Steve Lok lives for getting his hands dirty with transformative technology across industries with 22 years of experience evolving from ISP to DSP and most acronyms in between. He is currently a Head of Martech & Ops at The Economist, where he has been leading a venerated content brand into a new era of growth and recognition with transformative techops strategies in subscription marketing - though his team affectionately refer to him as “The Plumber”.
Steve has received numerous awards for his enablement of smart use of data in marketing acquisition and in helping his team execute on their visions of the future. Steve started his career as a web developer in the 90s with his own healthcare IT startup and then led agile project programs before accepting his current role in The Economist’s global circulation team. Applying those learned principles into marketing led The Economist team into taking home a Cannes Lions, DMA Grand Prix, and last week the IPA award for smartest data and technology use in The Economist’s recent Brand Response strategy and to the highest single-year increase in circulation revenue in a decade.
Following his presentation at the Big Data & Analytics for Marketing Summit last year, we sat down with Steve to discuss how the Economist uses data to influence its publishing strategy.
How did you kind of get started in your career what got you interested in data analytics?
It's a bit of a long story but, you know, to cut it short I actually started in technology directly. I was developing websites when I was 14 or so, back in 1994 when people didn't really know what the web was. Obviously, we've taken a whole bunch of different turns - very interesting ones in looking at today's actual enterprise and what the web is today, what data is today, what even analytics are today. But I now work in marketing. I'm the first technology person that's been pulled into the micromarketing organization at the company - that's been an interesting challenge, but also has created a lot of opportunity, because of the fact that there is a real core technology person inside.
The digital publishing aspect of the Economist, has that changed since you've been there?
Yeah absolutely, I think that we still produce content in essentially the same way - at least and you know at the Economist there's a real division between the editorial side and the commercial side of the business. And so, obviously, I work on the commercial side. We get content from, you know, the editorial side and they maintain their independence as to what they write about. I think what we've done commercially is we've had an opportunity to do more interesting things, as technology progresses and allows us to do that.
These days, we’re actually processing all of our content through AI and data science-driven types of ‘machinery’. This helps us to derive a level of context around everything that we produce to be able to match a particular piece of content to a particular person.
So I guess people expect more personalized content. So, if they subscribe to the economist they want articles that they’re interested in. Do you think that can get a bit spooky?
I think that there’s great thought around that, we think about it a lot. You know, can we end up being too smart about this necessarily? Personally, I'm in a very firm camp of loving it, but that requires that the predictive nature of what we're doing inside of this area now to be good, if not great. When it's great I don't really notice it, because my emotional self wants to engage with something that's put in front of it because it's it just knows me so well - maybe that's piece of advertising, maybe that is personalization or customization that happens on a website.
So the execution, and the quality of the execution, is key for whether a person thinks that something is a little bit weird or off. I think human beings, we react emotionally first, and if it's something that drives us to a positive emotional reaction then I don't think we'll notice it that much, or feel like it being a kind of, you know, Big Brother, SkyNet.
Have you seen more people get used to the idea of sharing their data and getting this personalized stuff?
Yeah, I think it's really interesting how we don't even know what is being personalized these days, and what isn't. I think again as human beings we have a really good sense to go after the things that we’re interested in, and ignore the things that were not interested in. I think we tend to get annoyed when we were presented with things that don't align with us in the right way, so I think that there's maybe a fine line between what is perceived as strange and what is perceived as useful. But I think there is a level of acceptance that happens that we often talk about in advertising.
People hate ads that aren't relevant to them, because then you notice because it's not about you. If we look at advertising that doesn't appeal to us in some way, that's when we find that to be irritating in some way. Again, personally, I love well-targeted types of personalized advertising or marketing, because it's providing a level of value to me, and that kind of value exchange is something that I find to be useful in my life.
The Economist's circulation as of June 2017
You mentioned earlier that you were the first technology person. Has it become more of a data-driven culture at the Economist across all departments?
I think that the generalized answer to that is yes. It's happening in an interesting way at the Economist, but I think that the important thing is that it is happening. Editorial has a data scientist now, marketing has data scientists, advertising - every nook and cranny of the organization that is responsible for the generation or capture of information has data scientists now. That’s because we all understand, based on our specific context, that what's important to you may not be important to me, and that exists inside of a company culture as well.
What we look at, and what we ask our data scientists on the marketing side to look at, is not necessarily what somebody else in the on the editorial side really cares about. But they may care about something that is specific around what kind of content people are consuming and why - it may not influence them, or it's really up to them as to how it influences them. What we're seeing is that we have multiple kinds of data science or data-driven pieces and towers that are kind of being built across the business. So that, from a conceptual point of view, is really great that that's happening. I'm not necessarily a person who thinks that function needs to be centralized necessarily in any way, because we do have different uses.
What sort of metrics do you think are important within publishing? I mean in terms of things like CAM, that pays authors per click, it seems a little bit simplistic. Is it important to look a lot deeper than that in the industry to gain an idea of engagement?
I'm just thinking about sort of answering it from a little bit of a lateral perspective. As a periodical, we package our content, for the most part. When you subscribe to The Economist, you get a weekly analysis regardless of the medium that you consume that content in. I don’t want to say there's a theme to the week but certainly there are leaders in that week as to what we're talking about. There’s always been a lot of discussion as to whether we should break that out and look at different models, around either talking about payment or getting people to engage. Do we break the content up into smaller bits for people because today's attention span is ‘goldfish-sized’? It’s a constant question that happens, and I think that the answer changes whoever you talk to, but it also changes depending on what models exist out there for monetizing content.
We don't operate in the same kind of format that somebody like Quartz does. We don't operate in the same format that some of these pay-per-click or pay-per-article kind of models go after, but maybe we should look at some of that. Maybe there's something to be gained from those types of models that we can learn from to evolve our business as consumption behaviors change. We want to be customer-centric around the way that people want to consume content and then honor that. I'm sure, on the other end of things, there are things to learn from the survival and success of an old, higher-end brand in publishing, in that we do just push these kinds of additions every week out there. And, if we can show that it's successful and there are ways to do this in a good way that is useful for everyone, and going to meet everyone's KPI from the company to the customer, then maybe we can learn to mix some of these things together.