gimmick did not offset the rent.

~ what it left ~

amazon kept the shelf-tag star-rating idea, which continued in whole foods and amazon fresh signage. the phone-checkout tech continued in amazon go. the 4-star brand itself ended cleanly, no franchise reopens, no other retailer picking up the format.

the wider legacy is what 4-star tried to prove: that a physical store curated entirely by online-review data could be a category of its own, distinct from a traditional department store or a specialty shop. the review-driven-store idea did not scale. amazon still uses the star-rating signal everywhere in its own product pages, but as a shelf-organizing principle for a lease-paying storefront, it turned out to be a novelty that ran out its interest inside four years.

@}-->--

              )
             (_)
            .-'-.
            |   |
            |   |
            |   |
            |   |
          __|   |__   .-.
       .-'  |   |  `-:   :
      :     `---'     :-'
       `-._       _.-'
           '""""""

« previous  ·  random  ·  next »