For someone hearing about Agentic Storefront for the first time, how do you describe it?

Agentic Storefront is a context-aware destination page. Traditionally, a landing page is where a customer arrives from an ad placement to be pushed toward a purchase. If not a landing page,  We're creating a one-to-many experience instead — you land on an AI assistant page that takes in the customer's context. If I was shopping for five men's tops pages before arriving at Agentic Storefront, that context comes with me, and every part of the page shifts based on what I'm looking for.

The big bet here is that context is true personalization — it matters more to end-users than zero-party data like your name, your city, or that you like watching sports. What I actually want as a consumer is to show up and say: I'm a man who needs long-sleeve top for an long weekend hiking trip in the mountains. Whether I type that in one go or answer three questions to get there, we're collapsing the path from browse to purchase. That's Agentic Storefront — a context-aware destination that moves customers through the purchase journey faster.

What's the thing about Agentic Storefront you're most personally excited about?

At the 30,000-foot view, I'm most excited because I think this is where shopping is going. Mobile traffic is in the 80% range for a lot of these websites, and yet sites are still designed for desktop and then squeezed down into a mobile experience. We say "mobile-first" all the time, but the websites themselves haven't really changed.

When you think about how people actually use their phones — click, click, engagement, talking into the device, hitting buttons — they don't want to type a lot. They want to get to the end result fast. We're in a time-based economy, and you have to create value as quickly as possible. Agentic Storefront gives you the front-end click-and-go experience, and on the back end, a model trained on our merchants' catalogs delivers exactly what the customer asked for. That feels like value to the end user — no filtering, no scrolling walls of product blocks. I tell you what I want, you give me the best match. That's the winning strategy as we move toward customer-forward experiences.

What does Agentic Storefront let a merchant do that they literally could not do before?

Three things. First, within Envive specifically, it gives merchants a destination. Historically we were only on PLPs and PDPs, which meant we were missing other high-intent surfaces — help pages, home pages, FAQs — places where users are looking for an answer, which is exactly what we deliver.

Second, it lets merchants drive traffic off-site. From an email, a QR code, an SMS — you can now send people to this page with context. Before, we could only work with traffic already on the website that landed on PLPs and PDPs. Now you can link to Agentic Storefront from emails, SMS, Google Ads site links, or a QR code on product packaging.

Third, it breaks the over-reliance on widgets. We have four widget placements today, but as a Shopify brand you now have the ability to link to a specific URL and control placement yourself. Put a hero block on your homepage that says "Talk to a personalized AI shopping guide." Update your global nav, footer, or product pages for organic search. There's a lot more optionality and flexibility for merchants.

When a shopper lands on the storefront page instead of a traditional PLP or PDP, what do you expect to change about their behavior — and what kinds of signals in the session data do you think will surprise merchants?

I expect the biggest shift to be how dramatically the journey gets truncated. On a PLP, you land on a wall of products, you've got filters, maybe you use the search bar, you scroll. People talk about dwell time or time on site, but I think about it as time from action to action — how long it takes someone to filter down to what they want, versus clicking three buttons or just asking directly. The gap should be significant.

In a time-based economy, time to click matters, and behind that, time to engage. How much can we compress the purchase journey? For some products, consideration used to take weeks, then a week, then a couple of days. With this experience, I'd expect it to come down to: tell us exactly what you want, boom, boom, boom. That's what I think will be most surprising to merchants — coming from 15 years in e-commerce, the consideration journey is collapsing fast, and the faster you can put the right product in front of a customer, the faster you realize the purchase.

From your seat in customer success, what do you think will separate the brands that get massive lift from Agentic Storefront from the ones seeing modest results?

I'd expect it to come down to two things. First, data cleanliness. AI in this space is a garbage-in, garbage-out problem. Everyone is trying to enrich their products and add more information, but there's no standard way to consolidate it. The brands that spend more time testing and iterating on their own data should perform best.

Second, volume. This is tough for any merchant, but the more people we have interacting with the storefront, the more our machine learning can identify trends and optimize. Once that flywheel kicks in — customers coming in, questions being asked, responses being optimized — the brands with higher volume should be able to turn the dials faster than those with smaller traffic.

The playbook frames Agentic Storefront as additive — keep your flows, just add an entry point. Is it really that simple, or is there a deeper shift the most successful brands eventually make?

The deeper shift is something we're still waiting to see, but for now, yes — it really is that simple. For a cart or browse abandonment email, you add a conditional targeting block or a standard block, maybe at the bottom or just below the main CTA. Cart abandonment emails convert maybe 20-30% of recipients, browse abandonment way less — that's a big chunk of high-intent traffic disappearing. If you can add even 1% absolute or 8% relative additional clicks by adding another CTA, that's meaningful incrementality pulled out of a high-intent bucket.

As we develop more technology — Envive MCP and other touchpoints — things will get more complex. But for Agentic Storefront right now, I'm telling everyone internally and externally: keep it simple. The long-term vision is that a lot of people will shop like this as the default front-end experience — show up, get asked a question or two, tell the site exactly what you want, refine, purchase. That makes a lot of sense on mobile.

All we're trying to do right now is earn the right to get more traffic. The only way to earn it is by making it work. So push traffic in through simple entry points — an email block, an SMS link, a homepage banner for a week — and as we identify which sources perform, we'll keep making updates that should make driving more traffic a no-brainer.

Which entry point are you most curious to watch — email, SMS, or one of the on-site placements — and what does your hypothesis say about where customers actually want an agent?

The one I'm most interested in is SMS. The transition from conversation to conversation just makes sense — that's how everyone we know already uses text. You're typing with your thumbs, hitting buttons, adding emojis. Going from a conversation into a link that drops you into another conversation feels more seamless than going from a text into a static web page where you have to scroll and click around.

The one I think has the highest variance is whether this experience just replaces search. A lot of non-technical people view AI as a better way to search, and in many ways that's true. So what's the purpose of search today? How many people show up to a site knowing the exact name of the product they want to buy? Depending on SKU, probably a low number. Replacing the search bar with a CTA into Agentic Storefront — how much does that actually change the user journey? That's the test I'm most curious to run over the next couple of weeks.

When you talk to a skeptical operator about Agentic Storefront, what's the objection that comes up most, and how do you respond?

The objection is: "Why would I drive traffic here?" And it's fair — traffic comes into these sites in all kinds of ways. This is where the "earn it" point comes back. You can turn the faucet on low or open the fire hydrant, but whatever traffic you send, you should be able to see results pretty quickly.

What I tell skeptics is: test one area. You don't have to roll it across your whole site. Test one email flow. It's a tool for marketers and e-comm operators to drive more customer conversations, and it's fully in your control. Let the results speak. If your customers like it, great. If they don't, you were right.

The main pushback is really just "who should we drive here, and how will we know it made a difference?" The answer is that you know where the traffic was going before — a landing page, a PDP, a PLP, the homepage — so you can measure the difference directly. The great thing about this space is that everything is measurable.

‍