Mark Thompson: The term “platform” is just slightly, if I say so, unsatisfactory, because different people mean different things by it. While we’re arguing about substance, I tend to use the word “platform” to mean a very big digital environment which hosts content from many players and may often have a big social component as well. I think of Google and Facebook as true global, digital platforms.

I think of Netflix as a subscription destination, in the way that we’re a subscription destination, if you like. We allow more sampling than Netflix — Netflix does its marketing through clips and so forth, rather than through allowing as much sampling as we do. It’s probably because, typically, our unit of content is a much smaller unit and we think we can let people look at a few and then they get hooked, they can pay to see the rest.

There are reasons why the usage is different from entertainment. Reed Hastings and company at Netflix are going for a much broader spectrum of content, and so in the entertainment space, Reed is going for everything in a way from very tony, foreign-language art movies, through to very mainstream, as well as what once would have been syndicated TV. It’s a very, very, broad spectrum. We’re not.

Thompson: Reed’s doing quite well. We want to be part of a successful content industry. Honestly, we want to be a globally scaled subscription business. There aren’t many legacy news organizations we compare ourselves to.

Thompson: It’s branded, and it’s very congruent with the brand, and we’re pretty jealous of the quality boundaries and the seriousness boundaries of the brand. Even our food coverage is rather Times-ian, in a way: the quality of the writing, the recipes we tested 50 times.

It’s like curated content from, in a way, a single editorial spirit, whereas Reed is going for great diversity of creative content. It’s a polyphony of lots of different kinds of creativities. It’s great, but we have something that is much more shaped.

My view of the Times has been that in its DNA, it’s thoughtful. It’s always been journalism for a certain kind of person — a person who prefers to devote some time to it. It’s not a generic or full-spectrum offering.

Thompson: I think Netflix is a complete entertainment solution for many, many different kinds of households. I think we’re for a certain kind of household. Having said that, there are many, many, many tens of millions of people like that on the planet. There’s no reason why we couldn’t have a very, very big subscription base.

In the 1950s, I think the Times was the fifth most widely read newspaper in New York City. Number five, I think. It was not New York’s most popular newspaper, or its second-most popular, or its third-most popular. It’s been for a certain kind of thoughtful reader, certainly since Adolph Ochs took it over. That character of recognizing that you’re not a General Motors or a Toyota, as it were — you’re a Mercedes-Benz. That’s not an exact parallel, but it’s a handmade, thoughtful thing for thoughtful readers. That, to me, is part of this brand.

Thompson: The Times was considered a very serious newspaper pretty much from the beginning of the 20th century onwards. Ochs buys it in 1895 with a thesis about a serious, thoughtful newspaper and a civil, intelligent opinion from every perspective, and all this other stuff. By the mid-20th century, it’s been considered like that.

Winston Churchill, in the 1940s, got the British Embassy in Washington to put Times leaders, editorials, in the dispatches so he can read what the Times is saying America should do in relation to the Second World War. The voice of The New York Times was very influential in America then.

The character’s changed. The Times, with some very, very famous exceptions, has only relatively recently gotten known for great investigative journalism. That was not the Times’ reputation, exactly — notwithstanding the Pentagon Papers, and there’s similar others, some famous exceptions to that. I think the Times’ strength in investigative journalism associates significantly with recent executive editors and particularly with Dean Baquet.

Thompson: We think it’s a virtuous thing to do, and we’d like, if we can, to figure out ways in which we could play a part helping more news organizations do that. That’s another conversation, but we believe at the state level and the city level, accountability journalism — which would be a mixture of beat journalism and investigative journalism — is incredibly important. We’re certainly not beginning to claim that we could do all of that ourselves.

Thompson: We made this significant change in July. We’re asking pretty much everyone who wants to read more than one article to register and log on, after which they get a greater number of articles.

Thompson: It depends.

Thompson: Still optimizing.

Thompson: Somewhat more nuanced way, somewhat more nuanced way.

Thompson: The biggest single effect is to greatly increase the numbers of registered logged-on users — registered logged-on non-subscribers, in particular. I have to say, though, it’s also helping some subscribers because it’s also dealing with another issue, which is subscribers who draw back on their logged-on state for whatever reason on a device where they never registered. Which happens a lot, and people appear as anonymous users.

Thompson: It’s confusing to consumers and confusing to us. Someone who is a very loyal, 10-year subscriber turns into an anonymous user.

The consequences? Being able to track a much larger number of people across devices and over time. Serve them better. Begin to learn how to, in ways which are respectful of their choices, put content in front of them, send them messages.

Thompson: That’s right. We are also very interested in how you, as it were, create more centers of content-based value in secondary products. The bones of capabilities in podcasting and TV are all examples of trying to create other chunks of value. So far, we’ve typically thought of newsletters and so on as journalistic exercises which are designed to add value, generally, to subscribers and non-subscribers alike. Unlike the FT, we’ve pretty much avoided the idea of an interior, premium layer of content.

I’m an FT subscriber. I bump into articles inside the FT I can’t read because I think I’m not above that many people. I’m fairly hazy about what kind of subscriber I am but I think I’m a non-premium subscriber. Rightly so.

[The FT offers “Standard Digital” and “Premium Digital” subscriptions, at $335 and $585 per year, respectively. Premium Digital includes the Lex column, additional “in-depth analysis,” and the digital replica edition.]

I’m a big fan of John online. I’m a very loyal, multi-year subscriber to the FT in different guises — sometimes even with my own money! But I get irritated by that. I like the fact that with the New York Times, you get everything. You get “the Times.”

Thompson: First of all, we accept donations.

Thompson: That’s our standard sticker price for all-access.

Thompson: One of the things we talked about on the call yesterday is firming up plans for that, price-wise, in 2020. For tenured subscribers. This may mean you.

Thompson: I think we’re starting differently.

Thompson: Once a subscriber has graduated from a trial offer to a main offer, and has been on the main offer some time, some of these subscribers —

By the way, I’m one of them. I became a digital subscriber in 2011 [when the paywall launched]. I became a digital subscriber, I would say, at least six months before I got a call from Arthur Sulzberger, but I was running the BBC. I love The New York Times. I think it’s one of the best journals in the world. I became CEO after becoming a digital subscriber, myself. That’s my order of events. I’ve been paying the same subscription, I think, since 2011.

Thompson: Okay, I’ve stacked up since then — you’re right. I was living in England and now I live in America — I get a physical paper as well. The digital subscription I bought then…I bought it in dollars and I think I bought it at $15, I think. I’d still be on $15.

Since then, we’ve added hundreds of more journalists; we’ve got eight years of not raising prices. As a subscriber, I think the idea that I might have a price rise after eight years is not unreasonable.

Thompson: We haven’t disclosed that yet. We did some testing this year with a significant number of subscribers and, generally I’ve found that, when we explain what we’re doing and why we’re doing it, their willingness to pay more was very high, indeed.

[As Thompson put it on the earnings call: “The success of our price rise tests, and our growing confidence in our ability to deliver discrete messages to different segments of our subscriber base has convinced us that we can execute a price rise for tenured subscribers with minimal risk of reducing new subscriber growth momentum.”]

Thompson: We did increase some prices. What we said yesterday is we’re firming up plans to roll out — we didn’t say for how many, but for a much larger number of equally tenured subscribers.

We are seeing some significant acceleration in volumes. We put on 273,000 subscribers in Q3.

Thompson: Yes.

Thompson: The question has been: Can we do it in such a way that we can satisfy ourselves, that won’t damage the top of the funnel? We won’t get too many people bouncing off entirely. A lot of testing has convinced us that we could execute without losing significant unique users.

Thompson: It’s an informal, internal number, but we think 145 million in September.

Thompson: We’ve looked. We haven’t disclosed this, but we’re looking closely at pageviews. We’re looking closely at how many days a week people are returning. We’re looking at a whole set of parameters. We think it looks pretty healthy, actually. The number of registered logged-on users onsite is growing every week. We’re seeing virtuous behaviors. We’re seeing better-than-expected behaviors from these cohorts coming in since July.

Thompson: More consumption and more conversion. We seem to have more conversion than we’ve seen before. We’re still seeing an uptick now, from that cohort — the first week in July’s cohort.

Thompson: I’m not convinced that we need to stay at three percent. I think by the end of 2019, admittedly including cooking and crosswords, we will be fairly close to, or may exceed, 1 million net new subscriptions in the calendar year.

Thompson: That’s right. 209,000 in Q3 were to the news product. That’s 46 percent more than last year. Last year was, I think, 36 percent higher than 2017. We’ve seen a lot of buoyancy.

Thompson: Accelerating.

Thompson: Yeah.

Thompson: We haven’t disclosed that.

Thompson: All I want to say is the rate at which that’s growing is very encouraging, and it’s large-scale.

I think one of the striking things — we’ve been aware of this for some time — is that of our very large number of unique users, 145 million, we had a very small number of people who were registered non-subscribers. There weren’t many very powerful reasons for registering.

Thompson: Significantly, yes. Now, I think we can say that, although we’re still building it, that’s going to be a very large number. There weren’t many other reasons, really.

Thompson: Again, the answer is yes and no. Yes, we know. No, I can’t tell you.

We’ve often said that the number of print subscribers who register and log on to use digital access is very high. I haven’t got an exact number for you, but if you think of a number, which is well above 80 percent, and maybe now even in the 90s, it’s a high number. I think many people who are on Sunday or weekend print delivery actually subscribe to that as a combined subscription, and it’s an efficient way for them to get both the Sunday newspaper and digital.

Thompson: Oh, of course.

Thompson: $800 million in digital revenue [as a Times goal] was three or four years ago. [October 2015, to be precise.] 10 million [subscribers as a Times goal] is, I think more recent. [December 2016.]

Thompson: Right. We’re ahead of track.

Thompson: Yeah, we’re heading to 70/30 — 70 percent subscriptions, 30 percent advertising — and the model is going to tilt further.

Thompson: Subscription is growing rapidly, as well. The new digital subscription revenue number — that was a 14.5 percent year-over-year rise. It’s not just that we’ve got declining advertising.

[Advertising was notably weak in the Times’ Q3 numbers, dropping 7 percent overall, including a 5.4 percent decline in digital.]

The key is growing total revenue. [Which the Times did, increasing total revenues 2.7 percent for the quarter, now a consistent trend.]

We’re holding margin, reasonably, while investing rather than just rising profit. Adjusted operating profits have moved up and down, but we’ve kept it in a band in the middle 200s. We’ve been investing a lot. Our investment bank is the print product. The print product is a very cash-generative, high-contribution margin activity. We love it. It’s a wonderful thing, but we love it also because it gives us the room to invest.

Thompson: If you include Opinion, which you probably should, we’re probably 1,700 plus, now.

There are three broad topics of investment. The newsroom, journalism; the product and tech teams, so engineers, product people, data scientists, and the rest of it; and marketing. Marketing has got two components. There’s a bigger effort towards brand marketing and performance marketing, as part of the subscription growth story. One of the encouraging things about Q3 was that, although we added many tens of thousands more subscribers in Q3 than Q2, marketing spending didn’t go up. The lifetime-value/subscriber-acquisition-cost ratio improved.

Thompson: Yes. These things bop around.

If there’s an opportunity, in a way which makes economic sense, to spend money to get lots of subscribers, we’ll do it to build a base. That’s an example of funneling, which we do think we’re going to see, which is basically improving operating leverage. The complication with our model is all this happens at the same time.

By the way, we’re trying to invest in, essentially, new businesses like podcasting as well. Podcasting is a new business, potentially with new revenue streams. You can think of that as exogenous to the core, whereas the business of selling subscriptions to The New York Times is core. With the core, I think our view is that over the next few years, we’re going to see progressively improving operating leverage. I can’t tell you, as it were, at the enterprise level, when all the effects play out completely. It’s not far away.

Thompson: The cost of performance marketing is beginning to grow at a discernibly lower rate than the revenue is growing.

If we are spending X on our newsroom to service a base of 5 million total subscribers, we’re not going to need to spend 2× that on the newsroom to service 10 million — although we may well see the rate of investment take off there. 2020 will be quite a big year because it’s an election year.

Thompson: It’s now 16. 2025 calls for 20.

Thompson: We’ve been cautious in adding costs. We’ve experimented in quite a few markets with newsroom investment in different countries in the world, some more journalists.

Thompson: Our findings would be that that’s been less effective than improved marketing tactics. First of all, recognizing that, in the end, our proposition — we will cover the whole world. We’ve got 32 foreign bureaus, and in every market where we think there’s a really good marketing opportunity for our subscriptions, we will have journalists. You can’t really go to a country and say: “We’d love you to buy subscriptions, but we’re not ever going to cover your country.”

Thompson: The Guardian, I think, has actually, I think, done rather well in creating a new news surface for Australians. The Guardian’s operations in Australia are trying to give people local Australian media. If you can get that to work as a model, why not? Why not go and compete in Australia?

Thompson: We cover Australia because our readers in America and around the world will be interested in the most interesting news happening in Australia. We are local for global, if you like. We don’t see having to scale granular journalism costs.

There’s something analogous about our coverage of the continental United States. We want to be present in the country. We’ve got bureaus all over America. But when we cover America, we’re not trying to compete and substitute for local media. We’re trying to cover America because our readers are interested in what’s happening in every part of the country.

Pricing is significant. Willingness to pay, in many markets, is going to be significantly lower than the U.S., and you price accordingly. We’re going to be judicious about this.

Thompson: We learned a lot from it.

Thompson: I would say we’re going to continue experimenting internationally, but I think our view of international, in light of our own experience — I have to say, looking at others and seeing what, in particular, major journalistic deployments look like, in the end, it’s difficult to make sense of economically, generally.

Thompson: That subset of people is most likely to want to subscribe to the Times. I think there are some nuances. We’re experimenting in language. Opinion.

People who are very comfortable with getting the investigative journalism and U.S. news and global news in English may still prefer the tidbits about their region of the world or their country in their own language. And they like opinion pieces being written in their own language. You can read political opinion emotionally if it’s in your own language.