Killing Features is a Product Manager's Job Too

Introduction

It’s a sunk cost. A famous excuse for releasing products that customers didn’t ask for.

So why do we do it? Why do we (Product Managers) release things that we know don’t solve the problem our customers have anyway?

It starts with a fundamental misunderstanding of what sunk cost actually means in product development and a product manager’s responsibility in capital allocation.

There’s pressure to complete a feature once it’s started. There’s pressure to push something you finally got on the roadmap to completion regardless of if it solves a true customer issue. There’s pressure to ship something just to say you shipped it.

As product managers, we are responsible for outcomes above simple coordination. And sometimes, the best decision is to stop.

Why Product Managers Fall Into This Trap

For those of us trying to block economics classes out of our memory, the sunk cost fallacy is a phenomenon where we are reluctant to abandon something simply because we’ve already invested in it.

It can be obvious that abandoning a project is the best path forward yet something inside of us will instinctively resist.

Your first round of validation for a new feature went great. All the customers you interviewed identified this problem and were ready to say yes to the demo you showed them. With this new information in hand, you continue to develop the idea. Competitive research. Resource planning. Selling it internally.

After all that work, your feature finally ends up on the roadmap.

You develop the demo more and show a functional prototype to your customers. Suddenly, they aren’t as excited as they were before and don’t see a good use case. Don’t you wish they would’ve realized this a couple months ago?

The trap is set. We start to think dangerous thoughts like “I’ve already done this much” or “I’ve already come this far” that cloud our ability to accept the fact our customers just said this new feature won’t do the job.

This moment right here is where you should stop and dig into the problem further with your customers. Figure out how you can better align the solution with their problem and get to work.

But so many of us don’t. Why?

• We’re incentivized to ship: Our performance reviews celebrate launches, not cancellations. Killing a project feels like failure, even when it’s the smartest decision.
• We become emotionally invested: That feature you fought for in roadmap planning? Walking away feels like admitting you were wrong (even though it’s perfectly okay to be wrong).
• We confuse activity with progress: As long as the engineers are coding, it feels like we’re moving forward. But activity without customer validation is just expensive procrastination.

Learning This Lesson The Hard Way

Yes, it’s true. The guy writing an article about product management has made a mistake. Quite a few of them.

This one was in my first product management job. I had put together a new dashboard UI with my team. Drag and drop builder. All kinds of visualizations. Our ATS customers could make anything they wanted. This reporting module could do it all.

Except actually give our customers the simple turnover insights they wanted.

I put on my collared shirt excited for my first demo of the new app. My mind was absolutely racing with all the amazing things our customers were going to say about our new reporting module.

The demo couldn’t have gone better. I breezed through all the best features and even paused for effect at all the places I had practiced. I got to the end and we asked for questions.

So where is the turnover reporting screen?

Unfortunately, this is the only direct quote I remember so I took some creative license with the next few.

You can make it yourself with our new flexible dashboarding tools. It has all these cool features, remember? I can show it again.

I’ve never been so grateful a call was on Zoom. My leg was jumping. Sweat rolled down my arms. I was doing anything to avoid this question

So you can’t show it to me now? And I have to make it myself?

The customer had gone from confused to upset in just a few minutes. Why?

Rewind a couple months, I had shown them a slick turnover report. They all loved it. Naturally I thought, if a single turnover report was cool then just imagine how awesome it would be if we could just let the customers build any dashboard they want including the turnover report.

Looking back on it now, I should’ve known better. I should’ve stopped in my tracks after I showed that first demo, but I didn’t.

Instead I pushed forward in the name of getting something shipped and wasted my team’s energy building something our customers didn’t even want.

So How do You Avoid This Mistake

It’s simple but extraordinarily difficult. Don’t get too attached to your clever solution. The customer is the ultimate judge of value, and I ignored that judgement because I thought my solution was better regardless of what they said.

Sunk costs are not the time we spend exploring and building. That’s our job. The real sunk cost is the resources we invest after there’s clear evidence it’s not the right solution.