There's an old line in marketing, usually credited to Theodore Levitt: people don't want a quarter-inch drill, they want a quarter-inch hole. Clayton Christensen and others built a whole method on the same idea and called it jobs to be done. Customers don't buy products. They hire them to get a job done.
Feature requests are the drill. Your job is to find the hole.
Read a request for the job
Take a typical post: "Add Excel export." The feature is clear. The job isn't. Some possibilities:
- I need to report our roadmap to finance every quarter.
- I need a backup in case we leave.
- I want to analyse votes by customer in a pivot table.
- My manager won't log in, so I have to send them something.
Each of those jobs has a different best answer. The finance report might be a scheduled email. The backup might be an API. The pivot table might be a better filter in the product. The manager might just need a share link. Building "Excel export" might solve one of the four and leave three people still stuck.
Ask "what were you doing when you needed this?"
The single best follow-up question on any request is some version of that. It moves people from describing a solution to describing a situation. Situations are where the job lives.
You can ask it in a comment on the post. You can also put it on the form, so people answer before you have to ask.
Look for jobs across requests
Once you read for jobs, you'll notice that several different requests are the same job. "Excel export", "share roadmap with a link" and "embed the roadmap on our site" might all be "keep stakeholders informed without giving them an account". That's one problem with one well-designed answer, not three features.
This is also the best reason to keep merged posts' comments. The comments are where people explain the job.
Don't over-correct
Sometimes the request is exactly right. People who ask for CSV export frequently do want CSV export, and a ten-minute interview to discover their deeper motivations is a waste of everyone's time. Read for the job when the request is expensive, vague, or contradicts other requests. When it's cheap and clear, just build it.
The aim isn't to second-guess customers. It's to make sure that when you build the drill, it makes the hole they actually needed.