The conference room had a journalist on one side of the table and a product manager on the other. The reporter was explaining why a proposed change to the article page would bury a piece of context that was essential to the story. The product manager was explaining that the current treatment was getting ignored by roughly nine out of ten readers who opened the page on a phone. Both of them were right. Our job in that room was not to choose a winner. It was to figure out what to do when two forms of expertise, both legitimate, both rooted in craft, were looking at the same pixel and seeing different problems.
That meeting, and hundreds like it, is what I mean when I talk about teaching a newsroom to think like a product team. It is not about making reporters use Jira. It is about building a shared vocabulary for decisions that used to belong to one tribe and now belong to several.1
Questions that serve journalism
I want product thinking to help reporters do better journalism, not turn them into growth hackers. Dashboards can inform editorial judgment; they cannot replace it. The journalism is the product, and anyone who treats a scoop as a conversion funnel is going to end up with neither.
Product thinking, as I use the term, is a habit of asking three questions before you ship anything. Who is the reader, specifically, for this thing? What problem are we solving for that reader? How will we know whether we actually solved it? Those questions do not replace the editorial instincts that made a great reporter great. They surround those instincts with a second layer of discipline that catches the things instinct cannot catch, like how the story renders on a five-inch screen at seven in the morning on a phone with two bars of signal.
The moment the vocabulary started to hold
The cross-functional team is the unit of change. When a designer, an engineer, a product manager, and an editor sit in the same standup and talk about the same piece of work, the categories start to shift. The editor learns how the designer keeps the hierarchy of the story intact on the trip from a fifteen-inch monitor to a phone. The engineer learns why a caption carries information essential to understanding the lead image. The product manager learns that the reporter's instinct about what the reader wants, formed over thousands of stories, is often more accurate than the A/B test, because the test measures the wrong time horizon.
I have seen this happen in small, specific moments. A live graphics treatment for a breaking financial story where the designer suggested a tap-through format and the reporter realized, watching the prototype, that the format would actually let her tell the story in a way print never could. An iOS app refresh where the editorial team and the engineering team discovered, by sitting in the same room for a week, that half of what both groups had wanted for years had been blocked by a process, not a technology. A hackathon where a working prototype came out of a twelve-hour conversation between a reporter and an engineer who had never worked together before.
Product thinking gave journalism a second pair of hands.
Making shared work last
The vocabulary takes practice. A newsroom runs on daily and sometimes hourly cycles. A product organization runs on weekly and monthly cycles. When you put those two rhythms next to each other, the daily cycle will always win on urgency and the weekly cycle will always win on depth, and the two groups will start to resent each other for exactly the strengths that make them useful.
What I have found works is making the cross-functional team the smallest stable unit. Not a reporter pulled into a product meeting on Thursday. Not a product manager briefing the desk on Monday. A team. Same people, same ritual, same shared metric. The team does not have to be large. Four people who actually know each other's work beats twelve people who hand off tickets.
The other thing that works is being careful about what you measure. The wrong metric, chosen lazily, will push the newsroom toward work that feels hollow. Page views without time-spent, clicks without follow-through reads, subscriptions without retention. I tell product managers in training: if your metric can be juiced by a reporter doing worse journalism, your metric is wrong. Find a better one. The metric should reward the behavior you want to encourage, which in a subscription newsroom is almost always depth of engagement rather than breadth of reach.
What gets left alone
Editorial judgment about what is worth covering is not a product decision. A product team can help decide how a story is told, where it lives, when it surfaces, what formats it takes. A product team does not decide whether the story should be written. If you find yourself as a technologist pushing on that line, you are over your skis.
The writing itself is also not a product decision. The sentence-level craft, the tone, the voice, the pacing, these belong to the people who spent decades getting good at them. I have seen product organizations try to A/B test headlines into oblivion and what they get back is a homepage that feels like it was written by a committee of algorithms. That is a failure mode to watch for.
A newsroom that thinks like a product team brings two forms of expertise into the same room, with enough trust to argue and produce better work than either group would produce alone. That trust is the thing you cannot hire for and cannot ship in a release note. It gets built one cross-functional decision at a time, and it gets lost faster than it gets built. The only real job of the person running the organization across that line is to protect the rooms where the trust is being built, and to keep the vocabulary shared enough that the two groups can still hear each other when the next story breaks.
Footnotes
-
Kathryn Thomas, How the WSJ iOS team promotes cross-team collaboration through OKR-driven feature requests, Dow Jones Tech, May 29, 2018. ↩