Changing the Currency in Advertising: Chartbeat Founder Tony Haile on the Future of Content Measurement
In 2014, Tony Haile was in the midst of leading a publishing revolution.
As CEO of Chartbeat, a web analytics company, he pushed the industry past the pageview and toward metrics that valued quality over quantity. His Time article “What You Think You Know About the Web Is Wrong” was a call to arms: “We confuse what people have clicked on for what they’ve read. We mistake sharing for reading,” he wrote. “We race towards new trends like native advertising without fixing what was wrong with the old ones and make the same mistakes all over again.”
Gradually, publishers started to listen. Many turned to Chartbeat for the tools and insights that could let them focus on meaningful engagement instead of clickbait. There was still work to do, but Chartbeat had helped correct course. Then, after seven years with the company, Haile suddenly resigned last February to try something new.
That turned out to be Scroll, a subscription service that plans to offer ad-free access to journalism from a variety of sites. As Haile continues to develop Scroll with over $3 million in funding, he spoke to me on the phone about Chartbeat’s evolution, going up against Google, the need for a frictionless free press, and what he hopes to accomplish in the next decade.
What brought you to Chartbeat in 2009?
My background prior to then is somewhat bizarre. I spent a good amount of my twenties leading polar expeditions and competing around the world in yacht races. Basically, most of my twenties were focused on doing stupid things in fairly risky environments. I had great stories for dinner parties.
Then I was in New York looking for money for an exhibition, when someone asked my opinion on a business plan. Because I was at a loose end, I gave him six pages of notes about what they should be doing instead of what they were doing. This got passed around to a bunch of people in Silicon Valley. Then they said, “Do you want to help us run this thing?” That was a Hawaiian-based startup, which had its perks, but it very swiftly became a disaster.
Part of that was my inexperience. I launched a coup against the CEO and then was fired. So I found myself in New York, again. I met up with the Betaworks co-founders, who introduced me to this incredibly creative guy called Billy Chasen, who built this early prototype of Chartbeat. We partnered up together. A little while later, Billy, whose heart is really in the consumer, gifted me Chartbeat for my own.
Once you took over, did Chartbeat evolve in a major way that you didn’t expect at first?
If you looked at the analytics space at the time, there were already a few players. There were your big players like Omniture, which was acquired by Adobe. And there was Google Analytics, which was everywhere. Everyone already had it, it was free, and it was pretty damn good. The challenge we faced was, holy shit, how does one even survive in this space? How does one make money?
I went to a friend of mine who’s a very, very wise man and laid out the situation. I said, “I’ve got this thing that does web analytics, but it’s new to the space and Google’s there. Am I actually fucked?”
The challenge we faced was, holy shit, how does one even survive in this space? How does one make money?
He said, “Yes, you’re probably fucked. However, you can’t succeed by doing things even close to the way Google does it. You can’t do something incrementally better than Google, because Google will just get better. They have the best engineers dedicated to that. So why not take the opposite of everything that Google does and do that? That might make the world’s worst analytics platform, but at least it will be different.”
We decided to do that. Google was focused on historical data, so we’re surely going to focus on real time. Where Google was focused on the pathway between pageviews, we were focused on what happens within the page. Where Google allowed you to pivot and dive deep and have all this complexity, originally we just kept it to one single page. We didn’t let anyone export data. We didn’t know what they would do with it, and we wanted them to come to complain to us about it so we could understand what they were doing. It was all a complete bet.
In 2014, you guys were a huge name, really pushing for that switch from vanity metrics to engagement. Based on what you’ve seen since then, has the publishing world embraced engagement metrics and been more sophisticated?
The shift has been both faster and slower. In the past, engagement was a very vague term. There wasn’t consistency around what we were trying to measure. One of the things that’s been good to see is the general sense that there’s this attention economy. Attention is accurately measured time spent. There is now a common language.
Chartbeat was one of the first to look at the problems in the market. The way time spent had traditionally been measured was so inaccurate as to be pointless. It was measuring between page slugs regardless of whether the tab was left open in the background or whether the person had gotten up to get coffee. Chartbeat built a new class of metric, which was engaged time. We looked for human-computer interaction and really tried to understand an accurate sense of someone’s attention better than just a computer measuring a tab being open. I talked with Contently early on, and I know you guys have been doing good stuff around that as well.
The most interesting thing beginning a year or two ago was that you started to see it affect the best-of lists. The New York Times and GQ and whomever else—instead of ranking the best-of lists by pageviews, which they’d been doing in the past, they were doing best-of lists by total amount of time. This showed a very different and better view of important content.
One of the slower things has been the movement of these metrics onto the economic side, and that’s to be expected. But there’s a tremendous amount of inertia. You’ve got great people pushing it forward. You have the Financial Times and the Economist pushing through new metrics. That’s very heartening to see, but it’s going to be a very slow process.
Do you have an educated guess on when that process may become a little more concrete?
I was reading the Zuckerberg manifesto the other day. He said people overestimate what they can achieve in three years and underestimate what they can achieve in ten. I think that’s a good phrase to think about engagement metrics from an advertising perspective.
I think most people would agree with you that sophistication with attention has gotten much better. But you also have these pervasive content recommendation modules that aren’t promoting the best quality content. So how does distribution fit into this discussion?
I think everyone develops to what they measure and what is told to be valuable. If you are a recirculation company, you optimize for clicks, because a click is what makes me money—not necessarily what happens after the click. And that lends itself to the most clickbait stuff we can possibly do to make money. It’s still the case in much of the world.
Unfortunately, right now, the incentives are toward what people’s index fingers do more than what their brains do.
It would be interesting to see if a recirculation company were to say that payouts are going to be based upon the amount of engagement that happens on the page after someone clicks through. In the way that those companies currently work, I think it would be very difficult. But maybe there are some unique possibilities around that in the future.
If you can change the economic incentives, you can change the rest. Unfortunately, right now, the incentives are toward what people’s index fingers do more than what their brains do.
Why did you decide to leave Chartbeat?
It was an incredibly difficult decision. It was my baby and I adore the people who work there. But there were two things for me: One was I put together a very good executive team. As I was able to delegate more and more to them, I found myself increasingly feeling like I was adding some value but I wasn’t doing as much anymore. I was able to sit back and think big things and get up on stage. The company was running itself quite nicely without too much input from me.
Then I also had this urge. The things I enjoy doing most is when former employees or friends who have startups come in and we do a whiteboard session. What’s the strategy for this? How do we acquire customers? The big questions, which haven’t been fully formed. I missed those startup days when you’re not sure of your hypothesis and everything is to play for. I missed a little bit of the danger. I wanted to get back to that.
How does Scroll connect to everything you’ve done in the past?
There’s been a guiding mission to what I’ve done over the last decade or so—a passionate belief in the importance of a free press effectively able to be sustainable, thrive, and stand up to corrupt politicians. The job of journalism has never been more vital. At the same time, the ability of the industry to fund that journalism has never been more under stress.
At Chartbeat, one of the things I was trying to do was change the currency in advertising so the quality of the content would affect the value of the page. With Scroll, I’m looking at it from a more direct approach. The mission of Scroll, quite frankly, is frictionless access to a thriving free press. The problem we’re trying to solve is how do you fund journalism by making the experience of the people who consume it better, not worse?
Traditionally, one of the challenges you had was in order to make money, you had to sacrifice user experience. I’m interested to see if there’s a way we can have our cake and eat it—have a great user experience and have revenue coming to journalists, where they’re trying to optimize for how engaging the content is, how loyal the audience is, and so forth.
What’s the progress been like solving those problems? Are we going to get to have the cake and eat it too?
I hope so. We’ve been lucky in that we’ve had the support of not just great VCs but also a number of publishing companies that gave us financial and other support. We took investment from The New York Times, Axel Springer, NewsCorp, and a bunch of other publishers that have been helping, testing things for us, telling us where we’re crazy. There’s still a bunch of work to do before we come out and tell people everything we’re up to in a broad way.
There have been similar services like Blendle that have gotten good press but not necessarily taken off the way some people thought. What is Scroll going to do that will work compared to what hasn’t worked for other companies before?
Part of that’s the secret sauce. By the way, Blendle has done extremely well in the Netherlands and Germany. It’s hard to break into the U.S. sometimes when you aren’t living here. I haven’t counted them out yet.
For us, the key is you can’t ask too much of users. If you ask them to change everything about how they discover content, that’s difficult to do. We’ve been trying to solve this by looking at a lot of the corpses in the graveyard. What were their Achilles heels? We’ve been able to learn from their experiences, and hopefully that will give us a pretty good start.
Have you gotten any pushback from people on the ad side? If Scroll works out, is that going to cut into a business model that is still making a lot of money?
If you look at any kind of freemium model, as you go, you’re incredibly lucky if 10 to 15 percent of the audience upgrades from the free part to the premium part. I would be very surprised if any publisher says that they have a 100 percent success rate.
The job of journalism has never been more vital. At the same time, the ability of the industry to fund that journalism has never been more under stress.
Also, sometimes people on the ad side of the business get unfairly maligned. There are people who care deeply about user experience and the content they’re creating. Some of them really hate some of the formats they have to sell in order to make their quota. One of the things I think Scroll can do is start to diversify the revenue sources. Advertising will still be a key part, but if there’s a little less pressure on the ad sales people, they can focus on the deals they really want to do.
Taking a page out of Zuckerberg’s manifesto, if you project to 2027, what do you hope you can accomplish by then?
The main thing I hope I will accomplish by then is that there are more journalists being able to bring truth to power, getting paid well, and reaching a wide audience. That is what we’re building for. And if we can do that, if I get hit by a bus in 2027, I will have considered my life well spent.
This interview has been lightly edited and condensed.Image by Chartbeat