Thomas, welcome to the 2X eCommerce podcast. I’m delighted to have you on here. Where are you calling from? 

I’m calling from Charleston, South Carolina. I’m also excited to be here. Thanks for having me.

I hear there are some weather challenges there but it looks like you’re faring well.

We are. It was called for a big storm, up to 70 mile an hour wind. It seems to have missed us, some rain, but we’re doing fine. Thank you.

There’s a bit of a cold spell here but it’s all good. I’ve had a conversation with you. You have an excellent backstory, which we have to share with the audience. A top-level overview for those who are reading, Thomas is the Head of Strategy at a CRO agency called Bear Group. He was previously the founder of Scoutside, which was acquired by the Bear Group. I’m not going to be your spokesperson. I would love you to share your backstory, please. 

Happily. Thanks for that intro. I the Head of Strategy now at Bear Group, an agency that focuses on growth for eCommerce, both D2C and B2B commerce, and helping merchants grow and scale through all phases of their needs through strategy, design, UX, great engineering, and integrations of systems to help for scale and growth. We do that with a data-backed approach. To get there, there’s always a backstory and mine is rooted in founding my own agency called Scoutside.

We were focused on the very same work focused on eCommerce and helping merchants scale and grow. That started from my education, which is computer engineering. I studied computer engineering in school and realized that I was being trained to write code all day every day but yet I loved working with people, interacting with people, customers, and clients, and interacting with team members and growing a team.

I started to shift to beyond just computer engineering and into management of technology is what they called it in school. Out of that, I started building websites and enjoyed it. I liked that work and wanted to do more of it so I continued to hire people. I focused on eCommerce. Out of school, I worked at a software company making in-store solutions for retailer, POS, back office, and fulfillment. That’s where I learned all of my retail knowledge.

I realized that I could work in any industry but if I like the industry, if I was a consumer of that industry, meaning I simply buy things, I was intrigued by the logic and work would be more fun. That was a lesson quickly learned. I didn’t want to make in-store solutions. This was 2009 and 2010 and it felt like the internet and synapse were where things would go. I started focusing on eCommerce and grew Scoutside. My goal was always just to do a great job at what was right in front of me. I didn’t have a big master plan for merging or acquisitions.

I loved what I did and I love what I do. I wanted to do more of it, I wanted to hire more talented people, and I wanted to grow a team and that’s how we set sail. We built a great team and company and went through a lot of different phases of growth and learned a lot. It’s always interesting to reminisce on that trajectory. There are a lot of different growing pains that you encounter as you grow and some of ours were going from 7 and 8 people up to 14 that had a big growing pain. We were then hitting another around 18 and 20 people and that’s where Scoutside maxed out.

It’s a little bit by choice but we started hitting some growing pains in 2022. As I looked into the future and thinking about how we grow and how we reach the next tier of growth, I started plotting out our org chart and we needed to hire seven more people. We needed to build out a sales department. We needed to have more strategists and account managers to help assist. That leap was pretty daunting. That would have been a difficult jump to make in 24 months, 18 months, or 9 months. That’s what we needed to do. We needed to do that within one calendar year, hire up 5 to 7 people.

In that realization came along Bear Group that also wanted similar goals. They wanted growth goals, they wanted to build out their team more, and they wanted to offer more services, specifically into strategy and design to help merchants grow. That seemed like a great opportunity for both agencies to align strategically, merge, and offer more services to all existing clients and set sail for growth. That’s what we did in all of 2023.

You’ve been running Scoutside for about twelve years. You were fourteen at the time and about 45 now in Bear Group. We’re talking about almost a 4X in your throughputs. I hope you’re settled in Bear Group now. Let’s jump into our episode. What we want to talk about is personalization in eCommerce. How would you define personalization in eCommerce, particularly in the context of all that’s happening in 2024 and all that’s happened prior to 2024. What is personalization?

As I would define it, and many would agree, personalization is the concept of delivering more relevant content to the visitor based on something that we know about that visitor. Not all visitors are the same of a digital experience. Not all customers are the same. How do we capture some data about that customer or potential customer or visitor and serve unique content to better put in front of them what they’re looking for or what they need?

The first bit of your definition, how do we capture data about customers, how do we and what data points set the foundation for personalization strategy? 

Great question. That’s a big part of our focus at Bear. In everything we do, we take a data-backed approach. We’re no longer guessing or even only using best practices, which what we do, but we supplement that heavily with data. When it comes to personalization and data, there are several things to consider. The terms often used are zero-party, first-party, second, and third-party data. Of course, what was working in the past up till about 2022, most merchants were relying on third-party data from Facebook and Google.

As privacy has started to become more popular or is getting more attention and consumers are more aware of tracking methods, which is a good thing and most certainly has a place in the market, what’s happening the relevant content that was being delivered is now becoming a little less relevant, which ad spend goes up and return on that ad spend goes down because we know a little bit less about the visitor.

How do we handle that now and into the future is into this conversation of zero and first-party data where you’re asking the customer questions. They can come in all sorts of shapes and forms. It can literally be a question. For example, if you’re selling hand soap, how many bathrooms and how many people live in the home? A question like that can help you understand how much you should buy and how frequently you may run out. You can ask a question and start collecting data about that visitor.

Of course, there are still older methods that are very relevant, such as any information you can glean from the browser, such as IP address when available, and start digging into demographics. A great example is TheNorthFace.com. If you go to The North Face from Seattle, you’re going to get rain jackets. If you go in Miami, you’re going to get bathing suits. It’s a simple example of personalization and how that is helpful as you land on a site. It runs the gamut on how to collect, ask for, and even not ask for information.

I like that North Face example, which is geo-targeted. It’s interesting. I was going to speak to a specific personalization question but now that you brought the Miami versus Seattle example of the raincoats versus the bathing suits as a first step, to me, that sounds more like geo segmentation. It’s not 1 to 1, from my perspective. It’s like, “You live here and you’re in this basket so we’re going to serve you this experience,” which makes a lot of sense.

For a lot of retailers reading this podcast, many people haven’t taken the first step, myself included, into personalization and thinking about personalization. A lot of us are very much concerned about acquisition for first time customers, as well as retention of existing customers, and trying to maximize CLTV. I know personalization is part of the story but for merchants like myself, and I’m speaking for many of the audience who are reading this conversation right now, what do you think are the low-hanging fruit or first steps we should take towards personalizing experience in our stores? Do we take The North Face approach or do we go a bit more granular, a lot more vocal, or a bit of both? 

The answer to that question lies on a scale, like everything. I believe in all new topics, meaning if someone’s reading and they have not started working in personalization, it’s about making that new concept and new technology feature approachable. We can talk about big solutions and we can talk about small solutions and ways to tiptoe but I like making things approachable. You bite off and you see the value of starting with a simple concept of personalization. You start realizing the benefits. You start feeling, “This is approachable. This isn’t a scary or big topic or expensive topic.” It can be inexpensive and start seeing the benefits of that.

Once you do, take that cost benefit analysis and start rolling it out into bigger solutions. That’s how I approach things. Following that lead, a quick example or advice is to look at your email marketing system. Of course, an industry favorite is Klaviyo. Take a look at that. Take a look at a newsletter. Are you sending the same newsletter to the entire list? If the answer is yes, you’re doing it wrong. It’s that simple.

What we know, and you brought it up too, is segmentation and that’s the root of personalization. What we know is that when the content is more relevant to the reader, the visitor, or the viewer, it’s more likely to result in the expected outcome. If that’s having them buy something, you’re more likely to have them buy something. If you want them to understand the brand and product value proposition, you’re more likely to achieve that goal too. You take any bit of information you can and you start delivering different content to that user base. Email is a great place to start.

Another approachable area is related product. This is not a new concept at all. If you think about it, back to my other point on collecting zero and first-party data, sometimes it’s a direct question and sometimes it isn’t. What does that mean? If we know that a visitor is on high heels, if they’re looking at high heel shoes, we can make some assumptions about that visitor. We can start displaying more relevant content. Related products is a great area.

If the customer is on a particular product page, you can assume safely a lot about that customer. They’re interested in high heels. They’ve made it this far to that product page. They’re interested in high heels. They may not be interested in things like flip flops. They may not be interested in hiking boots. They’re interested in something a little more formal. What you shouldn’t display as a counter example is, in related products on that product page, you shouldn’t show flip flops. You shouldn’t show hiking boots. You want to make the related products, this age-old concept, intelligent. That’s another great way to start.

There are a lot of apps that do this that are price-effective.  What they’ll do is the app will take into consideration the product page that the visitor is on and, in real-time, analyze what else shoppers are interested in while also interested in that product. It will return to screen the top ten products that are often considered when viewing a high heel and you start to display those items. An idea is you can start displaying other shoes that are in that category.

I’d also recommend expanding that and having not just related products but products that pair well with that product, maybe another ribbon of products in a different category such as pants that are relevant for those high heels or a top that are relevant for those high heels. Those are easy ways. Most merchants and most eCommerce themes come with a related product section. That’s out of the box.

The next step is making those recommendations intelligent. Those are two great ways to start engaging with personalization. That has a low-barrier to entry in both cost and implementation. Both of those will result in improved conversions and that’ll help the merchant start to bite off more into personalization. It makes it more approachable.

What I picked up from what you said is to start personalizing your email. Even if it’s from a segmentation standpoint, there will be certain behaviors or certain preferences segments of your database would prefer. Start from there because those are conversations that feel one-to-one to customers. Segment your email marketing strategy. This is beyond flows. This is more newsletters and the way you communicate offers to your audience.

The second is more in the merchandising section, which is the related products, recommended products, and intelligent recommendation. There are tools out there to help you expedite the pairing, whether it’s based on what you viewed or based on what you’re likely to buy together, the intelligent bundling. You mentioned the stack for email, the de facto stack. There’s Klaviyo. We must say that there are other providers such as Amass, Manysend, and a few others. In the merchandising piece for personalization, given the fact that most of our readers are running Shopify or maybe BigCommerce, what recommendations would you have from a tech stack perspective? 

Great question. That’s an easy answer. The platform is called Rebuy and it’s terrific. I know those founders and they’ve grown to a large platform, successful and effective. Now they’ve reached into other areas beyond just recommended products on the PDP. They have an intelligent cart drawer. It can fully replace the cart drawer inside Shopify and other platforms and makes that intelligent as well. You can start introducing recommended products based on what’s in your cart. It has some great subscription features and loyalty features that plug into other third-parties as well. It’s a great system.

I’ve come across Rebuy a few times. Given your strong recommendation, I would look into it again. The other question I have was more around a personalization side. Once you’ve got those basics sorted, the merchandising piece and email piece, what do you think the more intermediates eCommerce merchants or brand would start looking into for personalization or to personalize their user’s experience or shopper’s experience on their website?

On the scale of personalization, if we’re making three categories and one is maybe beginner, intro, and now we’re an intermediate and then there’s advanced. As you’ve asked more about the intermediate personalization, it depends a little bit on brand-to-brand. This is something that we love to do and a great app but we like to think of the brand holistically and understand their products, customers, and their different offerings.

Some great examples that we’ve worked on and are proud of is we worked with a great water filtration system for consumers. They’ve got pictures, in-fridge filters, and under sink filters. They came to market finding the current offerings in market a little trumped up by marketing and thought they could build it better and they have. The technology is there and incredible. Their approach to market is very scientific, which can be hard to convey digitally to the average consumer.

The challenge there was, as we redesigned their site for scale and growth, we also wanted to push the values of the brand and products but in a very approachable manner. This brand is clearly filtered and a terrific brand.  To personalization specifically, what they have done is create a relationship with the only database of municipality water in the entire nation. Well beyond the state, beyond the city, and right into the actual county, they have real time data on the health of water, what’s found in the water, and contaminants in the water throughout the entire nation.

They have access to this information, which is not easy to do and it’s not easy to get access to, meaning there is no API. Being able to surface that on the site was a challenge itself. Here’s where personalization comes in, back to the earlier conversation, there are multiple ways to interact with customers for personalization and getting that data here. We are asking the customer to provide their zip code.

They don’t have to but if they do, we can tell them exactly what’s in their water and make a better recommendation on the product and filter needed to remove the contaminants that fits their needs better based on their location. We can show them reports of how the filter helps based on their water source. We’re not letting them download PDFs. We’re doing it in a user-friendly way.

If that’s an intermediate solution, I can’t wait for the advanced. What’s the name of the brand?

It’s Clearly Filtered.

What I gathered from that terrific example is a unique blend of a content marketing strategy with personalization to drive a unique user experience to help them. It’s like, “We’re going to personalize the content and then we’re going to deliver a fantastic merchandising experience for the customer based on those preferences.” They’re going to see unique information to their location, their zip code, as well as the perfect products that match the challenges they’re having. That is fantastic. It’s a strategic place. How was it devised? Was it led by the brand? Did you brainstorm that use case? How did you come up with that solution because it is really interesting?

We, at Bear, are structured and wired. What is being asked of eCommerce and what’s being asked of brands and its consumers is a mutually beneficial relationship between consumer and brand. This continues to evolve and become more deeper every generation, maybe starting with Millennials. That is what the shopper wants. They want to care about the products they’re buying. They want to know more about the story of the brand and the value propositions and they want to share that with others. They want to be proud of the things that they’re buying.

It’s all in creating a mutually beneficial relationship between customer and merchant. This brand came to market with the genuine desire and success at creating a better filter and so that’s a message everybody would be receptive to. It’s a matter of how to get the customer to hear that message from this new brand, when they were new, they’ve been around for a while now, and believe in it in a way that sticks with them. That’s where the heartbeat of this is and where personalization can always play a role as you’re trying to solve the goal of creating a mutually beneficial relationship.

For this particular case, the merchant already had this feature on their site but it was cumbersome to use and it was hard to find and the recommendations weren’t part of a natural purchase journey. We took the concept and said, “This has a lot of power behind it. It’s being underserved.” When we went to redesign the entire site, which was not needed for this particular portion but overall, that was the goal that we were working with the client, we thought about this personalization feature with zip code and product recommendation.

We thought about it from the top down. It wasn’t an afterthought. It wasn’t a bolt-on feature. It was deeply considered through the entire customer journey as well as an overall vision of how to talk about the benefits of this product in a consumer-friendly way that is scientific. All of that came together to create a good user experience.

We are great thought leaders, we are great strategists, and we do this in a lot of different verticals and industry but what makes us really successful along with that is collaboration with our clients, and that’s the best part. Our secret weapon is working with the clients and digging in and then looking at the data and seeing how customers are using the site and then making strategic recommendations.

Oftentimes, you never know if you’re sitting on a gold mine until you join heads together to see the data and that’s what these clients did with yourselves. I’ve gone through the process and it took about ten seconds to go through that entire survey. It’s an initial survey, which is Get Started on their website, and I put in a New York postcode and it gave me the twelve contaminants there.

They need to make sure that the data is always accurate and then they sell based on data. There’s a significant segment of us humans. I was reading this research paper that said that for all human beings, there are 4 or 5 categories of what drives us. Some of us are very emotive so we need a story. Some are based on pure data. I can’t quite remember the other three parts. It’s interesting with regards to digging into what they already have and using it to drive a personalization strategy.

That’s the moat right there. You’ve used the moats to drive better merchandising on their sites. Site load time are quite good, under the two seconds. Moving on to the more advanced applications of personalization, what use cases you have? They don’t need to be enterprise examples, they could be startups that have grown quite rapidly and you think personalization is driven or has been part of their rapid growth.

I have many more examples and intermediates. Your reaction was you can’t wait to hear about advanced if that’s intermediate. On the scale, for me, advanced is using potentially a third-party CMS content management system plugged in with real-time data analytics. It’s something Ninetailed and refactoring all of the content on your website based on any type of information that’s being delivered to the browser from the visitor. That’s as advanced as it gets. The concept there is all of your content on your site, homepage, PLP, PDP, and blogs are all driven from a third-party CMS that has alternative images and content based on customer data.

From a CDP. 

Also data warehouse. Ninetailed can, in real-time, build segments within your CMS pulling in data from whatever data sources you have. As an example, Contentful and Sanity are great CMS platforms that work with those technology stack. What you’re doing is a little more into the camp of The North Face example from the beginning of the podcast but carrying it much further, not just the homepage hero but every part of the site. Now, if you’re in Miami, you’re going to prioritize weather dependent and geographical dependent information. You would suppress any type of pants, rain gear, Gore-Tex, insulation, and down jackets. You’re going to deprioritize through the entire customer journey in the navigation.

You have to be careful because there’ll be some Miami customers who want to go to Seattle or want to go to Antarctica and they’re in The North Face because they know North Face as the all-weather brand that stocks those products so they can pack and travel. It’s nuance, right? 

It is very nuanced. An IP alone is becoming difficult to rely on just by itself. The core of personalization is taking what you know and doing something with that information and how far you take that. Of course, in all of this, you have to proactively monitor the data and see if it’s rendering success. The concept is, “If we lose a couple of customers coming from Miami that want to go to Antarctica or we’re providing friction for those, is that worth the greater good that’s showing bathing suits, shorts, and short sleeves? Is that worth the conversion we’re having over there?” It becomes mathematical. It’s yes or no.

What you’re incinerating here is, in the front-end, a dynamic head, be like a headless set up so a front-end or a CMS that’s works off the back of data so it’s manipulatable. It’s not just static where you publish, it’s very dynamic. You can say, “Based on these data variables, we’re pushing from the data warehouse based on this display. On the back end, you have Ninetailed, which which could be powered by the data warehouse or a segment.com or CDP, which is a customer data platform.” Is that the rough architecture there? 

Exactly.

You’d need to be dealing with huge data sets to make this worthwhile.

You have to have the traffic too. You have to have enough traffic for this to be statistically significant.

Traffic on the one hand and also even the variety on your site. Even if it’s a simplistic website or a limited merchandise, data would dictate what you do if you need it. I prefer the intermediate example. We have a bit of time for one more example of intermediate use of personalization. if you have an example, please share. 

There are several. We do a lot of work with meal delivery. The lesson with personalization is as you’re starting to seek customer data, you can’t ask too many questions. You made a great point of the time it took you to get through the Clearly Filtered zip code quiz. You can’t have too many questions because customers get paralysis by analysis. There can be too much friction in trying to get that data. It also has to be natural.

If a customer ever says, “Why do you need that? Why do you need that information?” Often, brands will ask for date of birth unnecessarily. They want to send you a simple email for your birthday and people don’t want to give it and they don’t care. It has to be a good reason. Meal delivery brands have a natural place of asking food preferences and allergies, especially nut allergies, tree nut allergies, lactose allergies, or a dislike for fish.

There’s a great way to ask for that information that’s imperative to the function of the business and customer happiness. It also gives you the edge to market against that. You can start promoting more foods that are within their dislikes and that’s a way to start segmenting your email list as you release new food products. I’d say the bigger use case and what we’ve seen a lot of is around in the baby space so infants, toddlers, and children.

Cerebelly is a great brand that does this extremely well where they provide nutritional pouches for children. They formulize the pouches based on different phases of growth. They recommend consuming, for example, more carrots during the developmental phase of eye growth. The more you electively interact with the brand, the more it can intelligently recommend the products to your children, Even more than one child, you can start creating profiles for multiple children.

Of course, age plays a role in that. It can be as simple as asking for age as opposed to date of birth. That’s a little bit more approachable. They are successful at asking different questions around milestones such as, “Is your baby starting to crawl? Is your baby walking? Is your baby talking?” These are natural questions under the overall umbrella of trying to provide a better product to your children and it’s all elective.

I understand some people are protective of their children’s information and that’s perfectly fine, nothing wrong with that. It’s about finding that middle ground of adding benefit. Once you start collecting that information, if we know the child is 2, we’re going to start surfacing content that’s more relevant to parents with 2-year-olds.

We’re going to stop surfacing content around crawling and walking, assuming that that’s already happened, and start talking about the next phase of life, which is interacting and socializing with other children and being able to share. You can relate this beyond the product but into content like educational information and blogs and articles that help educate parents.

If I was to look at this from a data level perspective, if the readers were to look at the business from three perspectives, one is customers, what pieces of data can I get from customers? That’s a big question. The other is their content, “How can we categorize our content?” They should start building a content repository that’s tagged and somehow linked to some of those tags they’ve established with their customers.

The final piece, this is from what I gathered, is their products or merchandise, “How can we link our merchandise categories and product types with content as well as customers?” Everything is personalized at those three levels and the experience that you deliver off the back of it is a great merchandising experience that’s seamless with a content marketing strategy and that’s also quite relevant to what each and every customer is seeing because all of that tagging has been done effectively all around the board.  

Personalization is rooted in data and data is only as powerful as you have it stored and can access. You need to be able to action that data. It needs to be in a place that you can action. What’s great about Shopify and its ecosystem is the care in which these platforms and technologies integrate with one another.

Rebuy, for example. It gets intimidating quickly but the way to make this approachable is to start small. Rebuy, for example, will do the data work for you and then it pushes that data into Klaviyo. It integrates with Klaviyo. At that point, you can start actioning against it directly within Klaviyo and it’s less of a headache of, “Do I need a data warehouse and a data silo?” You can start small and start with these different systems that work within Shopify’s ecosystem and integrate with one another.

Thomas, thank you so much for sharing your insights on this part about personalization. I learned a thing or two. Before I let you go, there’s a final question I have, which is more around platforming. From what I could see at the Bear Group, you guys support Adobe Commerce, which is previously Magento, Shopify, and Drupal.

The focus of this conversation has been on Shopify merchants because most of you reading this podcast run Shopify, I’m assuming, and ome of you also run BigCommerce from the surveys I’ve carried out. What is your opinion about Adobe Commerce from a personalization standpoint as well as an adoption standpoint? Where is it in the landscape? Has Shopify eaten its launch or is there space for everybody to coexist? 

It’s a topic for the past ten years and probably the next. Yes, we are believers that there’s a large enough market for all players. Most certainly, Adobe is extremely relevant, reliable, absolutely applicable, and recommended for certain merchants. We start looking at it based on technical needs, especially from a back-end point of view in relation to unique ERP systems and other integrations for running back-of-house business for merchants.

Often, the leap from a physical store to eComm have a sophisticated and longstanding technical needs and back-end needs. Often, Adobe is a great spot for that. Another historically great place for Adobe is B2B. They’ve carved-out a huge place in the market for B2B. Shopify has released some great B2B features, even as recent as 2023. They’re getting rid of their wholesale channel and replacing it with a new feature they’re calling Shopify B2B. It’s young but there’s a lot of promise with it and we’re excited about that. Still, for the time being, Adobe is a great fit for B2B.

Thomas, thank you for coming on the 2X ecommerce podcast. I thoroughly enjoyed the conversation around our own personalization. For those who want to find out more about Bear Group, it’s BearGroup.com. Are you active on any other social media platforms? 

Pretty much LinkedIn is the best resource.

Thank you very much for coming on the 2X eCommerce podcast.

Thanks so much for having me. I had a great time. I appreciate it.

Cheers. 

Take care.