A shopping assistant announced from a position of strength
Best Buy did not launch Ask Blue quietly. The company unveiled its new conversational AI shopping assistant on its Q2 2027 earnings call, the same call where incoming CEO Jason Bonfig reported same-store sales growth of 4.1% year over year and quarterly revenue of $9.8 billion, up 3.6% annually. That timing matters. Best Buy is not reaching for AI as a turnaround story or a hedge against declining traffic. It is folding a shopping assistant into a results narrative that was already working, which gives the rollout more room to be judged on its own merits rather than as a Hail Mary.
Bonfig, currently chief customer, product and fulfillment officer, will formally take the CEO seat in fall 2026. His framing of Ask Blue on the call was operational rather than promotional: "Customers need solutions quickly when an essential appliance fails, and our ability to offer competitive pricing, strong product availability, and fast delivery and installation are imperative to compete effectively." That is a supply chain and service statement wearing an AI headline, and it tells us where Best Buy actually expects the assistant to earn its keep: high-stakes, time-pressured purchases, not casual browsing.
What Ask Blue actually does
Ask Blue leans on product details, customer reviews, real-time item availability, and pricing to help shoppers make comparison and compatibility decisions, the kind of research that used to mean five open browser tabs or a wait for a blue-shirted associate. It can tell a customer whether a specific dryer fits an existing venting setup, or whether two components in a home theater system will actually work together, using data Best Buy already owns rather than a general web crawl.
The more consequential design decision is what Ask Blue does when it hits the edge of its usefulness. Rather than optimizing to keep the conversation inside the chat interface, the assistant is built to direct customers toward self-service resources or a live support specialist when the situation calls for it. That is a deliberate rejection of the containment metrics that dominate a lot of enterprise chatbot design, where success gets measured by how many contacts never reach a human.
Why the off-ramp is the real story
Most retail AI assistants launched over the past two years have been judged, publicly or internally, on deflection rate: the share of shopper questions resolved without a human touching them. That metric is easy to game and often wrong for the categories that matter most. An appliance that just died, a home theater system that needs to work the first time, a compatibility question with a wrong answer that costs a customer a return trip: these are exactly the moments where a confidently wrong AI answer does more damage than a slow human one.
Best Buy's own positioning backs this reading. Bonfig tied Ask Blue directly to appliance replacement urgency, a category where speed and accuracy both matter and where trust, once broken, does not come back easily. Building the escalation path as a first-class feature, not an apology screen after the bot fails, is a signal that Best Buy's product team modeled the cost of a bad AI answer against the cost of a slower resolution and picked accuracy over automation share.
Digital assistance is shipping paired with staffed physical space
Ask Blue is arriving alongside the expansion of Best Buy's Meta Labs concept, staffed, roughly 900-square-foot experience zones now running in 50 stores. Reading the two moves together, Best Buy is treating digital assistance and staffed physical experience as one coordinated strategy. The company is investing in both at once, on the same earnings call, which tells us AI is being positioned as a complement to store labor rather than a quiet replacement for it.
That pairing matters for any executive building a business case for a shopping assistant primarily on labor savings. Best Buy's public framing leans on speed, availability, and service quality: appliance failures, installation timelines, delivery windows, and getting a complex purchase right the first time. Headcount reduction is absent from the pitch entirely. If Best Buy eventually reports labor efficiencies tied to Ask Blue, that will read as a secondary benefit disclosed after the fact, not the premise the company sold to its board or its investors on this call.
Where this sits in a crowded field
Best Buy is a late but deliberate entrant into a category already occupied by Walmart's Sparky, Amazon's Rufus, and Home Depot's Magic Apron, each iterating fast and each reporting meaningfully higher spend per order among users who engage with them. Ask Blue does not need to out-feature those tools at launch. It needs to prove that a narrower, escalation-first design converts better within Best Buy's specific categories, electronics and major appliances, where a wrong answer is expensive to fix and the right human is frequently standing thirty feet away on the sales floor.
That is a defensible niche strategy rather than a me-too launch, and it is testable in a way vaguer AI announcements rarely are. Same-store sales, attach rates on services like delivery and installation, and repeat purchase behavior among shoppers who use Ask Blue will eventually tell Best Buy, and every competitor watching this launch, whether an assistant engineered to hand off gracefully outperforms one engineered to hold the conversation as long as possible.
The build-vs-buy question this poses for your roadmap
If you run technology at a retailer or a commerce-adjacent SaaS company, Ask Blue is a useful benchmark for a design argument you are probably already having internally: should your assistant be scored on how much human contact it eliminates, or on how well it routes the moments that genuinely need a person? Best Buy just put a public, earnings-call-endorsed answer on the record, and it chose the second framing.
The practical takeaway is to build the escalation path before you build the containment dashboard. Instrument for the moments your assistant should not handle alone, price-sensitive compatibility calls, post-failure urgency, anything with a return-shipping cost attached, and route those deliberately. Best Buy is betting that customers will trust an assistant more, not less, when it knows its own limits. That is a cheaper hypothesis to test than most AI initiatives on your roadmap, and it is one you can validate with data you already collect.
