Walmart: ChatGPT Checkout Converted 3x Worse Than Website (searchengineland.com) 28
Walmart found that purchases made directly inside ChatGPT converted at only one-third the rate of traditional website checkouts, leading it to abandon OpenAI's Instant Checkout in favor of routing users through its own platform. Search Engine Land reports: Starting in November, Walmart offered about 200,000 products through OpenAI's Instant Checkout. Users could complete purchases inside ChatGPT without visiting Walmart's site. Daniel Danker, Walmart's EVP of product and design, said those in-chat purchases converted at one-third the rate of click-out transactions. He called the experience "unsatisfying" and confirmed Walmart is moving away from it.
Instant Checkout was designed to let users complete purchases directly inside ChatGPT without visiting a retailer's website. However, earlier this month, OpenAI confirmed it was phasing out Instant Checkout in favor of app-based checkout handled by merchants. Walmart will embed its own chatbot, Sparky, inside ChatGPT. Users will log into Walmart, sync carts across platforms, and complete purchases within Walmart's system. A similar integration is coming to Google Gemini next month. In other Walmart-related news, the retailer announced plans to roll out "digital price tags" to all U.S. stores by the end of the year.
Instant Checkout was designed to let users complete purchases directly inside ChatGPT without visiting a retailer's website. However, earlier this month, OpenAI confirmed it was phasing out Instant Checkout in favor of app-based checkout handled by merchants. Walmart will embed its own chatbot, Sparky, inside ChatGPT. Users will log into Walmart, sync carts across platforms, and complete purchases within Walmart's system. A similar integration is coming to Google Gemini next month. In other Walmart-related news, the retailer announced plans to roll out "digital price tags" to all U.S. stores by the end of the year.
Who the fuck (Score:5, Insightful)
would shop through ChatGPT
Re: (Score:2)
I'm frankly surprised they had any "conversions" at all. I wonder what the actual rate was.
Re: (Score:2)
would shop through ChatGPT
Apparently, people who don't want to be ominously "converted".
Re: (Score:2)
I hate all of this sales pressure BS - that the instant anyone hears your name you have to try to "convert" them to sales - and I literally run a webstore.
It's all so deeply anti-consumer, and I want no part in it.
Re: (Score:2)
Someone got an immediate answer on a subreddit post who said they spent 30 minutes asking ShatGPT for an explanation and not getting one.
Folks who used to go to Google for everything instead of going to a specific shop or Wikipedia are now just using ShatGPT instead.
It's like the people who would type searches into the URL bar before they turned URL bars into garbage.
Re: (Score:2)
I'm reminded of the people who used to formulate search queries as a question.
Instead of bringing those individuals up to a level of understanding the tool, they've brought the tool down, and made things difficult for competent people.
Although to be fair, Google search was ruined way before ChatGPT.
Re: Who the fuck (Score:2)
Journalists use it for summaries (and sometikes don't double check sources even). Both Nate Silver and Tech Dirt blogs talk about their use and use paid versions.
A lot of programmers use it to do tedium too.
These two user classes pay less than it costs to provide them service, but do get significant value from it.
I use it to fix some of the tedium of using pandas, and also to generate pivot tables in Excel.
I think when they AI companies start focusing on their cost rather than improved benchmarks we'll see
Re: (Score:2)
People who use ChatGPT as the next generation search engine. So they might ask it for "I want to buy a new computer" and ChatGPT would now provide direct links to buy them from Walmart.
Or they may ask ChatGPT to find the difference between two similar products and it might suggest a third.
That was the idea, i think. But in reality people are probably doing their research after ChatGPT suggests a product rather than just blindly buying it leading to lower conversion rates.
Re: (Score:2)
The way things are going, everybody who can't figure out how to turn if off (and they will make it as impossible as they can to do so).
What business doesn't dream of eliminating the one obstacle between them and your money, your choice not to do business with them?
Re: (Score:2)
The business model makes more sense than you would think. Let's say you are chatting with ChatGPT about repairing something and after some discussion and ChatGPT loading quite a few repair tutorials it presents you the solution and the question "Look up offers for the repair parts?" This is basically the scheme how YouTubers who make hundreds of repair videos make money using referal Links, only that that tutorial was by ChatGPT and the offers fetched via some kind of API which shops can use to offer their
Re:Who the fuck (Score:4, Interesting)
Actually, ChatGPT is too good at shopping, and not so good at converting.
I recently used it to identify "interesting and unique hotels" in a given price range, and to explain why they were interesting and unique. It came up with a really great list, not based on how much each hotel paid for a listing, but based on actual interesting criteria.
But don't worry, OpenAI will do some intense analysis after this "failure" and will make sure that the next shopping bot they release, will be sure to "convert" more of the time.
WTF does this mean? (Score:4, Informative)
"in-chat purchases converted at one-third the rate of click-out transactions"
Theres a lot of jargon in that summary.
Re:WTF does this mean? (Score:5, Informative)
It means when people asked ChatGPT, "What kind of diapers should I buy for my chihuahua?" three times more people ended up buying the diapers when ChatGPT just gave them a link to the Walmart website versus ChatGPT describing the item and saying "I know who you are and where you live and have your credit card info. If you want to buy these diapers, just blink twice."
I know, it's a surprising result.
Re: (Score:2)
AI has the potential to do so many good things (Score:2)
Why do AI companies keep coming up with these stupid ideas?
Re: (Score:2)
At any given moment in their history, AI companies have tried to get AI to do what it can do in its current state. These ideas may or may not appear to be "stupid" but they are waypoints in the development of the technology.
They have gone from games (chess, etc.) to machine-learning applications (autonomous planetary probes and online recommendation engines) to deep-learning applications (jeopardy, go, voice-recognition, image-analysis) to our current collection of human-like technologies (chatbots, LLMs, a
I don't trust AI agents to make decisions (Score:2)
Not that bad... (Score:5, Funny)
Re: (Score:3)
They forgot that their religion was capped at 144,000 people.
Whole Lota Nonsense (Score:2)
Walmart is moving away... openAI is discontinuing.
Walmart will embed its own AI... inside ChatGPT? With a similar system inside Gemini next month?
Users will complete purchases within Walmart's system.
What a whole lot of nonsense.
In CmdrTaco's day... (Score:2, Flamebait)
I hate talking for others, but I'm pretty sure CT would have prioritized the "Digital price tags", which has all kinds of technical and legal implications like the ability to do surge pricing and change prices between the shopper getting something and checking it out at the checkouts, over "Fraud tech whose fraud has become obvious still awful."
But you guys do you, and keep hyping this shit.
And another LLM business model dead (Score:2)
I guess there really is nothing left besides "somewhat better search" now, and that one is nowhere near able to generate enough profits to keep LLMs running and updated.
Re: (Score:2)
LLMs have no business model.
That's why OpenAI is trillions in the hole, with no profitable tier of product in sight.
It's a cute toy that costs far too much to generate and maintain, and relies on basically stealing the data of the entire Internet to keep itself up-to-date and vaguely relevant, and the lawsuits on that have barely started yet.
And if an LLM was actually "AI"... it wouldn't need customers, as such. It could be left to wander off onto the Internet, given a credit card number and it would: set
Re: (Score:2)
That is a bit extreme and not really true. What is true is that LLMs have no business model that is sustainable, unless they become, say, 1000x cheaper to train and run. Which is not going to happen.
As to "AI" doing its own thing, that is overstating the case massively. Even AGI would not mean "can to anything" and would not imply "can set up and successfully run a business". At this time, we do not even know whether AGI is possible, and we certainly do not know what it would look like.