US retailers are spending big on AI. Chatbots. Automation. Smart tools that promise to fix customer problems before they start.
AI shows up everywhere now. Earnings calls. Investor decks. Store signs. App updates. Every brand seems to be racing to prove it is smarter, faster, and more modern than the last.
But customer experience is not getting better. In many cases, it is sliding.
Something feels off.
It is like buying a fancy car and finding the windshield wipers do not work the first time it rains. Everything looks impressive. Until you need it.
Customers still get stuck. They still call support. They still get confused. And most importantly, they do not feel like anyone is listening.
For shoppers, the gap between promise and reality is growing. The technology sounds advanced. The experience feels familiar. And not in a good way.
Customer Satisfaction Is Flat, Even with More Tech
New data from the American Customer Satisfaction Index shows that the overall score for online retailers barely budged in 2025. The top performers like Amazon stayed steady, while big names like Target and Walmart saw slight dips. Best Buy and Apple also saw small drops.
That is not a small trend. It is more like a warning sign.
Retailers have rolled out more AI tools than ever before. Yet customers are not reporting better experiences. They are reporting the same frustrations, just delivered through new channels.
Consumers are saying, in effect, “Your tech is not helping me.”
That matters. Most retailers are betting on AI to solve the very problems customers keep complaining about.
Problems Still Boil Down to What Happens After the Sale
AI shines at simple tasks. Tracking packages. Answering easy questions. Routing calls.
But the toughest moments for shoppers happen after the click.
Returns. Refunds. Order issues. Loyalty glitches. Refund delays. These are the sticky spots where humans still matter.
These moments are emotional. Money is involved. Time is involved. Trust is involved. A bot can give a status update. It cannot always give reassurance.
And US brands are feeling it.
For example, Amazon is in the spotlight for its return and refund policies. The company has agreed to settle a nationwide class action lawsuit over how it handled refunds for returned products.
Customers online are frustrated, too. Some people report long delays, chat support that does not work, and confusing return flags on their accounts. Many say they are more likely to shop somewhere else next time.
That is awkward for the world’s biggest retailer.
And it is a clear signal that AI alone is not enough.
Walmart And Target Are Not Immune Either
Walmart and Target do many things right. Strong stores. Strong logistics. Huge investments in digital and automation.
But even they struggle to translate AI into better experiences when something breaks.
Both saw satisfaction scores that did not improve in the latest ACSI data. That is despite heavy investment in technology and logistics.
That means customers still hit walls when something goes wrong.
And when something goes wrong after checkout, it often lands in the retail contact center, they are the real frontline of post-purchase experience.
That is where promises meet reality. And where frustration shows up first.
Best Buy’s Mixed Signals
Best Buy improved its customer satisfaction score in one recent retail ranking. That is worth noting.
But even here, social media tells a different story.
One shopper tweeted about being charged a hefty restocking fee after being denied the opportunity to test an expensive camera in the store. That is the sort of disconnect AI cannot fix.
Another frequent theme around Best Buy is that returns can be confusing or costly, depending on the product and timing.
The point is simple. Customers notice friction in real time, even when the checkout was smooth, and the app worked perfectly.
The Tricky World of Returns and Refunds
Returns in the US are huge. They are no longer a side quest. They are the game.
NRF said returns were projected to reach $890 billion in 2024, or 16.9% of annual sales. NRF also said 93% of retailers see fraud and exploitative behavior as a serious issue.
A separate returns report by Appriss Retail and Deloitte puts total US merchandise returns in 2024 at about $685 billion and estimates that $103 billion of those were fraudulent.
This is why refund timelines are getting tighter, rules are getting stricter, and customers are getting more anxious. Retailers are trying to protect margins. Customers are just trying to get their money back without feeling like they are on trial.
And real incidents show how messy it gets.
On Reddit, one Amazon shopper described dropping off a return, getting refunded, then later seeing the refund pulled back because the system said the return was never received. They said the status flipped from “return received” to “return not received,” and they had to fight support to prove it.
“I dropped off a return at the UPS Store and got refunded… then they later took it back because they said I never dropped it off!”
That is the kind of thing that makes customers feel crazy. They did the right steps. Then the system changed the story.
This is also where AI struggles. A bot can repeat a policy. It cannot calmly resolve a situation in which two systems disagree, and the customer feels blamed.
Real Experts See the Gap
Marty Bauer, an ecommerce expert at Omnisend, explained the returns problem in a way that feels very real.
He said return abuse and “digital shoplifting” are pushing retailers to slow refunds and add more checks. It creates extra manual work behind the scenes, and it is one reason shoppers are now seeing stricter return rules and longer waits.
This is the kind of gritty truth you do not hear in most glossy AI press releases.
The problem is not that retailers have policies. Policies are normal.
The problem is how the policy gets executed when the situation is messy, and money is involved.
Why AI Alone Will Not Fix CX
Here is the thing. AI can be fast. It can be precise. It can scale.
But customer experience is not a trivia contest. It is not just “answer the question.” It is “solve my problem and make me feel safe doing business with you again.”
AI is often deployed where it is easiest. Order status. FAQs. Store hours. Password resets. That helps. But it does not touch the moments that shape loyalty.
Those moments are post-purchase. Refund delays. Returns disputes. Wrong item shipped. The package says delivered, but it is not there. Loyalty points vanish. The gift card balance looks wrong. A promotion did not apply. Someone is upset. Money is involved.
This is why retailers keep seeing the same pattern. The chatbot works until it does not. Then the customer is pushed to a person who is already frustrated, already tired, and sometimes already convinced the brand is hiding the truth.
More data backs this up.
Gartner found that 85% of customer service leaders plan to explore or pilot customer-facing conversational GenAI in 2025. Gartner also said many leaders feel pressure from executives to implement it. That explains why AI is spreading so fast.
But adoption is not the same as improvement.
Zendesk’s 2025 CX Trends report adds another clue. It found 64% of consumers are more likely to trust AI when it feels more human, with friendliness and empathy. In plain English, people do not hate AI. They hate cold AI.
Now connect that to retail. If your bot cannot explain a refund delay in a way that feels clear and fair, the customer does not just feel uninformed; they feel misled.
Also, returns are rising and getting stricter at the same time. NRF said returns are now shaping loyalty, and the returns process is a real “touchpoint” for customer experience. That is a polite way of saying returns can make or break the relationship.
So, AI alone will not fix CX because most CX pain is not “information.” It is “exceptions.” It is “judgment.” It is “I need a real answer.”
A bot can tell you your refund is processing. A human can tell you why it is stuck, what will happen next, and what they are doing to fix it.
That difference is still the whole ballgame.
Post-Purchase Is the Real Test
ServeRetail’s view is simple. The sale is not the finish line. It is the starting gun.
More retailers are starting to see this clearly. They understand that the real experience begins after checkout, not before it.
That is why post-purchase support matters so much. When something goes wrong, that is when trust is tested.
If a customer calls about a problem and gets a clear answer on the first try, they are far more likely to stay loyal. If they get pushed through bots and loops, they are far more likely to shop somewhere else.
It is not science. It is common sense. And it is backed by real satisfaction data.
If your satisfaction scores are not rising while your tech spend is climbing, something in your execution needs fixing. A good place to start is how the retail contact center handles those tricky moments.
For more on how contact centers drive real outcomes, see ServeRetail’s blog on retail contact center operations and how brands manage loyalty and return better.
If you want one real, public example of how post-purchase issues turn into brand risk, look at Amazon again. The Reuters report about the refund settlement is not just legal news. It is a customer trust story. Refund handling is supposed to be the easy part. When it becomes a dispute, customers forget the great checkout. They remember the refund fight.
The Bottom Line
US retailers are not bad at technology. Many lead the world in digital shopping.
But none of that matters if customers still feel lost after the purchase.
AI alone will not get them there. Real human support will.
And that is the part retailers are still figuring out.
