The full New York Times Innovation Report was leaked last week.

It’s worth reading if only because it lets you look at a paradigm – an entire way of thinking, laden with it’s own explanations of culture, causal factors, jargon, assumptions, myths, systems, and heretics.

It enumerates the preferences and aspirations of a small group of people (including their preferred org-chart re-org!) and highlights a long-standing tension between technologists and journalists. It may also serve as a wake-up call that continuous improvement and scientific management is already a reality at several disrupting media startups.

Let’s begin.

Summary if you didn’t read it (and won’t):

The document contains 97 pages.

The term “Competitor” is mentioned on 39 of those pages. Analytics on 18 pages. Personalization on 8. Decisions on 7.

There are two chapters, one labelled “Growing our Audience” and the second “Strengthening our Newsroom”.

The first chapter for growing our audience contains three proposals, called:

  • Discovery
  • Promotion
  • Connection

Each of those chapters contains specific recommendations:

Discovery contains a recommendation to create an audience development role, surface the New York Times archives, packaging multiple similar articles into a bundle/experience, personalization, and more robust meta-data tagging.

Promotion contains a recommendation for the ‘business side’ to work with the newsroom to promote stories, have editors promote stories (using a compact toolbox), and mapping influencers.

Connection contains a recommendation to facilitate user generated content by way of expanding the number of op-eds, creating TED/NPR style events (paid conferences), getting to know their readers more.

The second chapter contains three primary recommendations:

  • Collaborate with the business-side units focused on reader experience
  • Create a newsroom strategy team
  • Map a strategy to make the newsroom a truly digital first organization

In that last point, the authors enumerate three steps in “how to get there”:

  • De-emphasize print
  • Assess digital needs
  • Explore more serious steps


The greater the dissatisfaction, the greater the effort and the wider the search for alternatives (March, 2009).

This passage appears to be the stimulus for such a search:

“…over the last year The Times has watched readership fall significantly. Not only is the audience on our website shrinking but our audience on our smartphone apps has dipped, an extremely worrying sign on a growing platform.” (p. 3)

A point reinforced with four ‘key measures’ (p. 23):

  • Home page visitors are down (error on the axis – cosmetic design can have drawbacks).
  • Pageviews are flat.
  • Time spent is down.
  • iPhone active users are down.

The digital business is perceived as stagnating or in decline.

There is no single, unified, causal analysis as to why this is the case. The document suggests that money from well-funded new entrants and technological change are driving this. A direct link isn’t made. And perhaps this wasn’t part of their mandate.

“This report represents our best attempt to provide answers.” (p. 9)

This wasn’t designed for the world to read it. It’s org-centric. There are key passages that are indicative of why they believe their organization isn’t structured to win:

“The wall dividing the newsroom and business side has served The Times well for decades, allowing one side to focus on readers and the other to focus on advertisers. But the growth in our subscription revenue and the steady decline in advertising — as well as the changing nature of our digital operation — now require us to work together.” (p. 60)

And a few cultural points are made:

“Our mindset is to perfect, then release. This should always be the case for our journalism. But we must question whether everything needs to meet this standard. Our competitors are launching new products or features as betas, and then using vital feedback from readers — rather than another round of internal feedback — to improve.” (p. 86)

And especially because,

“Because we are journalists, we tend to look at our competitors through the lens of content rather than strategy.” (p. 24)

That strategy lens was particularly strong in the late 90’s when distribution was the dominant strategy. That’s what the whole ‘portal’ strategy was all about – Canal+, AOL, Yahoo all pursued a portal distribution strategy.

Many of the competitors listed in the document look at their competitors through the lens of audience. Several of them are extremely audience centric, they are not traditionally-strategy focused.

There’s a gap between what the news desks are, and what they aspire to be:

“To succeed in the coming years, news desks need to be building digital skills. Indeed, a major reason producers have seen their access to tools and templates curtailed is because of the concerns that the editors on these desks are unable to recognize substandard work.” (p. 77)

Which is in part because:

“There has been a longstanding tension between dispersing and centralizing our digital talent. The choices that result affect the ebb and flow of experimentation in the newsroom.” (p. 76)

So to solve this problem:

“We have identified five areas that warrant more investment: strategy, analytics, product, platforms and audience development. Our competitors — old and new — have been staffing up in these disciplines, and we must join the battle to better meet our digital needs and to build a deeper bench of digital talent.” (p. 94)


“Create and distribute a clear organizational chart of the Reader Experience departments that includes a contact person and their newsroom counterpart.” (p. 69)

Which means a new strategy group in the newsroom:

“A strategy group could help provide conceptual help, structure and guidance to experiments launched at the desk level, allowing more producers, editors and reporters to innovate and learn. They would be familiar with the tools and talents in the newsroom, Technology, Product and Analytics that could help bring such ideas into reality. This would allow us to use desks as laboratories to answer pressing questions and to develop best practices.” (p. 77)

And will enable:

“Unlike a printed newspaper (which is polished to near-perfection and “launched” once a day), a digital experiment should be released quickly and refined through a cycle of continuous improvement — measuring performance, studying results, shuttering losers and building on winners. The Verge, for ex- ample, redesigned its home page 53 times in two years. We must push back against our perfectionist impulses. Though our journalism always needs to be polished, our other efforts can have some rough edges as we look for new ways to reach our readers.” (p. 31)

So there’s a proposal here to bring the tools of the business-side into the newsroom to enable a test-bench that results in continuous improvement cycles, causing a sequence of best-practices.

In other words, asking the producers and newsroom to behave like a scientific startup.

What of Analytics?

What do they mean, and how many people are there?

“We use “Analytics” as shorthand for Consumer Insight and Business Intelligence” (p. 62). It has ~30 people (p. 64). 8 work in R&D (p. 64)

They identify a gap:

“Need: We don’t regularly use data to inform our decisions in the newsroom, which means we are missing out on an opportunity to better understand reader behavior, adjust to trends and drive traffic to our journalism. This makes it more difficult to set goals and assess progress. A strong analytics operation is essential to every one of these digital needs. Opportunity: Analytics skills are needed in many parts of the newsroom, including for top-level strategy as well as desk-level decision-making. We need to hire analytics experts to work with news, platform, and product editors, newsroom strategists and the people trying to grow our audience. We need to also work closely with data scientists in the Consumer Insight Group.” (p. 94)

But I see some pretty big cultural problems:

“Currently, our capabilities for collecting reader data are limited. The information is dispersed haphazardly across the organization and rarely put to use for purposes other than marketing. And the newsroom, which is perhaps best positioned to champion this effort because of its close connection to readers, has not played a leading role.” (p. 54)

Which are linked to the incentive structure:

“Most reporters know exactly how frequently they’ve appeared on Page One in the previous year — indeed, annual performance reviews often lead off with that figure. Similarly, desk heads are keenly aware when they have a dry spell of Page One stories, and backfielders spend countless hours each week to making sure the pipeline is filled with stories that could be offered for the front page.” (p. 87)

And then that hope is lost pretty quickly with a reinforcement of the existing paradigm:

This feedback — which a new analytics unit will make far easier to provide, based on data — is essential because New York Times editors and reporters will always be working at capacity. They just need clearer and more consistent signals. “They mostly seem to care about the front page and big, giant stories, and that’s great,” one desk head told us. “But if 13 million people need a news alert, we ought to know that. It ought to be somebody’s job on the masthead to tell us that. We do respond, but it’s a matter of shifting our burdens.” (p. 87)

The document does not contain a disruptive innovation

The findings may surprise senior management at The Times (which is likely its intent). But on it’s own, there is nothing disruptive.

Most of the tactics listed in the first Chapter are already executed by competitors. The re-org recommendations in chapter two, while perhaps edgy to established incumbents, are in place at several competitors. Moreover, nearly all of those competitors are after different audiences (markets). They have all developed content strategies that match.

Many media startups have an audience development director in the form of the CEO. Audience development is their core focus. Continuous improvement and scientific management are gospel at several of these competitors. It’s not done perfectly and long standing tensions between content and technology persist, but these principles are in place and it’s baked into how business is done.

This collection of tactics, if executed, might only bring them to a state where Mashable or The Verge was last year. Moreover, both organizations are flatter and do not have two strategy groups that liaise with one another.

It’s very hard to tell if the tactics enumerated in this document constitute a coherent strategy because we don’t know how it links to what New York Times readers value, or, which audiences the The Times wants to expand into.


The document raises the point about audience development.

As a researcher, I always expected the New York Times to be right. Whenever I was tearing away on a paper, looking for facts and quotes, I used The Times index. I expected it to be accurate. That kind of professionalism in content generation is very different from the What Is Your Inner Potato? style of content you’re finding elsewhere (that’s real, click the link.) The Times are just a much higher standard.

Audiences come in such great varieties. Some audiences really enjoy quizzes that ask “What is your inner potato?” – just check the Facebook Comments to see a sample of them. The academic audience is different.

The media startups that went after casual audiences didn’t ever have to worry about being accurate or right. That meant a closer alliance with technologists was possible and desirable. Technologists are far more likely to iterate. The current paradigm of scientific continuous-improvement entails releasing best-guess product/content into the market, and very rapidly iterating on it. The combination of those two factors, light-weight content combined with heavy weight technology, was itself disruptive.

That makes it all the more harder for the New York Times. The argument for a segregated newsroom is all the more convincing when one considers how much focus it takes to write accurate stories. They don’t need the distraction of audience development. Very specifically, a newsroom producer asked for signals, not anything else. How can collaboration happen if content professionals, in this case journalists, are burdened with a higher standard.

There is a stark choice here.




It’s worth reading if only because it lets you look at a paradigm – an entire way of thinking.

They have their own culture, and it’s striking just how deep the institutional lock-in is. There are a few pretty strong assumptions about the importance of content, and, you can detect a whiff of who they were fighting. It took some courage to put this together.


It enumerates the preferences and aspirations of a small group of people – namely, a new content-side strategy group. It seeks to pull in technologists.

It may also serve as a wake-up call that continuous improvement and scientific management is already a reality at several disrupting media startups.

It also says a whole lot about audience formation and an emergent mix of medias that people will select.

The NYT produces 300 URL’s each day (p. 27). Not all of that content is relevant to all audiences. And, sometimes, very sophisticated audiences can be in the mood for low-effort content. It’s likely not the case that I want to get my fix of Internet sub-culture from the same source that blows the whistle on those in power. It’s in the mix.

I thank the leaker.