How to fail forward

By Ava Mandoli

It’s probably not often that you get up in front of dozens of people to explain how your idea failed. 

But at the 2024 NPA Summit, three media startup founders – Bárbara Libório, Crystal Houston, and Maria Vitória Ramos – did just that. They poked holes in every part of their media startups, explaining where they went wrong and why. In doing so, they demonstrated how they failed forward. 

Libório, Houston and Ramos all spoke about their experiences launching very different types of media startups in unique environments: one fighting gender bias, one combating misinformation, and one working on government transparency. Every media organization has its own unique challenges, but there are common potholes along the main road of news product development. During the virtual session, attendees from media organizations around the world shared their commiserations over common failures and struggles.

Reflecting on past mistakes is important to learn from them, but the panelists said one surefire way to fail backward is to beat yourself up over every misstep. This session was to be a place for people to not only learn together, but also to heal together, they emphasized.

Here are a few “failures” that the speakers discussed. 

Assuming you know your audience 

Instituto AzMina always had a strong idea of who their audience was – or, at least, they thought they did, said Bárbara Libório. The non-profit Brazilian organization produces a gender equity-focused digital magazine and technological tools to track women’s legislative rights. They assumed, Libório said, that their audience was mostly middle-to-upper class young women who were interested in education, politics and technology. But after launching a newsletter about women in science that didn’t pick up much traction, they began to question their assumptions about their audience, according to Libório. 

“For 10 years we made decisions that didn’t really have their audience as a guide because we didn’t really know them well enough,” she said. 

Operating based on your assumptions of your audience’s wants and needs is “imaginary empathy,” Libório said. After conducting their first large-scale audience research project, they found their audience was actually older and had much more diverse interests than they had assumed. 

Libório said these research insights “gave us more confidence in our work.” 

Her advice: don’t start by creating a product and finding the audience later. Start with research and identify what problems your audience cares about, then develop products based around that. 

Not investing in in-house expertise

One struggle that many session attendees shared was that of actually developing news products with a tech team. Ramos, who co-founded a Brazilian government accountability startup called Fiquem Sabendo, discussed how challenging it was to work with a third-party tech team. It was expensive, time-consuming, and didn’t allow her to iterate on the product as much as she would have liked to. She said she wished she had invested in an internal tech team from the beginning. 

Even though many organizations may not have the funds to hire a full internal tech team, there are still better options than outsourcing everything, she noted. One participant suggested investing in hiring a few internal junior tech team members, but having an external consultant manage them; another replied they had a good experience doing so.

When working with any development team, internal or external, participants also expressed that  setting timeline expectations is always a challenge. In the chat, one former developer and current product consultant said they found tagging to-do items with S/M/L/XL stickers was helpful for creating shared estimates of project timelines. 

For Ramos, having a product manager who knew how to balance several teams’ needs was extremely helpful. She said her product manager even changed the direction of her startup entirely, steering it towards the non-profit organization it is now.

Building a homogeneous team

One thing Houston has learned from her experience co-founding several companies – including a misinformation-fighting media startup called Citefull – is how to build a good team. 

“Look for complements, not compliments,” Houston said to the audience. Your team members should not be the people you “vibe” the most with, she said – that might be a red flag that you are too similar. If you’re an extrovert who loves the human-interactions side of the media business, Houston suggested trying to find a partner who’s more partial to the analytical side of company operations. 

Reflecting on her time working on Citefull, she said she did not realize how vital it was to have conversations about and define her team’s shared vision on journalism ethics and professionalism. For media professionals looking to partner with team members from non-traditional media backgrounds, she suggested reviewing industry codes of ethics together. 

Communicating with your team and creating an operating agreement at the start is also critical, according to Houston. “It's important that you enter a co-working relationship with your eyes wide open, and that you realize the vast majority of all new businesses fail,” Houston said. Having a written document that outlines how you’ll solve conflicts is key.

Refusing to fail

As a self-described perfectionist, Ramos said one of the hardest things about being a media product leader was allowing herself to fail. When Fiquem Sabendo was selected by the Google News Initiative Innovation Challenge in Latin America, they received a grant so they could test their startup’s hypothesis. She said there was no requirement that they had to succeed or produce anything. Despite this, Ramos said, she never accepted the possibility of failure — “I tortured myself over it,” she said. 

If she could do it again, Ramos said she would have implemented more clear stop-loss benchmarks. That means predefining success or failure criteria to halt projects if they don’t meet goals. Not only can that help you minimize losses, but it also forces you to continue iterating instead of getting stuck on an underperforming idea, she said.

Stop-loss, to Ramos, is not giving up: it’s managing uncertainty and protecting your resources. 

Failure is, after all, a learning opportunity. Failing forward means taking those lessons, pivoting, and channeling your new insights into future projects.