A good hunch is just the beginning

Every product I’ve ever worked on started with a gut feeling.

It’s usually a hunch that a newsletter would land, that a guide would finally answer the question people kept asking, or that readers wanted something the site wasn’t giving them. In a small news organization, instinct can often drive the best innovative features and ideas. “Hunches” are inherently important. But if we stop there, we’re leaving a lot of even better ideas on the table.

Having a good hunch will carry you a long way. It’ll get you through the pitch to your peers or bosses and maybe even to a first version. What it can’t tell you is whether the people you’re building for actually want the thing. Sometimes we build it anyway, with a strong feeling and no evidence, and we find out whether we were right after it already exists. But that’s a costly way to learn something a reader could have told you in two weeks.

That gap is what our recent New Jersey Product Coaching Program was able to address. Seven small New Jersey news organizations, most of them with one- or two-person staffs, each picked one product in May and worked it through with a Blue Engine coach until the end of August.

On our last call together, coach Cynthia DuBose walked the group through how they had scored themselves on a set of product fundamentals in May and again in August. Across the seven teams, overall scores climbed 44%, and one of the biggest jumps was in making decisions based on data.

"We want you to stop just saying, ‘Oh, my gut is telling me to do this,’" she said. "We want you to use data to feel good about what you're putting out in the world."

Cynthia, Will Bergmann and I spent the summer coaching these teams through those decisions. The goal wasn’t for teams to stop trusting their instincts but rather how to stop, well, stopping there.

Every team came in with something they already knew. One founder knew her newsletter needed attention. She compared it to perfecting an omelet before trying to make a fancier dish, and she was right. Her gut told her a civic events calendar and government jobs board would make the newsletter more useful.

But then she asked readers. Very few picked either one when asked to choose the features they wanted most. Instead, they wanted more of the accountability and policy reporting she was already doing. She changed her priorities before spending time building something readers didn’t really want.

Another team had a different kind of hunch. Years of conversations told them that small business owners in their county were hungry for better resources, and their research confirmed the need. But it also caught a problem they hadn’t anticipated: the name. When they tested their working title, most people couldn’t tell what the newsletter was supposed to be. A small fix, but one that could have cost them subscribers.

For another publisher, research validated what she already suspected. Her assumptions about her audience were right, down to her subscriber demographics. She didn’t need to change course, but the research still gave her something valuable: confidence that she was building for the right people and could move forward without second-guessing herself.

Most of us skip asking because we think we don’t have time. We’re carrying five ideas at once, and researching all of them sounds like one more thing we can’t fit in.

So we asked every team to pick one product. They got specific about who it was for, what that person needed and what the product could do for them. Then they put the value of the product into a single sentence they could actually test.

Once teams knew who they were building for and what they needed to learn, the asking part took almost no time.

There was no reason for elaborate research. A simple form before creating a new product to gauge interest. A quick question to a handful of business owners on Instagram. A rough newsletter to 25 people to get a quick sense of response. These were 10-minute tests that cost nothing.

Who they asked mattered, too. One team went directly to business owners instead of its subscriber list and got better answers. Another started with her most loyal supporters, realized she was mostly hearing from her biggest fans, and went back to readers who had never given her a dollar. That second round changed her plan.

As one publisher put it: "It's so much better … you can really check that gut with actual data."

All seven teams still have their hunches, and they should. But they’re now more comfortable asking the people they were building for what they actually wanted, listening when the answers surprised them and adjusting their plans based on what they heard. By the end of the program, they had guides, newsletters and a magazine issue shaped not only by what they believed their audiences needed, but by what they had learned directly from them.

More importantly, they now have a process they can use the next time an idea comes along. 

The New Jersey Product Coaching Program was a partnership with the Center for Cooperative Media at Montclair State University and the New Jersey Civic Information Consortium, funded by Google.

As always, we're ready to answer your questions and hear your thoughts at [email protected].