
Your organic traffic is drifting down. Your paid social costs keep climbing. And you have no reliable way to know whether ChatGPT, Claude, or Google’s AI Overviews are recommending your brand, or quietly handing the sale to a competitor.
That’s the reality right now for Shopify brands scaling from roughly $2M toward $10M and beyond, and the data says this isn’t a passing blip. Shopify reported that AI-driven traffic to merchant stores grew 8x year over year in Q1 2026, orders originating from AI-powered search rose nearly 13x, and new buyer orders from AI search are arriving at close to twice the rate of traditional organic search.
Rob Langenback has spent more than 12 years inside 802, where search is the only thing the agency does. Last year alone, his team drove more than $1 billion in online revenue for clients including Orvis, Pure Hockey, The Company Store, and Garnet Hill. Across that client base today, LLM traffic sits at roughly 1% to 5% of sessions and revenue, but for brands actually tracking it, that number is up 1,200% to 1,500% year over year. Some clients booked a couple hundred thousand dollars in a single quarter from tools that barely registered 18 months ago.
In this conversation, Rob breaks down the AI visibility playbook his team runs for mid-market and enterprise brands: what belongs on a product page so the tools can actually recommend it, why collection pages are the most underrated SEO asset in ecommerce, where blog content still earns its keep, and a free process any DIY merchant can run this week to find out whether they’re showing up at all. Whether you’re doing $50K months and figuring this out yourself, or $2M months with a team ready to execute, this is the lay of the land.
Let’s dive in. 👇
✅ Why traffic can fall while revenue climbs: Rob breaks down the difference between healthy zero-click traffic loss and actually losing ground, and why LLM sessions convert closer to branded search than to cold, non-brand traffic.
✅ The one product page mistake almost every brand makes: there’s a reason the tools can’t recommend your bestseller, and it usually comes down to a naming convention your marketing team fell in love with years ago.
✅ What “good product data” actually means in 2026: use-case copy, explicit product types, dimensions, materials, care, fit, and the schema field most Shopify stores skip entirely.
✅ Why collection pages are the most underrated SEO asset in ecommerce: the shopper landing there hasn’t decided what they want yet, which is exactly why that page is your highest-leverage conversion surface.
✅ A free DIY AI visibility audit you can run in an afternoon: no tools, no subscriptions. Rob walks through how to use the AI tools themselves to reverse-engineer what your best customers are actually asking before they buy.
✅ The honest read on Amazon: why Rob calls it a necessary evil, what happens to your buy box once you start winning, and how to weigh marketplace revenue against the audience you actually own.
Plytix gives you a single place to create and manage all your product content on Shopify.
Import your entire Shopify catalog in one click, and from there, you can bulk edit thousands of products, variants, metafields, and images with AI that actually knows your products.
Create new listings, optimize for SEO and AI search, translate and localize content, generate and enhance images, and sync everything across your Shopify stores and markets without breaking a thing.
Your whole team can work from the same place with unlimited users, commenting, and version history, so everyone stays on the same page. With Plytix, your listings stay accurate, your products get discovered, and your product catalog finally feels under control.
Search stopped looking like search. That’s the shift Rob Langenback has been watching from inside 802, the search and performance agency he’s helped run for over 12 years, and it’s why this conversation matters more than another SEO checklist.
The mechanics have changed in a specific way. Informational, top-of-funnel queries have migrated into ChatGPT, Claude, Perplexity, and Google’s AI Overviews. What lands on your site now skews heavily bottom of funnel: people who’ve already done their research and are ready to buy. Rob’s clients are seeing stronger conversion as a result, even as raw session counts wobble. The catch is visibility. When someone runs four refining searches inside an LLM and never touches a website, you’re blind to all of it. You only see the click. As Rob puts it, the biggest challenge his team has faced this year isn’t ranking, it’s knowing where they’re being cited and recommended in the first place.
So how does a brand get onto that shortlist? Rob’s answer is unglamorous and specific: give the tools better information. That means product pages built to answer the questions a real buyer would ask before purchasing. What is this product for? What situations is it best in? Why choose this over the alternative? He shares a swimwear example, where the winning copy isn’t “premium long sleeve rash guard” but language explicitly stating this is the right choice for someone with sun-sensitive skin planning several hours on the beach, with the UPF 50 rating right there on the page. He also shares a client who named products things like “the Emma,” a lovely name that tells an AI model absolutely nothing. If you never call a shirt a shirt, you can’t be surprised when the tools don’t know it’s one. And he makes the case for schema most brands ignore, including variant markup for sizes, colors, and patterns, which feeds both LLMs and the product grids now dominating Google results.
Then there’s the collection page. Rob argues it’s the most underrated SEO asset in ecommerce, and Steve backs him up with case studies of his own. The logic is simple: someone hitting a product page has usually already decided, while someone hitting a collection page is still choosing, which makes comparison blocks, buying guidance, and granular subcollections built around specific needs far more valuable than another filterable grid of thumbnails. The goal, in Rob’s framing, is to keep shoppers from going back to Google or back to an LLM to finish their research.
For the DIY crowd, Rob offers something genuinely free. Ask Claude or ChatGPT how a customer would find you and why they’d choose you. Take those answers, turn them into ten prompts a real buyer would type, run them, and log what comes back. You won’t get the same result twice, and that’s fine. You’re looking for directional patterns: are you showing up at all, and who’s showing up instead? Then go read those competitors’ product and collection pages. That’s your gap analysis, and it costs fifteen minutes and zero dollars.
This isn’t a repackaged SEO pitch with new acronyms bolted on. It’s a working operator telling you exactly what’s changing, what still works, and where to start.
👉 Falling traffic isn’t automatically a problem. Falling revenue is. Informational searches are moving into AI tools, meaning fewer top-of-funnel sessions but a higher concentration of ready-to-buy visitors. Rob’s clients see LLM traffic convert well above cold, non-brand search, sitting closer to branded traffic that typically converts 2x to 5x higher. Judge the channel by revenue and conversion rate, not session count.
👉 If you never call a shirt a shirt, the tools won’t call it one either. Distinctive product names are a branding asset and an AI discoverability liability at the same time. Explicit product type belongs in the title, the description, and the schema, right alongside the creative name. This is a one-afternoon fix on most catalogs, and it’s the highest-return work on this list.
👉 Write product copy for the buyer’s situation, not the product’s features. The tools are matching recommendations to a use case, not a keyword. “Best long sleeve swimsuit for sun-sensitive skin on a long beach day” isn’t a keyword you’d have targeted a few years ago, but it’s exactly how people talk to an LLM now. Name the person, the situation, and the conditions your product is right for.
👉 Treat your collection pages as buying guides, not product holding pens. The shopper arriving there hasn’t chosen yet, which makes that page your best chance to influence the decision. Comparison blocks, price-point guidance, what to look for, and subcollections organized around specific needs all keep the shopper from bouncing back to Google to finish their research somewhere else.
👉 Structured data is how you become machine-readable. Skipping it is a choice. Product schema, availability, pricing, ratings, and the frequently skipped variant markup for sizes, colors, and patterns give both LLMs and Google’s product grids something clean to extract. More complete data means more surfaces you can appear on.
👉 You can audit your own AI visibility this week for free. Write ten prompts a real buyer would use, run them across ChatGPT, Claude, and Perplexity, and log the results. Results shift day to day, so look for patterns rather than a single snapshot. If a competitor keeps surfacing and you don’t, go study their product and collection pages. That’s your roadmap.
👉 Amazon is rented ground, and the rent goes up once you start winning. Rob has watched clients build real revenue on the platform, then lose the buy box to undercutters, get knocked off by scrapers, or get told the marketplace would rather buy wholesale and control the listing themselves. Play there if the demand justifies it, but the customer relationship, the margin, and the lifetime value live on your own store.
Robert Langenback
President and COO, Eight Oh Two
Rob Langenback has spent more than 12 years at Eight Oh Two and close to two decades in search overall. He was promoted from Chief Operating Officer to President in 2024, taking the helm of a Google Premier Partner agency that does one thing: search. That includes paid search, SEO, content marketing, generative engine optimization, paid social, and conversion rate optimization, all pointed at the same outcome, getting client products in front of buyers wherever the search happens.
The results speak for themselves. In 2023, Eight Oh Two drove more than $1 billion in online revenue for its clients, a roster that includes Orvis, Pure Hockey, The Company Store, and Garnet Hill alongside brands in home goods, pet, and professional services. And yes, the name is a Vermont reference: 802 is the state’s area code, and the agency was founded there.
What makes Rob a useful voice on this topic is that he’s not selling certainty. He’s candid that prompt tracking is still emergent, that the keyword volume data the industry leaned on for twenty years simply doesn’t exist for LLM queries, and that his team is actively testing tools to find what works. He’s equally candid about fit, telling brands directly when they’re not at the right stage to work with an agency. That transparency, paired with real pattern recognition across a wide range of client accounts, is exactly what this moment calls for.
Featured in This Episode:
Over 9 seasons, I’ve been incredibly fortunate to chat with some of the brightest founders building amazing Shopify brands, as well as the partners shaping the app and marketing ecosystem. Every conversation has taught me something new, and I’m grateful for the chance to learn alongside you.
What matters most is that this podcast helps you solve real challenges and discover new ways to grow. Your support, feedback, and stories have made this journey truly special. Thanks for tuning in, sharing your wins and losses, and being part of the eCommerce Fastlane community.
Stay Connected: Leave an Honest Rating/Review on Apple Podcasts or Spotify. Follow & Subscribe on YouTube for new episodes.
Steve Hutt:
Welcome back to eCommerce Fastlane. I’m your host, Steve Hutt.
Today’s conversation is an important one. It’s all about getting found, because search doesn’t really look like search anymore. The habits around discoverability and the research phase that goes into buying a product are changing. If you’ve noticed your organic traffic drifting down a bit, or your ad costs climbing through paid social and elsewhere, you’re not imagining it. That’s happening to everyone.
And here’s the bigger issue: you don’t really have a reliable way of knowing whether ChatGPT, Claude, or Google’s AI Overviews are actually recommending your brand, or recommending a competitor’s brand instead. I think that’s incredibly important, and it’s one of the reasons I have my guest on today, Rob Langenback.
He’s the president of a company called 802. I’ll ask him in a minute where that name comes from. I think it’s a Vermont area code, but I’m not totally sure yet, so I’m going to find out.
Rob Langenback:
802.
Steve Hutt:
There you go. I’ll put the link in the show notes. Search really is the only thing his agency does, and from what I understand, last year alone his team drove over $1 billion in online revenue for their clients. They have some incredible brands, mid-market to enterprise, and we’re going to dig into a lot of it today.
It’s so interesting. So, hi Rob, welcome to the show.
Rob Langenback:
Hey Steve, it’s great to chat. Love the podcast.
Steve Hutt:
Thank you so much. So, from what I can see, did I creep on your LinkedIn profile a bit? You spent about 12 years inside 802, and like I said, search is the only thing your agency does. What’s actually changed in your clients’ accounts? Let’s talk about the last year, since AI has really been dominating the narrative lately. You clearly know the SEO side of the world and being organically found, but what’s changed over the last 12 months from your perspective?
Rob Langenback:
Yeah, it’s a great question. I think the biggest shift we’ve been seeing is in the types of traffic coming to our clients’ websites. We used to think in terms of top-of-funnel, informational-driven searches versus bottom-of-funnel searches, where people are ready to buy. We’ve seen much more traffic coming at that bottom of funnel now. A lot of that informational-type search is happening more and more in tools like Claude and ChatGPT, or even in Google, where we see a lot of that traffic go into AI Overviews.
We’ve talked about the rise of zero-click for ten years in SEO, but in the last 12 months, we’ve seen the most dramatic changes. We see a lot more AI Overviews for everything you search, and a lot more product grids showing up in the search results as well. There have been some strategic changes in how we optimize sites and what we’re targeting.
I’ve been doing this for almost 20 years now, and the last 12 to 18 months have been the most exciting, and a little scary at times, but there’s been a lot of change happening in search. It’s changed how we do things, but it’s also presenting opportunities, especially for smaller Shopify brands trying to find their way and get ahead in markets that have primarily been dominated by big brands.
Steve Hutt:
Right. What’s interesting is the ongoing narrative that traffic is going down for a lot of sites. We talk about these zero-click queries, these blue links in Google, and some people aren’t even going to Google anymore, but they’re still doing their research. If they’re getting their answers through chat systems or AI Overviews, I don’t know if that’s necessarily always bad news. There’s a line between healthy traffic loss and actually losing ground. What’s your thought process on the fact that traffic in some cases is going down, but brands can still have revenue going up if they’re thinking about things from a different perspective?
Rob Langenback:
Yeah, absolutely. Probably the biggest challenge we’ve faced this year has been visibility tracking. We lose a lot of that when people go to Claude, ChatGPT, or another LLM to do their research. They may refine their search from there, making multiple searches without ever visiting a website, and we’re blind to that. The only time we can see anything is when they make a click. So if your brand is recommended in ChatGPT and they click through to your website, at least we can track that session, see a referral, and hopefully a conversion.
But visibility has definitely been the big struggle, trying to find out where we’re being cited and recommended. Across our client set, we do see stronger conversions because we’re seeing more of that purchase-driven search driving to our site. We’re losing visibility into where they’re finding us from the start, where our brand is being recommended, where we’re being brought into that conversation. We’re still working through it, but that’s where the biggest struggle has been.
Steve Hutt:
Yeah, it’s interesting because I think Harley Finkelstein from Shopify made some comments recently, sharing publicly on one of their earnings calls, that AI traffic is still under 3% across all the accounts he was able to talk about. But he also said that AI-driven discoverability of a brand, the conversion rate of at least that touchpoint, is significantly higher. I don’t know the exact number, I’m going to look it up in a minute. But it’s interesting that here we are in July of 2026, and even though that’s only 3% of attributed revenue or conversion, the conversion rate of that attribution is massive compared to traditional search.
Rob Langenback:
Yeah, that’s exactly what we’ve been seeing too. The overall number I tell everyone is that most clients are seeing anywhere from 1% to 5% of their traffic or revenue coming from the different LLMs. But we do see very strong conversions. For anybody who breaks out brand and non-brand traffic, we generally see two, three, four, five times higher conversion rates for branded search traffic.
We’re kind of in the middle between non-brand and brand for LLM traffic. It’s not quite as strong as brand, but it converts very well, which speaks to the tools. People go to them because they’re trying to find a product for whatever they’re looking for, running shoes, for example. They say, “I need running shoes for a beginner runner,” or “for my first 5K,” or “for an ACL repair.” They’re creating these really specialized searches, refining based on the answers, and getting recommendations from the tools.
Ultimately, if your brand or product is being recommended, that’s why we see great conversion, because the tools are doing a really good job of making good product recommendations, versus historically, traditional search has been hit or miss. We see a lot of shopping ads, and you’d have to go read the reviews yourself to see if a product is good for your use case. The tools are doing such a good job of making those shortlist recommendations that when people do come to your site, we see really strong conversion.
Steve Hutt:
Yeah, I just looked up Harley’s comment from the Q1 earnings call. He said AI-driven traffic to Shopify stores grew 8x compared to the same quarter last year, and orders originating from AI-powered search rose nearly 13x over the same period. He added that new buyer orders coming in from AI search are happening at close to twice the rate of traditional organic search. It’s quite interesting, and this is a true reality from Shopify.
Rob Langenback:
Yeah, and we’ve seen a lot of growth too. Even though the overall percent of traffic is small, when we compare it to 12 months ago, for clients who are tracking it, we’re seeing it up 1,200% to 1,500% year over year. What we’ve been tracking mostly is quarter over quarter. Looking at each quarter, we can see that number keep getting bigger for clients where we track it. If you’re not already tracking this, I’d urge everyone to make sure your analytics are tracking all your sessions and revenue from these tools, and look at that conversion, because a year ago it was very small for some clients.
We have some clients where, last quarter, they did a couple hundred thousand dollars in revenue coming from these tools, which is crazy compared to where it was just 12 or 18 months ago.
Steve Hutt:
It’s amazing. So let’s talk about a brand that wants to come up on a shortlist. Let’s say they’re using ChatGPT or Claude and doing this research. How does a brand come up on this list during their interaction with these tools? I know this is the billion-dollar question, but how does a Shopify business come up? The challenge on one side is that traditional SEO, a high-profile backlink profile, and an aged domain are the traditional SEO things that go along with how Google is ranking and building AI Overviews. But other than the Google side of it, when you think about Perplexity, Claude, or ChatGPT, how do you become part of the conversation with your brand in these systems?
Rob Langenback:
Yeah. I think the big thing is making sure you have really good product information. The tools are trying to match a product recommendation back to a use case. When we think about your Shopify store and your product page, it’s really about how we can build robust product pages and product information. We want it clean, structured well, and easy to understand, but also answering, for somebody who wants to buy your product, what are the types of questions that would lead them to the page? What exactly is the product for? What situations is this product best for? Why would somebody choose this product? How does it differ from similar products?
We also mention a lot that a product is best for a certain type of situation or type of person, because that’s what the tools are looking for when they make a recommendation. For example, we have a client that sells swimwear. Is this UPF 50? Make sure that’s in there. Is it best for certain locations or certain types of people? There are all sorts of things you can include. It’s about thinking about what would be most helpful for your shoppers, and trying to avoid generic descriptions, especially if they’re coming from vendors or used on other websites too.
Try to differentiate your product pages and focus on the strengths of those products. For example, in your product copy for swimwear, you might ask, “What is the best long sleeve swimsuit for someone with sun-sensitive skin who wants to spend several hours on the beach?” That’s a keyword you probably never would have focused on years ago, but that’s the type of search people make now.
So try to include copy around that on the page, saying this is a great swimsuit for somebody with sun-sensitive skin, or for someone who spends long days on the beach, whatever’s relevant. That’s helpful.
Steve Hutt:
I see. These are some great examples. I’d think adding FAQs and some schema adds a bit more of the long tail, but it’s also a mindset shift about not just going after keywords and the long tail of those keywords, which is traditional SEO. Now it’s more conversational, conversational-type queries. People are using the microphone icon, speaking into their phones in search systems. People are getting more used to these systems where they just want to speak into it quickly, on their desktop or phone, and the natural language is different than typing, and more succinct. You can speak 130 words a minute, faster than typing. I use WhisperFlow, for example, with microphone-to-text translation in real time, because I can’t type 130 words per minute.
I’m more of a hen-pecker. I think that’s what’s happening. A lot of people now are using these systems because they just want to speak into their desktop or phone quickly. It’s interesting to think about these queries, what people are potentially saying into these AI systems to have your product be discoverable. Maybe schema could be part of it.
Rob Langenback:
Yeah, absolutely. A couple of things: definitely schema. That’s super important. It’s always been important for SEO, and I think it’s gained importance again. There are a lot of different ways to look at that too.
When we think about products, we want to make sure we have schema markup in place, but really try to build it out. There’s a lot of optional schema you can include, so we highly recommend that. One that’s often overlooked is variant markup, which shows that you have an item in different sizes, colors, and patterns. That’s really important, not just for the LLMs, but also within Google, which is still the dominant player in search, because we see so many more product grids and organic shopping results, and all these opportunities for products to be visible. The more data we can give the tools, the more visible we’re going to be in their systems. I think that’s really important.
Steve Hutt:
It’s interesting too. I wrote about this recently, the whole structured data side of it, adding things like not just price but availability, having your brand name in there, maybe even some ratings. I don’t think it’s necessarily a ranking factor, but I think it’s interesting to have a machine-readable floor for your product. I think about whether you can extract all the answers out of that, meaning having more raw HTML, which Shopify would have, because there are some challenges right now with reading non-HTML content like JavaScript. I think Google can do it, but the others don’t. Having clean, extractable facts, like dimensions of a product, materials, fit, care, whatever, I think you have to think differently on your PDP page than you would in the past.
Rob Langenback:
Yeah, I think it goes back to the more quality information we can give the tools, the more it’s going to help. You mentioned sizes. If you’re renovating, my wife and I have a room we’re renovating, and we take a picture and ask ChatGPT for some decor ideas. It shows us a couch, a carpet, and all kinds of stuff.
But then the dimensions become important, because now it’s saying, “You need a couch that’s 80 to 84 inches wide,” in this style, whatever that is. If we can include that on the page, the odds of your product showing up for somebody using the tool to get ideas and make recommendations that fit that space go up. Those details become really important, because we have to be really explicit with the tools. We have to tell them exactly what the product is and all those details.
One thing we see a lot with clients is they try to create these really lovely product names, but sometimes they forget to call a shirt a shirt. We’ve had plenty of clients come on board where we look at their product names and they’re not calling the products what they are, the product type. The product type is super important to include.
If we don’t call it a shirt, how do we expect the tools to know that it’s a shirt? We had a client who named their products things like “the Emma.” That’s a nice name, but it doesn’t tell the tool what the product is. We have to be really explicit, give the tools good data, keep it structured and clean, and really understand what would eventually lead our customers to this product, making sure we include all those details on the page.
Steve Hutt:
So that’s the PDP, the product details page. Now let’s talk about the collection page, because from what I’ve read that you wrote, and I agree with this, the collection page on Shopify is the most, and I quote, “the most underrated SEO asset in ecommerce.” Can you make the case and tell me what separates a good collection page from a not-so-great one?
Rob Langenback:
Yeah, I don’t want to say it again, but it really is about the details. On the collection page, we want to make sure we have a great product assortment that’s been optimized and ordered, but really trying to help guide our customers. We like to add some copy to those pages, maybe some comparison blocks and information related to the topic. If you sell kayaks, for example, help customers make a purchase decision on your page. What makes a good kayak? What should you be looking out for? What types of kayaks are there for different use cases and price points? Really try to educate the customer.
I think the big thing, whether it’s on the category or product page, is that we want to provide valuable content for shoppers to help them convert. We don’t want them to go back into Google or an LLM to make these searches. We want to educate them and give them what they’re looking for right there on the page, so they’ll stay and hopefully get the information they need to make a purchase.
Steve Hutt:
It’s interesting too. I agree with this, because I think a lot of the searches that land on collection pages aren’t actually searching for a specific product. The query might be something like “best waterproof hiking boots.” They haven’t thought about a specific brand yet. The person searching for a particular model name is probably already sold, and that’s fine, they can go straight to the PDP page or Amazon.
But a person searching by category, that’s why I think category optimization is hugely underrated. Even Shopify’s own themes aren’t well set up to be collection-friendly. It’s more of a holding area that’s filterable. But the good companies, and I have case studies of my own, have collection pages doing phenomenally well because they have on-page content that answers the queries for that high-level product mix.
Rob Langenback:
Yeah, absolutely. When we organize those pages, there’s a lot of opportunity to create more granular, subcategory collection pages, organizing products around specific needs within that broader category. I agree that when someone comes to a product page, they’ve generally already made the decision, and it’s just whether or not you have the product that fits the need. On the category or collection page, a lot of that traffic hasn’t decided what they want yet, so that’s a great opportunity to educate and guide them into your purchase funnel, showing them why your products are the best.
Steve Hutt:
Right. Let’s talk a little about paid media, because it’s interesting how paid and organic reach and AI visibility reach seem to massage each other a bit. They’re all part of the overall journey of discoverability leading to conversion. When you do audits for a new brand, where are they in organic search, where are they in AI visibility, what’s your process for paid acquisition through paid social? What do you generally find with brands today?
Rob Langenback:
Yeah, it depends on whether it’s paid or organic. On organic, the first thing we generally do is try to get an idea of how well they’re currently ranking, both traditionally and in the AI tools. That’s a little harder because there aren’t any great tools that do a good job of looking at the prompts people are searching, especially because they’re so hyper-specific. So we make a few prompts on our side and see where they show up, at least for the ones we think they should be most relevant for.
Generally, we do an initial baseline visibility report, see where they’re ranking, where we think their traffic estimates are, look at keyword themes, and what their brand is known for. Then we take a look at their website, what they’re doing on category pages or product pages, and try to identify opportunities. Our audit looks at visibility, easy wins based on our experience of what works, and then we share the data. I think data is huge, showing where we see opportunities and where there’s room to lean in harder and drive growth.
A lot of times, there’s a disconnect between what brands think they’re known for and what they actually rank for. They don’t look closely enough at the keyword data to really see where they’re showing up and where they’re not.
Steve Hutt:
Right. So how do marketplaces fit into this? It’s usually a multichannel strategy for most Shopify businesses, being on Amazon, maybe Etsy, other niche marketplaces, Amazon being the largest. All brands want to have a home, and very few, though some brands start on Amazon and stay there, know that the long-term play is to have their own owned audience instead of a rented audience through Amazon. What’s your thought process on the multichannel side of things? It looks like Amazon has been on again, off again with who they’re working with for AI discoverability, on-page, they had Rufus or something they were running, and how it was powered.
And then Amazon still has really high organic reach for a lot of products, and even some collection pages have a lot of reach on a lot of things. How do you balance the Amazon side of a brand versus their own audience on their own Shopify store?
Rob Langenback:
Yeah, that’s a tough one. We have mixed clients with different perspectives on that. The one thing I’d say is when you’re optimizing your Shopify store and growing that channel, you own that. Once you move things to Amazon, it’s tough, because things can change. If you build out listings and try to be in these categories, somebody else can swoop in and undercut your price, and then they take over the revenue on that side of things.
Plus, you start to compete with Amazon. If your product is unique and you sell it on your Shopify store, on Amazon, and maybe Walmart Marketplace and others, when people start searching for your brand or product, you’re competing against all the different marketplaces, and they inherently rank really well. I think it’s a necessary evil in a lot of cases, because there’s such big demand. But generally, selling on your Shopify store is more profitable, and long-term, that’s where you want a lot of your focus, growing that channel.
It’s tough depending on what stage you’re at as a brand, because Amazon has that immediate benefit, you list your product and can generate a lot of revenue. But ultimately, you want to grow your Shopify store, get your customers there, and then nurture them long-term.
We can then sell to them again through email or offers, and have them keep coming back. There’s not that same brand loyalty on Amazon. The lifetime value of the customer is much lower, where if we get them that first time on Shopify or our web store, we hope they’ll come back and we can sell to them again and again.
Steve Hutt:
Yeah, exactly. I also find, and I don’t want to say Amazon is evil, but sometimes it happens where Amazon has all the data, query data, conversion data, they know what’s going on. There are lots of case studies where Amazon has knocked off brands or forced people into more exclusive programs, because they don’t want third-party sellers, they want to own the narrative of a particular brand. I remember managing a brand when I was at Shopify, a hair removal company, and they got a message from Amazon saying, we see that you’re a well-known brand, but we don’t want any third-party sellers on the network anymore. You can sell on the network, but we’re going to buy from you wholesale and publish it ourselves under your brand name, because we want to control the narrative and the pricing.
Rob Langenback:
Yes, I’ve seen the same thing, and that’s why it’s tough, because ultimately you don’t own anything. We’ve seen it happen a lot, where Amazon will buy your product direct, or buy from the same wholesaler, and they don’t care about MAP pricing. If they need to, they’ll get rid of it. If you start doing well within a product category on Amazon, what’s naturally going to happen is there’s going to be more competition, more sellers trying to undercut you for the buy box.
That’s why I said, long-term, you might have a couple of good years or a couple of good months, but you’ll lose that buy box or price advantage. Once people see you’re doing well on Amazon, they’ll try to knock off your product and sell it for less. There are a lot of great things about the platform, but if you really want to grow the brand, Shopify is a great place to be.
Steve Hutt:
100%. I had an interview a couple of weeks ago about Amazon and all the knockoffs and fraudulent things happening, IP issues and stuff, because it’s so easy for Chinese sellers and others to see what’s going on, even on Etsy. Same thing, they see the five-star reviews, see things trending, and there are tools out there to knock it off with a slight change. Sometimes they don’t even do it correctly, and it goes down an ugly path. There are even stories of people getting messages saying, “I bought your product,” even though they bought it from a third-party seller, and having a problem with it. And you’re like, what’s your order number? We didn’t sell you that. It looks like our product, but it’s not, and it goes down an ugly path.
That’s the sad thing about Amazon. I’m not saying you shouldn’t be on it, but you have to be mindful. I’ll put that IP episode in the show notes, because you never know if you’re being fraudulently knocked off. It really affects brands that are more at scale, because once you have traction and sales and brand awareness, it’s easier to knock you off.
Rob Langenback:
That’s actually, I listened to that episode the other day. It was a great listen, and we’ve seen very similar things with our clients over the years on Amazon. As I listened, I thought, yeah, we’ve seen that same thing. When you’re a nobody, nobody cares, but once you start getting any success on the platform and driving enough sales, all these scrapers see that you’re generating revenue, and they’ll knock the product off and sell it for a little less. There’s a lot of great things about the platform, but we’ve dealt with some nightmare situations with clients there too.
Steve Hutt:
Crazy. I noticed there are a lot of case studies. I don’t know if there’s one you’d like to talk about, or maybe more anecdotally if you’re under NDA with a certain brand, but I’d like to get the warm and fuzzy version, here’s what this brand’s life was without 802, and here’s their life now, being managed within your organization. Just curious if you can share any upsides that have happened.
Rob Langenback:
Yeah, absolutely. We work with a lot of different brands at different stages. Recently we’ve had a couple of good smaller brands getting into the space. We had one, I mentioned swimwear before, but also one in cookware, where they were basically a nobody. They didn’t have any traffic, and it’s hard to get ahead competing against huge players.
Going back to what we talked about earlier, from a strategic standpoint, we really focused on how we could highlight what makes them different, both on the collection pages and product pages, really trying to build those out. Calling out things like, “this product is best for all-day sun coverage,” as an example, or these types of situations, and getting really specific. I can’t share the client specifically, but the tactic works really well. In about 18 months with this client, they went from basically no traffic whatsoever from organic to now getting close to 30,000 visits a month through Google primarily, with some LLM traffic starting to come in. We’ve seen significant growth in a pretty short time. It took a lot of work with the client, but I think we’re in a good place, and we continue to see a small brand start to make headway in a large space.
Steve Hutt:
What’s interesting too is what about the DIY people listening, maybe in the early days, trying to figure out what’s the next strategy or tactic they can implement? Maybe they’re not quite ready for an agency partner to take over part of their business, which is fine too, everybody’s on a different journey. I think that’s why we’re here today, to educate people. If it’s a good fit, great, we’ll talk about that near the end of the show, but for early-stage people who want to check and find out if they’re being recommended or discovered in AI tools, you said there’s nothing really great out there because it’s hard to figure out the queries.
But is there any process you’d recommend for early-stage people to start their own personal education about where they come up in AI visibility? Is there a tool or process you could share today?
Rob Langenback:
Yeah, generally what we do is think about our best customers. How do they find you? How do you think they found you? What products are they looking for? Reverse engineer it a little, even by using the tools yourself. For example, if you have a couple of products you’re most well known for, or product types, ask Claude, ask ChatGPT, ask Google, how would a customer find me, or why might a customer come to me? Ask the tools some of those questions, find out what the questions might be from customers, and then go back to the tool and search for the same thing.
Using running shoes as an example, you may find that ChatGPT is recommending your shoes for people in their 40s who had an ACL tear or need extra support, things like that. Lean on the tools to understand your best products, how people find them, what types of searches they’re making, and then honestly go make those searches yourself. They’re so hyper-personalized that you’ll never get the same prompt twice, and people get different results. But you want to find patterns, if you search three or four of these prompts you think are relevant, do you show up? Does a specific competitor show up? Try to find those trends.
We can lean on AI even to understand our customers better, if we don’t already.
Steve Hutt:
Yeah, I wrote about this too. I think there’s a free-ish process. Using any of these AI tools, take 15 minutes and think about maybe 10 questions a real buyer would ask before purchasing in your category. Things like, “best waterproof boots for wide feet under $200,” or “what’s the best natural deodorant that actually works.” If you think about it the way a real human would ask, and run those 10 prompts separately in ChatGPT and Perplexity, and log the answers on a spreadsheet, or ask the tool to log it for you, then in the early days you’ll see, oh, I am coming up, or I’m not. And then you realize you need to work on your PDP page, or your collection page. You learn where you’re not coming up, and who is, so you go look at their PDP pages, their product pages, to see why you’re not being discovered. Then it goes down a bit of a rabbit hole.
That’s the DIY approach. I think those things are available. There are some mid, early-stage tools out there too, like searchable.com, and a few others that will ingest your website, sitemap, look at Google Search Console and Google Analytics, and come up with some hypotheses about where you fit organically and through AI systems. So there are some tools that could help. But once again, you don’t want to buy tools and then not do anything with them. That’s the flip side of the mid-market and enterprise clients you serve, where you have the tools and systems in place, but also the execution layer to actually get things done. Can you talk about your sweet spot of merchant and what your process is?
Rob Langenback:
Yeah, absolutely. Like you said, there are a lot of different tools emerging right now that you can probably pull some data from. Honestly, even just looking at where you rank in Google, using a tool like Semrush, or Moz and others, you can see where you rank in Google, and then maybe take some of those same searches and add some additional prompts within the LLMs.
For smaller stores, it’s more manual right now. For mid-market and enterprise clients, honestly, we’re similar, still using emerging tools. We’re starting to do some prompt tracking for those clients. Visibility is tough. It’s not the traditional “here’s how many keywords you rank for and these are the positions.” It requires a lot of manual effort, deciding what prompts are worth tracking and optimizing for.
What we’ve really lost is that keyword data that says, “there are 1,000 people a month searching for this keyword or prompt.” Part of it is understanding the customer, thinking about which prompts matter most to your business, and trying to get a sample size, knowing we’re not going to capture all of them. We just want to know directionally if we’re showing up, if a competitor is showing up, who’s showing up for these terms, are we being cited, are we being recommended. And it’s changing quickly, two different days can give two different results.
It’s still very much emerging, so we’re testing a bunch of different tools currently to see what works best. For your larger mid-market and enterprise clients, getting some kind of baseline prompt tracking in place, and understanding if you’re being shown at all, is really important.
Steve Hutt:
The other thing I’d add, that I know has been quite helpful for brands, so we have the PDP for product details, then collection page optimization, and then I’d also argue that creating a brand is really about creating community around the business, having some kind of blog or way of interacting, maybe not just on social, which is an owned audience but also somewhat external, though still a necessary part of building social proof around a business. What’s your mindset around blog content, and is this still traditional SEO built into the topics and conversations happening? It would seem to me that LLMs are crawling these blogs on ecommerce sites and citing certain things, because there’s no reason you can’t have a blog post about how to wear a particular product in a hot summer in Europe, or whatever, and then link to a couple of pieces that make sense and are trending right now, still leading back to a product page. What’s your thought on blog content as it relates to discoverability?
Rob Langenback:
Yeah, it definitely is still a great tactic. I think blog content is a great way to influence the results and recommendations. Using the running shoes example again, if ChatGPT is going to make running shoe recommendations, it’s a great opportunity to publish blog content around, say, the best running shoes for different situations, types of people, or use cases.
The biggest shift we’ve seen with blog content is asking, will this content help a potential customer buy our product, or convert? We write it at different stages of the buyer’s journey, but the big thing for us is asking if this will help someone convert, help them make a decision. If so, it’s still a great tactic. Top-of-funnel content, like how to use the product, or even post-purchase content to support customers after they buy, is good too, but we really want to focus on content that helps drive conversion.
That content helps the purchase decision, but it also influences the results, so it’s good to talk about the products you sell in a way that explains different use cases, what sets them apart, the benefits, why someone might choose one over another, and how it compares to alternatives. Hopefully the LLMs read that and surface some of it in their recommendations.
Steve Hutt:
Yeah, this is amazing. I know 802 has a lot of other internal services available, paid media, PPC, and traditional SEO, I believe is still there, plus content production, which we’ve talked about. There are a lot of acronyms floating around too, answer engine optimization, generative engine optimization, and that’s a whole separate service you have around becoming AI discoverable and this agentic commerce future, however that looks.
I want to understand what services you have available, to give people an opportunity to see if there’s a fit wherever they are in their journey. Can you share the services you offer and the type of customer that makes the most sense?
Rob Langenback:
Yeah, absolutely. At its core, it’s everything search for us. We do SEO, search optimization, and paid search, think pay-per-click and Google. We also do some meta advertising, but primarily it’s about showing up in search results, or LLMs, if you want to think of those as search too. Even on the ad side, we’re seeing ChatGPT roll out ads, and in three months they’ve made so many changes to what your budget needs to be, what it takes to qualify, and now anybody can do it. We do that too on the paid media side.
SEO, at its core, when we throw around acronyms like AEO, GEO, AIO, ultimately the way we look at SEO is that it’s inclusive of all of those. When we optimize, we want to show up in traditional Google results, but also in Google AI Overviews, ChatGPT, or Perplexity. From a strategic perspective, we put together strategies and tactics that make our clients visible across the whole search landscape, however you want to look at it.
Most of our clients are ecommerce brands, a lot on Shopify, and I’d say when you’re scaling, newer stores under $2 million in revenue are often doing a lot of this internally. But as they try to scale to $5, $10, or even $100 million, that’s where we really excel, helping brands scale to that next level, trying to double revenue over the next couple of years, whatever that may be. Obviously, they need to have a marketing budget and be trying to grow their team and continue scaling.
Steve Hutt:
Yeah, this is amazing. All right, so it’s 802.com, the actual words. I’ll put that link in the show notes. For those who fit into that sweet spot, where you believe there’s some discoverability challenges you folks can help with, there’s a contact page, you can go to the link in the upper right corner. Have a chat with your team, almost like a business partner, audit where you’re at today and where you believe some challenges are, and come up with your own hypothesis. You know the old saying, you can’t see the forest for the trees.
Sometimes it’s smart to have an independent third party, at no cost, take a look at where you’re at today. If it’s a good fit, or we believe we can help, because sometimes you can’t help everybody, knowing where they’re at today, you might be able to come up with a process that could assist their business. I think that’s great.
Rob Langenback:
Yeah, absolutely, and always happy to have those discussions. As you mentioned, if you visit our site, we created a URL for you. Go to 802.com/fastlane, and feel free to reach out. We’d love to offer a free 30-minute consultation for all your listeners. Reach out, schedule an appointment, and you’d probably talk to me or somebody close to me on our team.
Really, it’s just about having that conversation. We love to walk through where we see opportunities for your business, and if it’s a good fit, we’d love to work with new brands. But if not, we’re pretty transparent, we don’t want to try to sell you something that’s not going to work.
We’re going to be upfront if you’re probably not at the right stage for us now, or if we think this is what we can do for you. That’s one thing about our team, we try to be really transparent and look for situations that are a great fit for us and for you, where it’s going to be a long-term partnership if we do move forward.
Steve Hutt:
I love it. So 802.com/fastlane, that’ll redirect you. It’s a really nice page with a book-a-free-consultation option, and they’ll take a look and see if your store’s showing up in AI search or not, and what the next steps might be. This is great, Rob. Thank you so much for recording today, and thanks for being honest and transparent. It’s such a moving target right now.
Rob Langenback:
Yes.
Steve Hutt:
But I appreciate you sharing here’s the lay of the land, here’s where we’re at today. It’s a nonstop process of learning, iterating, and sharing information that’s out there. At the end of the day, there are some basic principles that need to be implemented, but it is a moving target when it comes to discoverability and tracking that, and some on-page things we’ve gone through with the PDP, collection pages, and blog content. This has been really cool, I’ve learned a lot, and hopefully my listeners have too. Thanks again.
Rob Langenback:
Oh, I appreciate it, Steve. It’s great to be on. Hopefully any listeners who want to reach out, I’d love to chat. Thanks for having me.
Steve Hutt:
All right, sounds good. Have yourself a great day.
Rob Langenback:
Thank you.