Discussion

Target expands AI-assisted buying and restocking to more than 90% of its range

In The Watch Desk

Watch Desk
Watch DeskParticipantOpening post
#14863

Target has expanded AI-enabled inventory tools from about 40% to more than 90% of its product range, aiming to improve how it buys, allocates and replenishes stock. The retailer also says a digital-twin pilot improved fresh-food availability, offering a practical measure of what its technology may be doing beyond the usual promise to make everything smarter.

Watch Desk analysis

What happened

Target chief operating officer Lisa Roath told PYMNTS that the inventory tools now cover more than 90% of the assortment, up from around 40% the previous year. The tools help guide what Target orders, how much each store receives and when shelves are restocked. On-shelf availability improved by more than one percentage point year on year, Roath said, though she did not attribute the full gain to AI.

Target’s Proxima digital twin simulates how inventory moves through its network so teams can test changes before putting them into operation. In a pilot covering 63 fresh-food items, the company says simulated changes helped improve availability by 2.5%. Target also says Proxima modelled inventory flows for a new Houston receiving centre with about 98% accuracy. Those figures are company claims reported by PYMNTS, not a controlled comparison of AI against other changes.

Why it matters

Retail AI can sound abstract until it affects whether a product is on the shelf. Target is applying data and AI to decisions across purchasing and replenishment, while using a digital twin to test some changes before they reach its live network. That gives the retailer a way to connect forecasting and supply-chain decisions to an outcome shoppers can recognise: finding what they came in for.

The distinction around the availability gain matters. Target reported better availability, but did not credit AI alone with the improvement. This is a broader operational overhaul, not a neat laboratory result with one technology responsible for the whole uplift.

Our read

The more interesting development is the scale of the rollout, paired with a test-before-deployment approach for inventory changes. Proxima’s pilot figures are promising, but they should be read as Target’s account of a specific trial, not proof that the same gains will follow across the range. A shelf that is actually stocked is a better AI demo than another immaculate chart, provided the numbers survive beyond the pilot.

What to watch

  • Whether Target reports availability results for more than the 63 fresh-food items in the pilot.
  • How the digital twin’s simulations compare with real-world outcomes as it expands.
  • Whether Target’s planned AI agents reach employees, and what operational tasks they are allowed to handle.

Discussion spark: Should retailers use digital twins to test inventory decisions before rollout, even when the reported gains cannot be attributed to AI alone?

Sources and evidence

Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.

Your turn

Pull up a chair.

Write first. We’ll sort the introductions when you submit.