Many books about digital marketing are short-term and transaction-focused. They look for immediate ROI. In his book Converted: The Data-Driven Way to Win Customers’ Hearts, Neil Hoyne argues that such short-term thinking is wrong. Hoyne has written it in his personal capacity; as such, it reflects his own opinions and independent research. He advocates building long-term relationships with customers. “Long-term thinking is not only a better and more successful way to approach customer relationships, it’s also more profitable, and the data supports it,” Hoyne says. A shorter version of this conversation was published in May in AI Business. What appears below is a more comprehensive version of Hoyne’s conversation with the Coller Venture Review.
Short-term thinking is wrong, instead, you want to build relationships with your customers. Long-term thinking is not only a better and more successful way to approach customer relationships, it’s also more profitable, and the data supports it
Coller Venture Review —
What inspired you to write Converted? Did you see any limitations in existing books about digital marketing? What gap were you trying to fill?
Neil Hoyne —
The first limitation in existing books about digital marketing is simply that they are short-term and transaction-focused. They look for immediate ROI. This is the single argument that carries forward in my entire book, which is that short-term thinking is wrong, that you want to build relationships with your customers. Long-term thinking is not only a better and more successful way to approach customer relationships, it’s also more profitable, and the data supports it.
My second goal, which focused on the medium of a book as a whole, was to reach more people. In my role, I talk to the 16,000 largest advertisers, but generally these are companies that are spending at least $5 million to $6 million a year with Google. That’s a lot of money. But these lessons and ideas apply to companies that are much smaller. My objective is to offer those cutting-edge best practices in a simple and intuitive way to the small businesses that may not have the resources or the marketing spend that larger organizations do, but they are still looking at opportunities for how they can grow and compete.
My third goal was just the accessibility of data to everyday business people, to everyday stakeholders. This was my attempt to say, “I want to bring everybody into the conversation to show that data can not only be interesting, it can be accessible. And so my third challenge was, could I write something that makes data interesting, intuitive, and accessible for more audiences. And those are the three goals.”
Many brands seem to believe that the solution to their problems lies in a four letter word, “data.” But having more data, as you point out in your book, does not necessarily mean that it will be used well. Do you agree?
As it relates to capturing data, I think that companies put a disproportionate amount of emphasis on collecting data, and the systems that support that collection. Companies seem to believe that data is the new oil, and therefore the more they can capture it, eventually they will be able to convert it into some type of value. That leads to a dramatic underinvestment in the second part of the equation, which is naturally, what do we do with all this data? Most leaders are not necessarily taking a step back and saying, what is it that you’re hoping to get out of that data? What are those questions you’re trying to answer? When we’re talking about this within the context of the book is how you apply that data, and the questions that are important to ask.
Companies seem to believe that data is the new oil, and therefore the more they can capture it, eventually they will be able to convert it into some type of value
How can companies learn to use the data better to convert casual browsers into long-term customers?
Two things are worth mentioning. First, the idea of converting casual browsers – non-loyal or low-value customers into high-value customers – may be a difficult premise. In the real world, it would be similar to your meeting a friend who says, “I met this person, they’re terrible for me, but I can change them once they see how good I am.”
It’s the same premise with many companies. It’s a lot easier just to acquire great customers overall instead of trying to change or convert those customers. What I’m talking about here is to use the data that you’re capturing not necessarily to convert casual customers into great customers and great relationships, but to use that data to go back to the beginning to say, how do I find these people that are great for my business? How do I find people that already have that natural fit? Where else can I go? What else can I find, and what do those people look like?
Don’t use the data that you capture to convert casual customers into great customers, but go back to the beginning and find people that are great for business
The second component, though, is to say not only are you using data to understand where those high-value customers are, and how to acquire more of them, you’re also saying, how do I use that data to look inside myself and to become a better company, to offer things more for those high-value customers so that the fit and the attractiveness is already there.
As more brands invest in AI and analytics solutions, how is this changing the kind of insights they can capture about their customers?
Now, this is interesting because in my experience a lot of companies that are pursuing AI are being driven more with the attitude, “We need this tool,” rather than, “We have this business question.” They’re so worried about falling behind in the race for AI and AI competency that they’re not necessarily sure what they’re supposed to pursue. It seems like an afterthought.
Then you go back to those people and you ask, “What are you trying to solve with AI?” And they give you blank stares. It’s like, “Well no, no, no, we’re investing, it’s just a capability our business needs” without really thinking of what the end goal is. Now, for those companies that are using it, and they’re often using it carefully, they’re going back to really what the essence of analytics and AI is supposed to be. So they already have their question in their mind, what are our high-value customers doing that we’re missing?
The other component of AI which is fascinating is that they’re using it to build a scalable process. The only way to do that in a scalable way, given all the data and signals we’re collecting, is artificial intelligence, machine learning, which is really to take over that human component of analyzing the data and saying, really what’s going on here and can we do this in a repeatable, scalable way.
Can you offer a few examples of brands that are doing this well? What can small companies with limited marketing budgets learn from larger organizations in this regard? Are there any experiments that they can try?
Within large organizations, the general understanding that they have more data, more capabilities, more systems, larger marketing budgets. But what we miss is that they also have large, overwhelming bureaucracies where prioritizing a question, getting alignment and action on that question is often difficult.
Smaller, more entrepreneurial startups, well they’re smaller, they’re more nimble, everyone is kind of aligned – plus they don’t have those strict bureaucratic silos. But they all have that same objective, which means we need to succeed and grow as a company, otherwise somebody is going to be pulling the plug. What both recognize is that it’s really the strategy by which you want to apply that data that allows you to compete.
Now, are there any experiments that I can try? What I generally advise is for companies to try calculating lifetime value. For a small, medium sized business it’s not to say “I don’t have enough data.” Maybe your several thousand customers are plenty. Your two years of data is just fine. For mobile gaming it can be as little as a month’s worth of data.
How can brands become better at using AI for mass personalization at scale?
There are a few points here. When we really talk about personalization we have to broaden our lens a little bit to say, “what do we know about these customers and what are their actual needs? How do we deliver them”?
Here’s just one small example, and I like to use travel companies, for instance. Oftentimes they’ll underinvest in that personalization to improve the customer experience, because in their mind it doesn’t immediately lead to anymore incremental reservations. But if they took a step back and they said, let’s look at the lifetime value, for instance, of customers that we service well, that we provide ancillary benefits to, they’d probably see that those customers stick around longer and probably spend a little bit more money.
Another example in the travel space is simply — this is a case study one time with one of those airline apps. What they found was that the lifetime value of people with their app versus just web users was roughly the same. Because their mobile app, even though it had a permanent place on their device, wasn’t adding anything incremental to the experience. All they were trying to do was say, I need your money now.
And when they started adding additional features to say, let’s help you manage your reward points, let’s help you manage your way around the airport, to manage your travel reservations, they weren’t finding immediate bookings. In fact, they found that it actually distracted some people from booking again. At the same time though, they found that the customers were happier with the airline, that in the future they booked more, and that their lifetime value went up.
So when we talk about how brands can become better using AI for mass personalization, again, it comes to what’s the objective? Is the objective for using AI and personalization to drive that short-term conversion, to drive the acceptance of your immediate proposal, or is it to build a longer-term relationship? Once we go into the latter question, then we see how much more room we have to work.
Which industries are doing particularly well at using AI and analytics to build long-term relationships with their customers? What are they doing differently than industries that are doing less well?
You know, this is probably taking the easy way out. They may argue with it, but the industries that are doing particularly well are the industries that are entirely dependent on long-term relationships from the start. So if you think about subscription-based businesses, it would be silly for Netflix or HBO to say, we’re only looking at short-term revenue. For a telco company to say we’re only looking at your first monthly bill, that would be ridiculous for them. They have to forecast out that relationship, they have to look at how that relationship will change over time.
The laggards in the group, just for completeness, are generally ones where there’s a lot of time between transactions. So large-scale purchases, when we’re thinking about things like automobiles, that’s 10 years in between purchases. The idea of lifetime value isn’t really there. Long-term relationships are nice, but there’s no way to directly measure its impact so it becomes a little bit harder for them.
And also in that category, consumer packaged goods or CPG. In CPG, it’s not because these long-term customers don’t exist, it’s because they just don’t have the data for it. They’re getting aggregate data from their retail channels to say, look, I’m not going to tell you Neil bought x numbers of your products, I’m just going to tell you how many thousands of units we sold. And in that case, the only real long-term relationships they’re able to build are with the retail channel partners who are selling their products through to the end consumer. So you do see them starting to push a little bit into direct-to-consumer selling. It’s not because I think they want to compete in those areas. But they just want a little bit of that data, because they feel just by understanding a small glimpse of their customers they can make better long-term relationship decisions.
You’ve worked with thousands of companies on data-driven marketing projects involving AI and analytics. What are the most common mistakes you have seen them make? What advice do you generally offer to avoid those mistakes?
Number one, the challenge is I don’t think enough people know what the business objectives are that they’re trying to solve. As I mentioned before, they start with technology. “We need to be in AI, we need to be in machine learning, we need to have a bigger data presence.” They do not necessarily know what business questions they’re trying to answer.
Now, there are other mistakes. For instance, companies are unsure as to what level of transparency they need. Do they need to understand how these models work, how this data works? Do they have biased data? You know, we talk about these transformational technologies like AI, one of the things we neglect to discuss is in transformative processes and strategies, there’s often a significant reallocation of company resources. Data is never binary. That wiggle room sometimes is just enough to create organizational deadlock.
Those are just some of the mistakes that they make, and you’ll notice that these are not on the technical side. It’s not about how to build the models or the technology that should be used or who should own the technology. It’s simply decision making from an organization to say, “how are you going to handle something new that may not necessarily be transparent but may be disruptive.” And as a leader in your organization, how do you have those conversations and guide people through that process.
What can brands do to establish relationships of trust with their consumers and increase their comfort level in sharing their data? What questions should they be asking to identify their most profitable customers?
So let’s work with the first part of this question. Establishing trust with customers is an important part of collecting any type of data, and generally what comes out of it are three things. First, consumers are looking for transparency, e.g., what are you doing with my data, what are you capturing? They are also looking for control, e.g., “can I remove my consent to that data, limit, or correct the data you’ve collected?” And finally, they want to understand the value, e.g., “how is the information that I’ve given you going to somehow benefit me?” Or are you simply going to use data to know which customers you can charge more for, or follow me around the internet with ads?”
The goal of these efforts in building trust with your customers should simply be better than what your competitors are doing. Remember, this is an auction environment. You just need to have slightly more data, slightly more trust, slightly better understanding of your customers than your competitors do. That should allow you to make better decisions and build better models. That’s just one way to look at it. You don’t have to be perfect, just better. Right now the bar seems to be fairly low.
What are the main takeaways of the book for entrepreneurs, venture capitalists, and private equity investors?
Well, I would hope for entrepreneurs, much like marketers, that they take away a sense of confidence that these techniques can be a part of their portfolio, even with their current capabilities, even with their current set of data. And that they start to recognize unique advantages that they have in developing their team and developing their processes.
A lot of the third section of the book, self-improvement, talks about incremental change, experimentation, and making sure that you can actually act on the data. And my hope would be that entrepreneurs who read this early on will embed that as part of their culture and their processes, because they’ll certainly have an easier time than a large company that’s trying to change theirs.
Now, for the venture capitalists and the private equity investors, this would be related to how you really judge the performance and the effectiveness of the businesses that you’re investing in. Who are your most valuable customers? Who are your least valuable customers?
There is also this emerging area where instead of looking at the individual customers, we take all the customers of the business, all of their lifetime values collectively, and add them up. We get a number called customer equity, which is how much your entire customer base is worth. This is the most valuable asset of your business – and we’re able to put a value on it to say “this is how much they’re going to spend.” It allows you to better understand the valuation of the firm. Directionally it allows them to measure the full impact of what the teams are doing instead of just setting arbitrary short term metrics.
If the CEOs of brands in the U.S., Europe, or Asia were to ask you where they should start to apply the lessons of your book and make smarter use of AI and analytics to build customer lifetime value, what would I recommend they do?
Well, step number one is just to acknowledge that you probably don’t need to buy more big data systems. You have all the data you have. Step two is to calculate lifetime value. The techniques are already available and proven, and you likely have data scientists already that can perfect the models if there are weaknesses. The third step though is actually making sure the right metric is front and center alongside all of your existing KPIs and metrics. And simply having that understanding, having that metric, even if you’re not incentivizing people on it, even if you don’t have a specific plan of action, encourages people to talk, to discuss, to understand how they change course. And then it comes full circle where the company starts seeking out more information.
What new areas of research are you working on these days to build upon the foundation of the book?
Well, there’s a lot of different things in the book that I’m curious about. The application has infinite variations to go into – what are new techniques for acquiring customers, developing customers, retaining customers? And all of those are fascinating paths that could be their own guidebooks in themselves.
I am incredibly curious about how companies develop these assets and these people. I said early on that one of the things that I don’t expect companies to do is invest more in software, because I don’t think that’s a solution. I think the solution is to invest more in people. But it’s not simply hiring, it’s being able to train and develop, and to understand those motivations, those incentives, those processes, how you build those functional teams. I think that’s the next area of data science, because everything else by comparison is limited.