Hello Retail Conversations
Episode 04: Shopify and The Age of Agentic Commerce - Peter Sommer
101 views
View transcript
Today I'm super interested in two particular topics. I want to hear more about the Shopify community and its quite explosive growth these years. And I want to talk to someone on the consulting side about the roadmap to shopping beyond the website. And with that, I want to welcome you, Peter Sommer, from our partner Details and just want to give you the chance to introduce yourself. Thank you very much, Rasmus. I'm the CEO of Details, which is a partner that has worked since 2016 solely with Shopify e-commerce. So what we strive to be is the best within e-commerce when it comes to Shopify. We have two things we help our customers with. We help them migrate to Shopify. And then once they are at Shopify, we help them get the best out of Shopify, grow their business and make sure that they have a leading competitive edge. What about your own background? What did you do before that? I've been within the IT industry for more than 25 years. As a personal background, my entire career, I've strived to become the best leader for the people I lead. So that's my personal aspiration. And I've worked with a number of tech companies recently before joining Details. I was with IBM, where I was a partner in the consulting division for four years. Before joining Details, I was with IBM, where I was a partner in the consulting division for four years. I mean, there's a lot of people will know about Shopify for very small businesses, the whole self-serve part of their business. But that has obviously expanded so much beyond their roots with Shopify Plus and so forth. Talk me through the experience of being part of that community. So I think first and foremost, what comes to mind is that it's never boring being part of the Shopify partner community. Shopify has had an explosive growth and keeps on having it for many reasons. I think first and foremost, they have more than 3,000 software engineers that produces more than 100 new features every six months. So twice a year, we celebrate together with Shopify all the new things on the platform for the last three months and celebrate the things to come for the next three months. And looking back in the last two years, I've never ever come across a software platform that has grown with such a high stability on new features. So that's one thing you get with the Shopify community. And the whole Shopify Plus sort of change for the Shopify company, how has that worked out from your perspective? The whole change. Can you elaborate a bit? There was a time before Shopify Plus, I mean, and that sort of their whole rebranding exercise of let's split up into having two different paths or sort of the Shopify Plus brand as opposed to the Shopify roots. Is that, I mean, you're a Shopify Plus partner, right? That's correct. Yeah. So Shopify has rebranded their partner strategy and has created more tiers. Today, we are a Shopify Plus partner and has been for many, many years, actually ever since we started back in 2016. I think it's a natural evolution that Shopify is undergoing. When I joined the Shopify community a little more than four years ago, the biggest customers of Shopify were quite limited in size. Today, you now see the biggest brands globally joining Shopify. So I think recalibrating your partner strategy is a very important part for Shopify to keep fueling the business as they see it coming. Also in the years, they've now split out a couple of years ago and now they have a particular focus on B2B. Details have worked with Shopify and B2B for more than five years. So we actually built a additional functionality that lacked on the platform five years ago and we're just thrilled to see that Shopify finally took the step, which has also fueled our business quite significantly. So today we have a much bigger business on B2B than we had before Shopify launched this, which is a huge step for us. And in general, when customers come to you from other commerce platforms, what is it they're hoping for as they sort of migrate onto Shopify? It's an excellent question. I think there are two things that comes to mind when you ask that. If I summarize the customer meetings I've had over the last six, 12 months, I think there are two things they're looking for. They're looking for speed. They're looking for self-service. And on top of that, you will get a lower cost. So I would say the business people want faster time to market and higher self-service. The financial people, they want a significantly lower TTO. And they really want to flip the 80-20 rule because today they spent 20% of their IT budget building new new features and functions and 80% maintaining the platform. They want to flip that. They want a platform that can help them grow with speed so they don't have to do all these customizations themselves. They want to be part of a community that takes care of the business problems as they arise. Ranging from anything from integrations to payments to all the different disciplines you have within e-commerce. You mentioned that Shopify is sort of pushing out features in the hundreds every year. Usually it takes quite a while for businesses to adopt new features. And you as their consultant will be sort of a key part in advising customers on this feature is important for you. This feature is not important for you. How do you sort of get through that many new additions to the Shopify platform? And figure out what is advisable for our customers to adopt and what is just noise? So I think at a day-to-day business, we follow the change log very closely. Every six months, you have a celebration from Shopify where they launch this and puts all of these changes together and shows a little bit about what's to come in the coming three months. We take great pride in details that we have all been born on the other side of the table from where we sit today. So the customer facing team have all been merchants themselves. They've been in the shoes of a merchant. So when they get new features, when they see this, they can quite fast find out what's relevant for this market, what's relevant for that particular industry or customer. So that we can curate the massive number of features for our customers. So that's actually a very important part of the value that we bring to our customers is that we curate what's relevant and what's not. And then we keep up making sure that your platform is current so that you can harvest all the value that you get from Shopify. That makes sense. I know that if we turn this into a slightly different direction, I know that you're sort of quite aware of the changes happening with internationalization and sort of the ability of customers to launch new markets in a different way. I just want to sort of hear your thoughts on why is that suddenly working? I mean, that's been the ambition of merchants for forever to sort of broaden their reach and get into new markets and do that effectively. Why are they suddenly able to do that now? So I would say that going back to your previous question, what do you get as part of being a member of the Shopify community? You get a tech partner that is leading within their industry, but in that they also think forward quite aggressively. So if you look at the top from Shopify, Toby has very clearly stated that Shopify is going all in on AI. I think there's been a lot of buzzwords and talks about AI, but I think we are now seeing that the platform that the platform that the platform that Shopify has built is excellent to actually grasp both long -term AI relevant functionality and features, but also today you can actually get value from the way Shopify has been built. Shopify differentiates the way I see it in having a fantastic strong data model. With that data model, you can easily get data in and out, which is key towards AI. If you look at internationalization, what we see customers are doing today is that when you create products, getting a product title and product description is typically designed to follow a template so that you fill out your products in a certain way and give your customers an unified experience across your products or across your assortment and vendors, depending on if you are multi-brand shop or if you are a multi-brand shop. If you are a single vendor. So creating that in an engaging way where it makes it relevant for the human SEO search words in the past and now LLM models is key. So that's one of the examples that we see today is that you can actually quite easily create an engaging product text, product description and fill out some of the gaps you have in your data model in a much more efficient automated way. When you then speak about internationalization, what we see is that in the past, everybody was excited about Google Translate. Everybody found out that wasn't good enough. That was far from being something that was acceptable. I think we all became very excited with ChatGPT that kind of took the commercial part and storm throughout people. And now we see that using these LLM models correctly, you can actually translate your language within respect of the high-end brands that really cares about tone of voice, cares about pronouns, cares about their audience in their local markets. So today you are able to actually auto-translate things with super high quality and really take out many, many conversations. Are you seeing customers sort of launch into new markets, like real-life scenarios here where they're now doing that at a different pace than before or suddenly being willing to go into new markets or daring to? What's the real-life cases you're seeing right now? So we have a manufacturer of high-fidelity equipment. equipment that for 25 years has produced some really high quality content videos text articles blogs FAQs community everything in English because they had a global market and had at the time decided they would not translate this content because it was simply too big a task too costly and the business case wasn't there what we found when we did a pilot with them is that German was their first market was that using a editor for a highly recognized magazine within high fidelity as the test customer we were able with AI to actually tailor made the tone of voice where the German language you still have the ability to speak to your customer in the more polite way using see or in the more direct way using do but they didn't want to offend any of these customers they didn't want to take side in this so they actually decided to take a third person pronoun so what we've seen is that with the correct use of AI you can actually translate content in a way where they had been using more than a year to translate 10 100% and within a month calendar month we were actually able to translate all of their content and two things here actually because if we hadn't had Shopify I would question whether this was actually doable and feasible because as a partner it's so easy for us to take data out of the platform that being a product text metadata an article a blog post a page the whole webshop experience you can actually take all of that and then put that data out translate it out translate it accordingly to your instructions qualify the quality of that and then put it back in fully automated so very practically speaking we have shared customers you guys are obviously implementing Shopify on the hello retail side we have a range of products two of those are search and recommendations which Shopify also provides so talk to me about sort of what's the tipping point when you're you have a customer you have a customer you have a customer or an existing one that are sort of tempted to say okay let's just go with Shopify native solutions what's the tipping point where you feel okay they have they need something more advanced than what Shopify can support specifically on search and recommendations I think there are two ways around this there's a customer experience way around this answer and then there is a business business case so I think there is a business case so I think what Shopify has done over the last couple of years is to improve their search recommendation engine dramatically but you still have quite a number of customers for which that doesn't serve their customer experience so when I enter this and that what what do I want to show and how do I want to be in control of that so that's a UI where it's clear that many of our customers has a different expectation to what Shopify Shopify is able to do Shopify is able to do today some is language based some is about the vast amount of products they might have in their assortment so so what is it that they want to provide to their customers when you enter the search field if you look at the business case the way that Shopify have set up their search recommendation engine is that you have to curate a lot of this in a manual way you have to set it up and maintain it where you have to cut it up and there's you have to cut it up and then there's you and others have automated this quite significantly and then it becomes basically a business case saying if I automate this what's the cost saving versus what's the user experience and how much more business do I get from this and we clearly see that the larger the customers are the bigger their product assortment is and the further up they are in their advancement of being a mature e-commerce merchant, the more you need tools like Hello Retail to fit the needs of their customers. I wanted to talk about that other big topic as well, sort of the roadmap to shopping beyond the website. And the background is, of course, that we're all sort of hearing all of the stories about here's a way for you to sort of just use your LLM chat and then you can sort of do a lot of your research into whatever product category you're interested in. And then eventually get links to actual products you can buy and even at some point do the checkout as part of your sort of chat experience. So I'm curious about the kinds of conversations you have with customers right now that are sort of in that direction of how do we prepare for that future? What can we already do right now? How far are you on the Shopify side in terms of sort of getting product data and parts of the shopping experience into the LLM and where it's not just happening on the online platform? online store that's completely controlled by the customer. It's an interesting topic because I think we all kind of assessing right now how fast the market will move. I think we all see the direction. I think like with other technologies like self-driving cars, you never know how fast that will move. You predict something and then you need to kind of adopt your business to actually accommodate these significant changes. So if we zoom in on commerce and LLM models, which I think today is the contours of what we see will be happening is that you will have a conversational AI that will then interfere or be part of your shopping experience. I think there are two things that we need to take into consideration when you enter such a conversation. First and foremost, how is this possible? Is it possible at all to actually enable all of this data sets? In a structured, in a structured, easy way where it provides the correct data to the LLM models so it doesn't hallucinate and give you false information about a product type or factual specs and stuff like that. And Shopify actually just launched what they call agentic commerce just very recently over the summer. What agentic commerce means is that Shopify will basically build a catalog of all products across all websites. So all of their customers, they will curate all the product data where you want those products to be part of that catalog. They will curate that information so that a LLM model can easily access that data and get facts at the fingertips of that conversation. So they've been able to have a completely new. So they've been able to launch this with speed. I'm actually amazed how fast they came out, which is a big announcement. On top of that, they've launched a universal cart, which then enables you to have a completely whole new shopping experience. So you are able to, with this interface, to curate a AI conversation, ask the LLM agentic agent that you set up, can be any technology you want, you want, or you want to have a completely new product that you want, or you want to have a completely new product. So you can actually put that into a unified universal cart. So kind of like your big, magazine warehouse where you have different brands and vendors on the different floors. And you can actually exit at any checkout. In any floor, you can check out, pay your bill as a unified experience that you don't have to pay for everything in each area of the shops. You can put this in one universal cart and actually carry out the goods from there and then process your shopping experience. Is that a conversation that you are having with customers now? How to prepare for that? Or is that sort of two early days for them to even wrap their heads around that? We just sort of, maybe they've just been sort of pushing a direct -to-consumer strategy for a number of years, and now the whole world is sort of changing underneath them? Or where are the customers in those kinds of conversations? I think we're having two conversations. So looking at this unified commerce where you can actually have this agendic commerce experience, that's a dialogue we only have with our customers who are big within internationalization because the feature set is right now only enabled in North America. So the products that you offer on your webshop, if that's enabled for shipping to the US or Canada, then you can have this conversation. Which of course limits the number of customers to those who have that as a primary market. I think we discussed with more or less all of our customers is the big shift from actually, if you look at the buying experience, I think it changes quite a lot. Just very recently, I was asked to help a family buy a printer, and I'm not very tech-savvy within printers, so I promised my mother-in-law to help. I helped her curate the big offering of printers. I helped her curate the big offering of printers, and I think it's been a couple of years since I bought my last printer. So I start from scratch and have to do the research myself. I entered a few search words for color laser printer because that was kind of the type of printer I was looking for. And in my Google search comes up a lot of black and white printers that I ended up spending time actually believing was a color laser printer, so I was directed to a lot of time wasted because my search and the product shown were not relevant. So I actually flipped to one of the LLM models and again maybe damage from my line of business. I wanted to test out how can this shopping experience look like if I use a LLM model. And I found myself actually getting the advice where I simply put up saying I had four or five criteria. I wanted a wireless printer. I wanted a color laser printer. I wanted a multi-function printer. And then I would love to have a printer that could print on both sides of the paper. And I gave that information. And then it came back with a range of five printers that matched those criteria. Based on the price range, I could easily narrow this down to two vendors and two specific products. And then I went back to Google and found the vendor who offered this product for my market where I wanted a chip with a warranty and all the things that are important on how to proceed with my procurement. So I think that actually reflects the conversation we are having now with customers is that it's not massive, but it's definitely changing. And I think it will change a little bit more rapidly than we believe at this point in time. We have several joint customers that are obsessing over brand DNA, sort of creating shopping experiences that they control completely down to the last pixel. They want to be sort of in control of their brand and the shopping experience they provide for their customers. So a lot of that goes away, but probably not all of it. What you're mentioning is that the sort of the whole initial research phase might happen elsewhere. But then you might also still want to be connected with your favorite shops in a more sort of intimate way or you want to sort of be exposed to that shopping experience of the online store brand. But sort of what do you think will happen, honestly? Is the whole thing just going to move to shopping by the LLMs? Or will there still be a need for you guys as an agency to create online stores with a lot of care into the shopping experience that is happening there and the brand and campaigns running on the site and so forth? What do you think is going to happen? I think that we need to really help our customers understand the changes because with changes comes uncertainty and doubt. We need to make sure that we need to make sure that we need to make sure that we can advise our customers on how to proceed, build the roadmap and help them execute that to achieve their goals, which is the nature of the consulting business that we have today. If I circle back to what you said just previously around will this whole thing change completely and the shopping experience, I don't think this is unlike for the vendors who actually produce and manufacture a product today. Many of these have many channels. Many of these sell to shops, so not online shops, some do as well, but actually physical shops. And today the shopping experience in a physical shop is more like the agentic commerce just on the internet. Because if I walk into a big sports magazine, they might have many different brands. For some of them, I might have a preference. But if I want to buy a pair of new running shoes, I would definitely go to a shop that knows how to advise me on picking the right shoes for my running patterns and that feeds my needs. So I think we need to see the change that will slowly happen and maybe faster than we can predict will become like the advisory that you see within a shop. But if you look at the shopping experience where you actually made up your mind and you just want to procure it, the need for an advisory agent is different than for the one that can actually support you in your decision making process. So I think there are two different roles that the LLM models will play. One is being the advisory, like the person in a shop, and then help me pick out the shop that fits the needs. So if I go back to my example for the printer. I kind of started from scratch and needed advisory. I actually found that using a LLM model. If I step back and look at the advice I got, that was because I got good quality data feeded into an engine that understood the needs and could pick up the conversation I had and help me narrow down to a few products that fits the criteria. That's the criteria. That's the criteria. That's the criteria I set for myself for this situation. Then you would have the second part of my own customer journey was then to pick the vendor that was right. My experience was that Google was actually much better for that. So once I narrowed down the part number or the two part numbers, it was faster, better, more safety for me to go that route. So as soon as you actually get your, you as a shop, enabled to be able to be part of that conversation. So kind of like steward in a big mall where you have multiple competitors. You might have two store magazines that offer sports goods in one magazine. Then you can send the person to the right store that fits your needs. Whereas today, I found that advisory around the products, it was super capable of that. And I think that's what Shopify is trying to solve as well is to help the merchants fit the merchants fit the needs of their criteria. Is the product in stock? What's the price? What's the commercial conditions around procuring from this particular vendor if it's a multi-brand shop? So I think a lot of changes will happen. And I think you can actually see, look at Amazon as a competitor. You actually have that multi-vendor experience today. I think that's what we are now going to see being mimicked by AI and Shopify. So enabled by Shopify and AI, you will actually have a unified experience where you can shop across all the shops and get the advisory. You can get your selection criteria narrowed down. You can even get decision support if you are insecure. What is important? I could have started by saying what's important when I select the printer and I could get the guidance there. So I think we are now starting to see a big change that will happen in steps and in phases. And I think the first thing that will hit us and has actually become reality is that if you put in a long search word, a long search in Google, you would often get the first response as an AI generated response. And it's a whole different game becoming part of that response than getting your key search words that we know has been a game for many, many years. So I think you will see this change in stages. Yeah. Do your customers, the e-commerce businesses that you were talking with today, do they sort of realize by now the criticality of loyalty then? Those for both B2B and B2B and B2C actually. Do they sort of realize the, what sort of the degree to which the earth is shifting underneath their feet at the moment? I think that is the elephant in the room. I think that's the big conversation to have because honestly, nobody knows. I think there are a lot of smart brains out there that makes projections. I've been in this game for long enough to be part of these new, exciting announcements. If you look at the whole voice commerce, I think it's seven, 10 years ago, it was super hot with Alexa in the US where everybody thought you would shift all your procurement to Alexa and these smart devices, smart speakers. I don't know about you, Rasmus, but I haven't made a procurement of anything with smart devices in the last year. Same for me, for sure. I've never tried that. So I was part of that announcement and actually was kind of trying to ingest to myself what that shopping experience would look like and trying to imagine what the world would change into. And I think it was a huge enabler. I just don't think that it provided the shopping experience that the common user, the common customer wanted. So I think it all comes back to the shift in trends. I bought the iPhone one back in 2007 and was on that. I think that was a game changer from adding mobile to your number of channels and your devices. But again, today with all the dynamics going on, is that a big thing? I think it's all about a channel that serves the purpose that fits my needs. So what you're saying is that the mature choice for a mature e-commerce business is to see LLM, Agentex Shopping as another sort of channel that can be used across all of the other channels that they're already using. Is that what you're saying? That's step one. That's step one. And I think we would miss out on a huge opportunity if we don't grasp the value you can harvest today. But I think I'm a little bit skeptical about being the one to project where the market is going in five or 10 years. Because I think I think I've seen too many big changes that actually didn't really radically change the world. So I'm more into the next step and the next step after that. So what we advise our customers to do is to look at reality today and then harvest the value. I think many of our customers produce very nice products, have a brand. And I think commerce and the way to procure this is just one part of that strategy they have to really understand and embrace. But I think what we advise our customers to do today, whereas there was a whole industry for many years about just keywords and search, trying to outsmart Google's algorithm. I think those days are definitely over. Now it's back to real content and honesty and trust and loyalty. So that if you have a brand that is not trustworthy, then I think you will miss out very soon. Because if it's not commonly accepted that you as a brand with your user feedback, with forums, with all the places where people talk about your products and assess them and exchange ideas, that being cars where you have complete blocks and sites where people talk and help each other, those communities, those communities, they are out there. And the big changes that any customer will know about that through the AI and the advisory. So I think the good version of this is that the good will become the victory of the future. Because if you do good and you do well and you excel, the bad part is that if you have enough money, you can buy yourself into the dominance. But I think today you would need to make the steps for you to really succeed tomorrow and then really focus on this, make a plan and then make sure to harvest value as you move along. That would be my best advice. That's the perfect way to end. Thank you so much for coming here today. You're welcome.