The Data-Driven Way to Win Customers’ Hearts

0:00We almost get into this mindset for a lot of retailers where they say, I'm going to win the customer over by facts, by data. But when you start telling a story about where the product was from, how the product was made, now they become almost a character in that story and thinking about how it could fit into their life. Hey, what's up? Benjamin Gottlieb here, and this is Shopify Masters, your companion for starting and building a business. If you're listening to this show right now, there's a very good chance you either own your own business or at least you're thinking about starting one. And so inevitably, You think a lot about how to sell more of your stuff.

0:35There are so many articles out there, YouTube videos, podcasts like this one. They're full of advice. But if you ask Neil Hoyne, most of these strategies are actually short-sighted. Neil is the chief measurement strategist at Google, so he knows a thing or two about what I'm talking about. And his recipe for success, not clicks, not conversions, but building relationships with your customers through conversation. That's just some of the advice from his new book, Converted, the Data-Driven Way to Win Customers' Hearts. And guess what? He's my guest today. Neil, thank you so much for doing this.

1:11Hey, thanks for having me. Are you available for hire? I'd like to just, you know, have you walk around with me to meetings and just introduce me like that. That's fantastic. And I paid Neil to say that. So thank you so much, Neil. I'll give you$100 after the interview. No problem. Hey, my pleasure. Neil, let's start with this. If I am a marketing team, or a company for that matter, why are conversations just as important as clicks and conversions? I mean, clicks and conversions, that's where the web started. It was all about immediate gratification. Somebody goes to your website, maybe you paid money to get them there, and then they do something that you want.

1:43But we also know relationships are being built. Oftentimes, there's going to be customers that come to your website, but they're not sure if they want to buy right away. Maybe they need to ask somebody else in their family what they think about the idea. Maybe they want to go to other websites to see alternatives. Maybe they buy and then they come back again. Those relationships are just a natural part of doing business. And as you pointed out in the introduction, clicks, conversions may be a little bit short-sighted. You lose touch with those customers who may come back multiple times, contribute a lot of value, but are lost in the noise of people who are simply transactional.

2:18And in your book, you have this example of a customer who's come back multiple times. Perhaps That's an underestimation, right? This is a shopper who visits a store 262 times, if I'm not mistaken, but not visiting the physical location, but the website, right? Looking to buy a pair of heels. How do you have a conversation with somebody like that who's hopping in, checking out their virtual cart and then leaving so many times? Well, first, the requisite disclaimer. I don't understand how people buy shoes. I understand some people have hundreds of pairs of shoes. I have six. For some people, I imagine that that is a longer process.

2:57They're thinking about where they're going to wear them, if they're comfortable, how much they want to spend, do they like the brand. I look at durability and comfort. That's it. But for those people that have a longer process, maybe spending significantly more money, we might expect them to take multiple interactions. Now, if you had a physical store and you saw a customer come in, now this more than 200 times, This was over a period of 14 days. If you saw a customer come in that many times, I want you to think about in a physical environment how you might change your approach. I would think they're maybe casing the place, right?

3:29Oh, yeah. What are you doing? Who has all this time?

3:39You may say, all right, maybe not on their first time, but maybe on their fifth or sixth time, I might try to ask them different questions. What are they doing? On their 10th time, maybe I may offer them a discount or an offer that I didn't give them the first. I may recommend different products to them, or there may become a time where I just stop paying attention to them altogether. I don't know what the person is doing coming into my store. I've offered to help them. I've spent all my money on marketing. I'm going to stop and let that take its course. And that's what we're really getting at.

4:10We're getting at these people actually represent part of your customer base. Now, not everybody, of course, not everyone has this amount of time. 200 times to a website is absurd, but you start to think that these 2 % or 3 % of customers that often behave this way can take up a disproportionately large amount of your marketing budget.

4:31In this case, this particular customer clicked so many times that even though she did, to everyone's celebration, buy those shoes in the end, the company spent so much money marketing to her that they still lost money.

4:48and we went back and we looked at their data. We said, what did you learn about this customer? And they only looked at that last click and they said, well, she came to the store. She bought a product. Here's what she did right before it. They didn't know how many interactions she came because they weren't paying attention to it. They were paying attention just to that moment. She ended up buying the product. Now, really the thesis, the central part of the book is asking this next question. What do you do if this customer comes back? Was it worth it? All the time you spent building that relationship?

5:19Do you want to spend more time with her? Would you be more prone to ignoring her? Is this the type of customer you want to build your business around? Well, that's a great question. Is that the type of customer you want to build your business around? Well, that's where we get to. And so that metric we add to this, and I want to explain this, is this metric that's very common to a lot of marketers, but now we're putting it into practice. Customer lifetime value. What is it? It's saying, what do we think is going to happen with the relationship? Do we think this person's going to be the best customer we ever had?

5:46or someone we're never going to see again. And then we can act accordingly. So in the case of this woman, if she turned out to be a high lifetime value customer, someone that's likely to come in and purchase again and again and again, it was worth our time and our effort. However, if it turns out to be a customer we're never going to see again, we might rethink trying to acquire people like her in the future. Well, that's very interesting. I'm chatting here with Neil Hoyne. His book is called Converted. I've got it right here. The data-driven way to win customers' hearts. And he's also working at Google.

6:19Now, Neil, I've got to ask you about the storytelling because as a storyteller myself, I was really, really excited to see this in a marketing book. You write in your book that conversations are integral to the human experience. That makes sense. I mean, storytelling is as old as language itself, but marketing a product, making that as interesting as say a campfire story, that's not easy to do. So how do you do that? How do you make a story about a product but something that is very intrinsically human and accessible. Well, I think that's a great part about storytelling is it is uniquely human.

6:54We all know how to do it. We just condition ourselves not to do it. We almost get into this mindset for a lot of retailers where they say, look, I'm going to win the customer over by facts, by figures, by data. Data always wins. And oftentimes when we present stories from that perspective of data, thinking our features and our characteristics of our product or our price win, customers will follow your lead and they will look at those stories at a data-driven lens. They'll say, all right, I will compare your attributes against your competitor's attributes. But when you start telling a story about where the product was from, how the product was made, now they become almost a character in that story.

7:32They see themselves playing that role with that product, with those shoes, and thinking about how it could fit into their life. And sometimes their decisions are not data-driven at all.

7:47They may pay more for those products or features they didn't think were important. In fact, even when we look at the stories that companies tell, if you happen to tell a story about how your product is made and the people that contribute to it and build it, consumers rate that as being higher quality. They have a greater sense of loyalty, understanding what normally websites hide from them, which is what's actually going on beyond the page. How did this product end up at my door? those very core elements of marketing, if they're part of your story, the benefits are immeasurable. It's just, we get off too often to say, well, we'll throw the data at the customers and they'll figure it out.

8:23Customers really want that human interaction behind it. Does this speak to this concept you bring up, which I found fascinating, which is that people think of brands almost like they think of individuals. I love this company. I hate that website. They do. And it's, by the way, it's tough to tell people that because you're like, no, no, no. I don't think of Coca-Cola or Apple the same way I do as my spouse. That's true. Absolutely true. Those are in different categories. However, that doesn't mean you still can't have that love, that affinity towards a brand to do irrational things, to find yourself connected with it.

8:57And so that's what we have to appreciate is that those brands are very intimate relationships that we have with consumers. And it's that intimate relationship that for folks listening right now, most of whom either have a business or want to start one. They're thinking about this constantly, Neil. And I'm curious, you brought up this example of the customer who comes in 262 times to the website. Obviously, that's an extreme example. But when it comes to making mistakes or practices to avoid, what should a business owner do when looking at data like that or other data about their customers and making determinations about who to target and who not to?

9:34Oh boy, how long do we have? We have several days, several weeks. All the mistakes you make. By the way, you'll learn I focus a lot, especially on the book, on mistakes. I think we learn a lot more from mistakes than looking at companies where we think a CEO or entrepreneur points in a direction and they all head that way and it works out. It rarely is the case. Let's talk about mistakes and what people do wrong. Number one is, as we've already talked about, they're too short-sighted. Their metrics are what happens today. In that moment, consumers take time. They want to build that connection with a product.

10:05Number two is oftentimes when we're capturing data, when we're trying to understand customers, we don't do it with an intention of what we're going to do with that data. So most companies' table stakes is we'll capture conversion data. Here's what I need to get you that order. But we never think about what else we could do with that data. And so here's a practical example. In my inbox right now, I will have hundreds of promotional messages and email campaigns from companies I've purchased from in the past. All of them know my name. The research supports that adding somebody's first name into the subject line of an email significantly improves the likelihood that they will open that email, they will engage with that email, and decreases the likelihood they will unsubscribe.

10:49Yet in my inbox right now, no company, despite having my name, actually uses that data to build a better relationship. So companies may oftentimes collect data without thinking about how they can apply it. They just think, well, can we ask customers this? Like, can we have your credit card statement? What do you want it for? And customers are okay with sharing that data if they know the value exchange, if they know what they get back. We want to build you better products, make you better offers. I also tell people the most underutilized page for asking information. People often think they need to ask during the checkout process, which is a difficult bargain because now you're saying, you're ready to buy my product.

11:27do I really want to ask another question or two that may distract you from finishing that purchase? Of course, right. And the answer is no, don't distract them. But here's where you want to ask them these questions. The height of trust that most of the data supports that customers have with your business is after they give you money. I gave you money. I really hope your product shows up. I hope I love it as much as I do on your website. Please give it to me. And then what do we do as companies? We give them a thank you page with an order number that nobody writes down telling them that their email will have all their information, inviting them to close that window.

12:01It turns out that after they give you the money, that's at the height of trust where you can ask them those additional questions. What other companies did you consider? How valuable did you think our products were? How much money are you giving to us versus a competitor? That should be a playground for you to ask some of these questions to your customers to learn a little bit more about them and their behaviors. Just at the same time, always think for the questions you're asking, don't collect information just for the sake of collecting it. Think about how you might use it to personalize your emails, their experiences, or deliver better value to them.

12:31So Neil, I've got to ask then, in your estimation, many companies are collecting information, but they're not using it properly. You just alluded to some of the ways that they could use it a little bit better, but in your expertise, what are some ways that companies, brands that are listening, entrepreneurs that are tuned in right now might be able to use the data they're collecting a little bit more efficiently. So the starting point is not the reports. Far too often I see entrepreneurs go in and say, here's my analytics report. Let's see what's in it. And then they'll stare at it and be like, well, traffic is up in Brazil.

13:03Why is that? I don't know. What do I do with that? And you can't do anything with it. I can't do anything with it. I'm having PTSD, by the way. No, it happens. It happens. Small companies and large companies. This is where you're looking for data and you hope it tells you a question or an opportunity. And it never does. Instead, what I encourage you to do is take a step back. Before you go into that data, write down your hypotheses. What do you think you could do to better connect with your customers? What information, if you had it, would change the way that you build products, that you interact, that you service with customers?

13:34And then you get to the data.

13:39So you come up with this huge list of the hypotheses you have. And then the next question you want to ask yourself is, do we have any data that helps either prove or disprove this hypothesis. You may now find that in your analytics reports. If you don't, this may be an avenue for you to ask friends, family, colleagues, other people in marketing to say, have they seen anything that supports or disproves this hypothesis? And if you don't have any data, then you move on to the valuable third step, which is, can you test or find a way to capture that data on your own? But here, really what we're doing is we're starting with the question, what you think you should do differently, where you think there's an opportunity, and then all those reports work to support that as opposed to the other way around.

14:20Let's look through all the data and see if there's something there. And then we get frustrated when we're never sure about the question or the action we should be taking. We've been talking with Neil Hoyne. He's my guest today, chief measurement strategist at Google and also the author of Converted, the Data-Driven Way to Win Customers' Hearts. I'd like to take a moment here and thank you, yes, you, for listening to our podcast. You know, each week we bring you stories from Shopify founders and also industry experts like today about everything entrepreneurship. So if you're digging what we've got going on, make sure to give Shopify Masters a follow or subscribe to the show wherever you're listening now.

14:55And you can let us know your thoughts by leaving a review. We really appreciate you and thank you so much for listening. Neil, I would be remiss if I didn't ask you about recent advancements in artificial intelligence. After all, you serve as a senior fellow in AI at UPenn, the Wharton School. Beyond these existential questions about open AI, chat GPT, how are these advancements changing the mechanics of having conversations with your customers? I mean, there's no doubt that a lot of the things we've seen from these large language models, things like chat GPT, help on the creative side, different ways that we can tell stories, express ourselves in very unique and creative ways.

15:33And especially for entrepreneurs, that's a great gateway to say, here's how AI can be actionable to you. Where I'm a little bit more concern, though, is also that AI sometimes can be that solution in search of a problem, where more companies are saying, well, now that we have AI, how do we integrate it directly into our customer experience? How do we build chatbots? And that's why I also offer a little bit of restraint. So starting again to say, this is what AI does, I find companies are successful saying, where can it start to fill in and add immediate value in the processes and things we're doing day to day?

16:04Can it help us understand our data a little bit better or build presentations or some of these copywriting campaigns? The answer is yes, absolutely, and it's easy to get started. But when you start seeing these more heavier integrations to say we need to change our entire business or our value proposition to our customers, I urge a little bit of skepticism. As with all things AI, part of the adventure of this technology is that we're exploring what it can do and its limitations. And we're still in the early days of it. And you may have seen a couple weeks ago there was this lawsuit with an airline where— You're talking about the Avianca suit where an attorney filed a lawsuit basically using AI to find a court case, and that court case didn't exist.

16:43Yeah. I think it was like five or six court cases. And when you read it, it all makes sense. But it wasn't. It was just made up. And so now what does the judge say? The judge has said, I don't want anything from AI unless it's backed up by a human review, a human source. And that's the same way that we have to look at applying this technology within businesses, that it's very powerful. but there always needs to be that human element to say, is this message the right thing for our customers? Is this technology what our customers are really asking for? And so I say for anybody out there, you should be very curious about what this technology can do.

17:15You should play with it for yourself, but you shouldn't look at it as something that you have to immediately integrate into your business or worry about being left behind. Your business is successful because of everything you've done to this point. AI should help to accelerate or transform it, not to dramatically change it and head in a different direction. You're talking about trust here and the risk of perhaps peeling away some layers of trust that you've built up over the years with your brand, right? Anything can happen with automation. I like to use very human examples for it. Every holiday, we just passed Mother's Day here in the States, I will write a card for my wife, right?

17:54My words, not as eloquent, but I like to think that I can write. Now, if I go into chat GPT and I ask it to write a card, it will do a brilliant job, but it doesn't sound like me. And the meaning is lost. And part of what the value is, is that human connection, that imperfection, the time spent, the effort, just as we were talking about as companies build products, of sitting down there to try to come up with words. And so even though technically these AI models can give us the right answer. Is it really the right thing for building trust with people, for building that genuine connection, that intimacy?

18:28I would argue no. And that's simply about knowing the limits, what your consumers want and how you deliver value to them. And by the way, I'm not calling my wife a consumer. I'm just saying in that lens, that also applies. I hope our audience is listening because this is advice from someone who's written about customer acquisition, having genuine conversations with your customers, is an expert in data, and he is telling you right now that there is no replacement, as of now, for this genuine, imperfect at times, human interaction between yourself and your customers. But I think where maybe some of this gets lost, Neil, is it's not just you're on the phone talking to your customer.

19:08There's many ways to have conversations with your customers, right? Do you believe in this idea of coming at your customer from many different angles, right? So in my career as a journalist, if I wanted to get a hold of somebody, I'd call them, I'd hit them up on social, I'd email them, I might knock on their door. Is there a point where it's too much, too much conversation? Where do you find kind of the happy balance if you're trying to market a product to a customer? You know, I'd say in the initial stages, it's hard to find too much because you are trying to break through the noise and make sure your brand is known.

19:42Once that relationship and that conversation starts, which is really what you're talking about is the first thing, if they never hear you, if they never know you're around, you're not having a conversation. You're just messaging out there, hoping that you just capture a little bit of their mindshare. But once you have it and they're aware of your brand, they're aware of your products, a lot of the same relationship rules that guide our day-to-day conversations are surprisingly applicable to digital. And I'll give you kind of a silly example. And by the way, as I get older, people have told me that this is less applicable.

20:10So for the younger audiences, I'm sorry. But when I was dating, there was a very simple rule that said, look, if you went out on a date with someone, there was informally referred to as a three-day rule, which is you wouldn't call them or text them right away. You would give it a couple of days before you message them just to show you were interested, but not desperate. You weren't going to wait a week. Three days was appropriate. Not seem over-eager is what you said. Not seem over-eager because if the date gets finished and you immediately call them, that might decrease the chances you may look a little bit desperate.

20:42Now, are we surprised that when we look at consumer behavior to advertising, the same type of rules apply for those advertisers that immediately start messaging, emailing, serving up advertising to customers as soon as they leave the website, actually see a decrease in the interest of those customers in buying the products? If you do anything in under 24 hours, that likelihood of them actually buying that product declines significantly. Now, what's the ideal stage? around 48 to 72 hours, around three days after that interaction where they leave their shopping cart is the optimal time to increase purchase intent.