- The Retention Edge by MobiLoud
- Posts
- Your Customer Is Not an API
Your Customer Is Not an API
An API doesn't wear jeans.

If you spend any time online, you’ve seen this thrown around a lot lately.
It’s some version of: "Your customer is about to become an API." AI agents are going to do the shopping. Humans won't browse anymore. You need to optimize your entire business for machines.
You might have even seen it here. I’ve definitely been guilty of buying into the hype.
And I’ll be clear: the world, particularly the ecommerce world, is going through a monumental change right now with AI.
But I had a eureka moment this week reading one of these posts. A realization that a lot of brands are going to get themselves in a lot of trouble with this thinking.
The Take That's Everywhere
You can't scroll LinkedIn or open an ecommerce newsletter without seeing the “your customer is now an API” pitch.
The idea: before too long, AI agents will shop on behalf of your customers. They'll compare products, evaluate specs, check reviews, and complete purchases, all without a human ever visiting your site. Your product pages need to be machine-readable. Your data needs to be structured. Your checkout needs to speak API.
Google just launched Universal Commerce Protocol, an open standard for AI agents to browse, compare, and buy across retailers. Shopify rolled out Agentic Storefronts so merchants can sell directly inside ChatGPT, Perplexity, and Gemini. OpenAI has checkout built into ChatGPT.
The agentic commerce shift is happening.
But, contrary to what you’re being told, your customer is not about to become an API.
An API doesn’t use your product.
An API doesn't wear jeans. It doesn't use razors. It doesn't take supplements or drink soda or wear makeup.
We've Seen This Movie Before
This is not the first time retail evolved.
When ecommerce first took off, you could have had people making the argument that “you're not selling to a person anymore, you're selling to a computer.”
That didn’t happen.
Yes, people started buying through screens instead of in stores. The transaction was mediated by technology.
But it’s still the people who buy, the people who consume. The people who matter.
Agentic commerce? Largely the same thing.
The Risk: Brands That Forget Who They're Selling To
Agentic commerce is a big shift. It changes a bunch of things on the technical side of ecommerce.
Your PDPs need to be machine-readable. Your site needs to be plugged into the Matrix.
The majority of brands will get it. That it’s a new technical layer for ecommerce, not a 180° shift.
But I can already see that a lot are going to take it way too far.
“AI-native” brands, all in on “agentic optimization”, redesigning their entire brand for bots.
They're going to strip out the brand storytelling, the editorial, the personality, because an AI agent doesn't care about your About Us page. They're going to treat every product like a commodity with a spec sheet.
The real danger is when you start to see an uptick in traffic from AI agents. You see it as validation that your AI-first pivot is working.
So you go deeper. And doing so, you quickly strip away what makes your brand resonate with real people.
And the people you’re selling to are real.
How You Should Approach Agentic Commerce
You need to adapt to the changes that are happening. To ignore it completely would be a mistake. I genuinely believe that AI surfaces (ChatGPT, Google’s AI) are going to be a significant sales channel before long.
The risk isn’t in acknowledging agentic commerce as a thing. It’s in treating it as the only thing.
Here’s
Make your data clean and structured
Accurate product data. Good schema markup. Clean feeds. This isn't some revolutionary agentic commerce strategy. It's just good ecommerce hygiene.
If your product catalog is a mess, fix it. Not because AI agents demand it, but because it makes everything work better - search, marketplaces, social commerce, all of it.
Don't confuse the channel with the customer
An AI agent is a channel. Like Google. Like TikTok. Like email. Like your app.
Optimize for it the way you'd optimize for any channel: make sure your products show up correctly and your data is accurate. But don't reorganize your entire brand around it.
The brands HBR recently profiled as best positioned for agentic commerce? They're the ones that "speak to both humans and agents." Not one or the other.
Double down on the stuff AI can't replicate
Brand. Community. Experience. The emotional connection that makes someone choose you over a cheaper alternative that an AI agent surfaced.
73% of consumers can now spot AI-generated marketing and more than half actively disengage when they sense it.
Human-first marketing and content is going to become much more sought-after.
The irony of the "optimize everything for AI" crowd is that consumers are simultaneously developing antibodies to anything that feels automated.
Use AI for internal efficiency; work on being more human externally.
Protect your customer relationships
AI search turns brands into commodities.
The brand that fits the algorithm best wins. The right intersection of price/features/availability.
Anyone who’s built a business on an algorithm (Google SEO, Amazon) knows how perilous this is.
The brands who go all-in on agents at the expense of humans will have no customer relationships to fall back on. They’re at the mercy of machines.
Business is about relationships. Make sure you have contact with your audience, through email, your website, communities, your app. Make sure you know your customers, what’s important to them, how they think.
You can make the backend of your business hyper-optimized for bots. But the front-end of your business should be the opposite.
Hyper-personal. Real.
The Bottom Line
The infrastructure around agentic commerce is real. Google, Shopify, OpenAI, huge companies are all building it. You need to make sure your products are machine-readable and your checkout can handle agent-initiated transactions.
But if someone tells you "your customer is an API," push back.
Your customer is a person. They have a medicine cabinet, a closet, a kitchen. They have preferences that aren't in your product data. Their loyalty can't be earned through structured markup.
Build a brand that humans love. Make sure the machines can find it.
Not the other way around.
In Partnership with Arka

Are You Overpaying for Packaging?
Your packaging could be costing you more than it should, and we can prove it. Arka is offering a free packaging audit to help brands uncover hidden costs, compare prices, and explore smarter, more sustainable options.
We’ll review your current boxes, mailers, or inserts and show you where you can save without compromising on quality or brand experience.
📩 Email [email protected] to book your free audit today. Limited slots only.
Build a Future-Proof Brand Asset with MobiLoud

We've been talking about owned channels and protecting your direct customer relationship.
That's exactly what a mobile app does. It puts your brand on your customer's home screen, not on an AI agent's comparison table. When they want to reorder, they tap your icon. When you have something new, you send a push notification. No algorithm decides whether they see it. No agent shops around for a cheaper alternative first.
The whole point is that this is your channel. Your customer chose to install it. They come back because they want to, not because an AI sent them.
MobiLoud builds native apps on top of your existing site. No rebuilding, no duplicate work. Your full experience, on their phone, live in weeks (not months).
Get a free app strategy call or DM me on LinkedIn to talk about it.
Quick Hits
Is Reddit about to become a shopping platform too? Reddit is currently testing AI-powered product carousels in search results, surfacing products directly from community recommendations with pricing and buy links. Reddit has quietly become one of the most trusted product research platforms on the internet. Turning that trust into a transactional layer is a natural (and potentially powerful) move. Worth keeping an eye on.
Amazon (35.7% share) and Shopify (14%, up from 12% last year) now account for roughly half of all U.S. e-commerce. One is a centralized marketplace, the other is decentralized infrastructure powering millions of independent storefronts. Everyone else, including Walmart, eBay, and TikTok Shop, is fighting over the other 50%. If you're a DTC brand, your entire commerce world increasingly runs through these two rails.
Amazon hit $716.9 billion in 2025 revenue, edging past Walmart's $713.2 billion. The symbolic shift matters less than what drove it: AWS, advertising ($68B alone), and third-party seller services now make Amazon a tech and services company that also sells stuff. Walmart's still winning grocery (60% of its sales), but the question is whether it can make the same leap into higher-margin businesses.
Andy Jassy's argument is simple: horizontal AI agents like ChatGPT can aggregate selection, but they can't match a retailer on pricing, delivery, trust, and shopping history. Amazon's betting its own AI agent Rufus (300M users in 2025, 60% more likely to convert) beats any third-party middleman. Whether he's right or not, the subtext is clear: the battle for who owns the AI shopping layer is on.
Meta's $2B+ Manus AI acquisition is already showing up in Ads Manager, offering automated data analysis and report generation for advertisers. Zuckerberg says agents are valuable because of "the unique context they can see," including user history and interests. For DTC brands running Meta ads, this could mean less time pulling reports and more AI-assisted optimization. Early days, but worth watching.
The funnel is collapsing. This holiday season, a larger share of purchases happened with fewer touchpoints, with conversions at or near first exposure. Digital wallets, seamless checkout, and social platforms where discovery and purchase happen in the same environment are compressing what used to take weeks into moments. The implication: if you're not winning at first impression, you might not get a second one.
On the Pod
This week we hosted Brian Massey on the Retention Edge podcast.
Brian is the Founder of Conversion Sciences - a company that uses science to help you get more conversions.
He had a lot to say on A/B testing, common UX mistakes that kill conversions, analytics blind spots, and a ton more useful CRO insights.
Recommend giving this a watch/listen. Check it out below:
That’s all for now.
I’ll be back in touch next week, with more on how successful brands are doing CX and retention right.
If there’s any topic you’d like to see us dive into, for either the newsletter or the podcast, just shoot me a message here.
Until next time,
Pietro and The Retention Edge Team
PS: want to boost retention, revenue and profitability? If so, launching your own app could be the best move you make this year.
See how: go to our website to get a preview of your app for free, or shoot me a DM on LinkedIn to talk about it.