Why is Almost Everyone Still Struggling to Make Their AI Work?
Plus, a cool new tool that can actually click buttons and type things to help your customers.
📅 December 9, 2025 | ⏱️ 4-min read
Good Morning!
Everyone is talking about how AI will change everything, but let’s be honest: is it really making a big difference for your company’s bottom line? Today, we’re looking at why most companies feel like their AI projects are stuck in a testing phase that never ends. We’re going to look at what the biggest and smartest companies are doing to make AI pay off in billions, and what happens when the simple AI systems break down.
Here’s what you need to know today.
🛍️ Ashley Furniture Builds a Shopper AI That Handles the Entire Checkout
The furniture store Ashley recently teamed up with Perplexity and PayPal. North America’s largest furniture store brand built an AI shopper that can help you find products, add them to your cart, and finish the entire payment, all while you are just chatting with it.
Shopping used to mean clicking a lot of different pages. Now, it’s just one easy conversation. This shows that the smart companies are connecting their entire store—all the inventory and payment systems—to the AI, not just building a simple Q&A bot. They are focused on making money, not just giving out facts.
⏳ Customer Patience is Gone: More Than Half Won’t Wait Longer Than 10 Minutes for Help
A new Forethought report says people are getting super impatient. 57% of customers won’t wait longer than 10 minutes for help. The biggest problem is still automated systems that won’t let you talk to a real person when you really need one.
When companies use weak AI that doesn’t work well, it just makes people angry and they might stop buying from you forever. It’s not the AI they hate; it’s the bad setup. Before you launch an automated system, you absolutely must have an easy, clear button for customers to get to a human agent.
💰 Salesforce Reports AI Tools Drove a Huge $67 Billion in Cyber Week Sales
Salesforce announced that AI tools helped drive a huge $67 billion in sales for businesses during the busy Cyber Week shopping holiday, showing that these agents are making real money.
This proves AI isn’t just for saving money; it’s a big revenue driver. These AI programs are doing complicated jobs, like dealing with returns, discounts, and subscriptions all at the same time. The goal isn’t to have the cheapest bot, but the one that can handle the most complex and valuable tasks perfectly.
🧠 People Are Starting to Trust AI With Big Life Decisions Like Health and Money
The ‘What’: A Smart Communications study found that for the very first time, almost half of all customers are okay with AI giving them important life advice, like what insurance to get, how to manage money, or even health recommendations.
People are starting to trust AI with big, serious stuff. This is a huge shift. It means if your basic customer service bot messes up, customers see it as a major trust failure, not just a technical mistake. Your company needs to ask: “Is our AI good enough and safe enough to handle the serious questions customers are now willing to ask?”
🛠️ Tool of the Day: Fullview
Fullview is an AI tool that can actually “see” the customer’s screen and product interface. It gives step-by-step visual help and can even take direct control—clicking and typing—to fix problems for them within the product.
The ‘So What’: Companies often think that just giving a good answer is enough. This tool says, “No, you need to solve the problem.” The main lesson is that a truly useful AI needs to do the physical steps a human agent would, like clicking buttons and filling out forms. If your AI can’t be trusted to do the work, it’s not really automating anything important.
📊 DCX AI Data Stat
Minimal Financial Impact of AI, Despite High Adoption
Even though 88% of companies say they use AI somewhere, only 39% say it has any clear impact on their profits (EBIT). And for most of those, AI adds less than 5% to total EBIT.
Takeaway: There’s a big gap between the hype and real results. Most companies are still stuck testing AI in pilots, not using it at full scale. Until they make the bigger changes needed to run AI across the business, it won’t move the bottom line.
Source: McKinsey Global Survey, “The state of AI in 2025: Agents, innovation, and transformation.”
Your 1-Minute Action Plan
Ask your Head of Data: “If our smartest AI agent messes up a very important task, is the mistake a technical problem, or is it because we have a missing business rule? And how quickly can we find and fix that rule?”
The Signal
The main takeaway is that almost everyone is struggling to roll out AI because they focus only on saving money, not on solving real human problems like making people wait, earning their trust, and setting clear rules. The winners use AI to handle tough jobs perfectly.
That’s the rundown for today.
See you tomorrow!
👥 Share This Issue
If this issue sharpened your thinking about AI in CX, share it with a colleague in customer service, digital operations, or transformation. Alignment builds advantage.