The art of unshipping: How to deprioritize features, phase out projects and sunset products
In 2017, Apple did something no one expected. They removed the headphone jack on their latest iPhone. The media called it a mistake. Users took to Twitter to announce they’d rather jump ship to Android than deal with another dongle.
By the public response, the announcement was a huge mistake. But why would the world’s most valuable company risk upsetting their legions of fans?
Their answer? Courage. The real reason? Progress.
While keeping customers happy is essential for any sustainable business, they don’t always know what will make them happy. As Henry Ford famously said, ‘if I’d asked them what they wanted, they would have said faster horses.’
Jump to a section:
When you veer away from your product vision out of fear of upsetting users, your product is bound to become outdated, overly complicated, and bloated with technical debt.
This isn’t to say you should ignore user requests or phase out features for no reason. But that sometimes, the best thing you can do to create a better user experience is to delete rather than add.
So how do you deprioritize features your users want, sunset ones they love, and commit to an innovative yet risky roadmap?
In this guide we will go through exactly that.
What is ‘unshipping’?
Unshipping is a term coined by the team at Mixpanel to describe removing a feature or product that is already live to users. You could also call this killing a feature, deprecating, sunsetting, retiring, or any other range of terms.
However, unshipping can also go beyond removing a live feature.
Product teams become unnaturally attached to their product roadmaps. But a great user experience relies on simplifying your product, not making it more complex. ‘Unshipping’ isn’t the only path to simplicity. It’s just as important to deprioritize features or products that are planned for future sprints or even close to completion.
These two elements–removing live features and deprioritizing planned ones–provide a more comprehensive picture of what ‘unshipping’ really is.
But why would you want to remove features users want or throw a wrench into your perfectly planned product roadmap in the first place?
Sunsetting a product may feel like the end, but the sun always rises.
The real benefits of unshipping, sunsetting features, and deprioritizing projects
Developers and product teams are almost always focused on more. More features. More users. More projects. But being stuck in ‘build more mode’ can also lead to unwanted additions, such as more technical debt, bugs, and complexity.
While there’s nothing wrong with adding the right features to your existing product, the problem is when the rate of addition overwhelms any attempt to clean up your code, simplify your user experience, and stay aligned with your overall product strategy.
Every new feature has the potential to become your big differentiator. But they also require ongoing maintenance, upkeep, and support. Even worse, users don’t judge your app, tool, or site based on how much it does but on the value they get out of it.
|Adding Features||Removing Features|
|What it looks like 👉|| || |
|Potential rewards 👉|| || |
|Risks 👉|| || |
More than often, simpler is better. And that means regularly pruning your product like a gardener tending their plot.
Why it’s so hard to sunset features or deprioritize projects
Even the most experienced product and project managers are more likely to approach problems with a mentality of ‘what can we add to this to make it better?’ rather than ‘what can we take away to make it simpler, more intuitive, and valuable?’
But why is that?
Unfortunately, the human brain seems hardwired to hold onto what we already have rather than look at better options, thanks to a few cognitive biases:
- The Sunk Cost Fallacy is when to give too much value to tasks, features, or products you’ve spent time on. For example, when you continue to support an outdated and failing product because you’ve ‘invested’ so much time already in it.
- Confirmation bias is when you focus on facts and data that support your beliefs and ignore ones that don’t. For example, listening to that one big customer who loves an old feature instead of the hundreds of others asking for an update.
- The law of least effort is when you pick the path of least effort while ignoring someone else’s experience. For example, creating a workflow that makes sense to you but is confusing for customers without the same context and deep understanding of your product’s history.
- Loss aversion explains how you fear change and would rather keep what you have than try to build something new. For example, listening to customers who say they don’t want you to remove a feature when you know it will help the overall user experience.
- Completion bias is when you prioritize finishing tasks instead of questioning if you’re working on the right ones. For example, committing to your sprint plans or roadmap without testing assumptions or looking at other options.
Evernote: A case study on how complexity kills products
The real risk of these mental biases isn’t that you blow up, fail, and lose customers. Instead, it’s that you end up in product purgatory, doomed to a cycle of building more disjointed and lackluster features for a dwindling userbase.
If you need proof, just look at the story of Evernote.
The original productivity app quite literally got lost in its success. In the early 2010s, Evernote expanded from a note-taking app to a chat and recipe app and even sold Japanese-made socks described as ‘smart covers for your feet.’
Growth fell off a cliff a few years later, and new, simpler products took over the space. By losing sight of their original vision, Evernote expanded to the point of near obsolescence.
One of the best things you can do as a project manager is to help get rid of things. In other words, you need to look at your roadmap or current product and deprioritize features and future projects that no longer align with your vision and strategy.
More than often, simpler is better. And that means regularly pruning your product like a gardener tending their plot.
9 steps to sunset a product, deprioritize a feature, and design a phase-out plan
Streamlining your offering takes an unwavering commitment to your product vision, a deep connection to your users, and more than a bit of swagger.
But the results are clear. From a user perspective, simpler products are more intuitive and enjoyable to use. And from a business perspective, better alignment around a core purpose and vision leads to faster growth and higher profits.
So, where do you start once you’re ready to cut?
Removing a feature or full product requires balancing your expertise with user research, product vision, data, and gut feelings. While sunsetting projects or deprioritizing features you’ve been working on can be an overwhelming process.
To make it easier, let’s break it down into a few actionable steps.
1. Look for features that fall into the 8 ‘can we kill it?’ scenarios
The first step of unshipping is to identify live features or products that you should cut.
While you’ll probably already have a good idea of which ones are problematic, it can help to have a list of qualities on which to judge them.
Typically, there are eight scenarios when you should seriously consider removing a feature:
- Strategic alignment: Does this product or feature align with your long-term vision? Has your vision changed since you shipped it?
- Buggy: Is it hard to use, or does it underdeliver value to customers?
- Novelty: Was this feature once exciting but now has diminishing returns?
- Niche: Is it only being used by a small group of users?
- Obsolete: Does this feature no longer provide a solution to a customer problem (or solve a problem they don’t have anymore)?
- Redundant: Does it still fit in your overall product ecosystem? Or is it competing with a better, newer product?
- Incompatibility: Does it work with your product?
- Tech costs: Has it become too expensive to maintain for the return you’re getting?
It’s best to start at the top of this list and work downwards. Revisit your product strategy and use it as a lens to view all potential features or products on the chopping block. Then, go through your list and ask if the features or products fall into any other categories.
2. Dig into product usage data
Unshipping starts with your connection to products and features. After all, you know what’s coming in the pipeline and how you want users to see your overall product vision.
However, this human element of unshipping can hold you back. Think of it as cleaning up your closet. We all have a sweater we haven’t worn in years, but that has some sort of sentimental value that won’t let us throw it out.
That’s why it’s important to pull from other data sources before you start making a phase-out plan.
Product data can be a powerful source of ‘end of life’ signs that you might otherwise miss or ignore. Here are a few you can check:
- Usage metrics: Are users still interested in this product or feature? If usage stats are dwindling, this feature is a good candidate for either unshipping or updating.
- Sales metrics: Is your sales team pushing it? For example, how often are they closing deals that include this feature? This can help you differentiate from a product or feature that’s dead versus one that’s still valuable but in a slump.
- Support metrics: Is it taking time away from your support and development teams? Think about the opportunity cost as well. What’s the value you can unlock by re-deploying resources to a more successful feature?
3. Conduct user research to find out why features aren’t being used
Not every product with low usage should be unshipped. Instead, you might have just missed the mark or done a poor job onboarding new users, explaining features, or pushing it to market.
Data can only show you what features aren’t being used. You need to talk to users to understand why they’re not using them.
Set up a few user interviews to understand your users’ goals and how your features or products are missing the mark.
Sometimes you’ll learn that users want something completely different. Other times, you’ll discover they’re simply confused by it or unaware the feature even exists!
4. Segment your user base to look for patterns
On the other hand, you might end up with regularly used and sold features that you should still remove because they don’t align with your larger vision and goals.
The most common scenario is that you’ve built something that’s attracting the wrong users. For example, if small companies are only using a feature but your goal is to work with Enterprise-level ones, it might be time to sunset that feature.
Segmenting your user base can help you find patterns in usage. Use the data available to you to segment by:
- Company size
- NPS score (i.e., is this feature used mainly by NPS detractors?)
5. Set up triggers to help you revisit past decisions
A final source of features and products to deprioritize or unship is in your past decisions.
Every new feature or product was the result of a decision. As a product manager, you make hundreds of them a week, and it’s impossible (and unproductive) to question each one. Instead, you can set up specific triggers to help identify when you should revisit a decision.
Here are a few simple triggers to start with:
- Knowledge-based triggers are when something in the market has changed your assumptions that led to a decision. For example, if a competitor comes out with a better offering or some industry change makes a feature obsolete or outdated.
- Metric-based triggers are when you use your own analytics to see if usage has dropped beneath a certain threshold. For example, if a once-popular feature stops being used as much.
- Time-based triggers aren’t based on a change but are scheduled moments to reflect on past decisions. For example, you might set aside time each quarter to review decisions or look at feature performance during a sprint retrospective.
These triggers can help you keep on top of features that could be unshipped. As a bonus, regularly revisiting past decisions like this is a great way to improve your overall decision-making process.
6. Refine your product roadmap and deprioritize features
At this point, you’ll have identified several public features that you can sunset and developed a clear process for finding more. However, it’s significantly easier to deprioritize a feature before shipping it than it is to roll back once it’s live.
Unfortunately, changing your established roadmap can be a difficult pill to swallow.
The last thing you want is for your team to feel like they’re wasting their time or following a misguided leader. Yet, you need to deprioritize features or tasks when your vision or market factors change.
Start by grooming your current product backlog.
A typical backlog grooming session takes place before you plan a sprint and involves re-prioritizing and updating user stories. However, this session will go much deeper into those tasks and issues that have been lingering.
A project management tool like Planio is essential for keeping your backlog organized and up-to-date. Planio lets you create custom issue categories to organize tasks so you can quickly dig into the features, products, or sub-categories that are on your radar.
Approach your backlog with the same eight scenarios you used for public features (i.e., ask if they’re strategically aligned, focused on a niche audience, based on novelty, etc.) Add a special tracker to these issues so you can identify them right away and pull them all up at once.
Then, plan a longer sprint retrospective to go over what’s currently being worked on and discuss how it fits into your overall strategy.
Changing an established roadmap or deprioritizing features your team was excited about is challenging. It’s important to include them in the conversation, so everyone understands what’s happening and the reasoning behind it.
Start this meeting by grounding everyone in your product vision and strategy. Then, bring up the items from the backlog and ask your team to suggest others either planned or in-progress, that should be added to the list.
Finally, adjust your next sprints to reflect the new plan. In Planio, you can drag and drop issues into a new sprint with one click.
7. Create a phase-out plan for features you’re going to remove (and deprioritize)
Now that you’ve identified what features or entire products should be unshipped and deprioritized, it’s time to put those plans into action.
There are two aspects to a phase-out plan: the internal process (i.e., how your team handles the change) and the external process (i.e., how your customers learn about and handle it).
When it comes to announcing to users that you’re sunsetting a feature, you’ll want to take a coordinated approach. Here are a few steps to follow:
- Talk to stakeholders first. Sunsetting a product isn’t your decision alone. It can be a large strategic decision that requires approval from key stakeholders. The data and research you’ve collected will help make your case. But the choice needs to come from above.
- Work with sales, marketing, and support. Removing a feature is a team effort. Make sure to connect with other team leads to talk through how it will impact them and get their input. Sales will need to stop selling. Marketing will need to remove it from your website and ads. And support will need to know what to tell customers.
- Tell your existing customers: Your users don’t like surprises. Especially when that surprise is removing a feature they’re grown attached to. Create a list of whereyou will communicate the removal–in-app, on your site, on email, social media, etc. Communicating the change is essential, and there are a few best practices you should follow:
- Be transparent about why you’re removing a feature. Show that you respect your users by telling them exactly why you made this decision. Don’t sugarcoat it behind marketing speak.
- Give appropriate time to prepare. A good rule of thumb is to take what you think is a reasonable amount of deprecation time and leeway and double it. For large products or features, this could mean messaging users 6–12 months before the change. Give a clear timeline and continue to message them as you get closer to the deadline.
- Provide a substitute or guide them to a new workflow. If you have a new feature that should work, help them migrate over. If not, suggest an alternative (even if it’s a competitor!)
- Remove the feature from all marketing and sales materials. Once you’ve decided to phase out the feature, new users shouldn’t know it exists. Help desk can stay live for the time being, but make sure you highlight that they’re for a product that no longer is being supported.
- Know where the freed-up resources will go. Don’t wait until after removing a feature to decide where those development and support resources will go. Give your team a new and exciting direction to head in. They’ll forget all about that old feature you just retired.
While it might seem like the external aspect of sunsetting a feature is more challenging, it’s usually the internal one that faces a bigger pushback. Momentum is a powerful force. And changing your team’s direction requires a delicate touch.
Here are a few best practices you can follow:
- Decide who has the final say. Roadmaps aren’t made in a silo. The features in the works most likely came from cross-team collaboration and company-wide feedback. This means that deciding how to deprioritize them can get tricky. Instead, make sure a specific person is designated who has the final say.
- Be transparent internally. Respect your team the same way you respect your customers. Be open and honest about why the change is happening and what led to it.
- Create a change management plan. In a survey of over 26,000 workers, 49% said their company lacked the focus or capacity to change current processes and plans. To help your team tackle added uncertainty, create a clear plan of action for the next steps.
8. Don’t get stuck in limbo
Finally, don’t forget the ‘courage’ aspect of sunsetting features.
Making the choice to remove a feature (either public or in-progress) is only half the battle. The real work comes in going through with it.
You may think that leaving an all-but-retired product live isn’t going any harm. However, being stuck in a limbo state causes tension among your team (i.e., why are we still working on this?) and can result in even more technical debt as you compromise new features to maintain one that should be cut.
Instead, move quickly and confidently once you’ve identified a feature that needs to go.
Making the choice to remove a feature is only half the battle. The real work comes in going through with it.
9. Track your results
Every decision is a risk. Even if you’ve done your due diligence, there’s no guarantee that phasing out a product or feature is the right move. That’s why it’s essential to have a few key metrics to track to see the results of your change.
The goal of unshipping is to get your product more aligned with your overall strategy, so your overall KPIs are a good place to start. However, you can dig deeper into some specific metrics that tell you how your users react to the change.
For example, if you offered an alternative solution, track adoption and usage rates. You could also return to your user segment data to see if your ideal customers are more actively using your product. Finally, you can reach out and conduct user interviews to hear how the customer experience has changed or improved.
Metrics and feedback not only help put your mind at ease, but they also provide quantifiable proof that more isn’t always better. If project stakeholders are nervous about removing features (and they will be), this can help everyone feel more confident.
Finally, don’t forget to recognize and reward removing features
One of the most significant obstacles that holds teams back from unshipping, deprioritizing, and sunsetting products is that few people recognize the work it involves. Most companies prioritize short-term growth. While removing features is part of a long-term plan.
As Neil Rahilly, VP of Product and Design at Mixpanel, explains:
“Joining an All Hands meeting to demo a new feature is always going to feel better than coming to the meeting as the grim reaper saying, ‘We killed this feature.’ To change this culturally, you really need to work hard at emphasizing quality throughout the user experience.”
One of the best ways to start celebrating feature removal is to attach metrics to it. A few of the most visible metrics that your unshipping efforts will impact include:
- Decrease in support tickets and incidents for engineering teams
- Lower data center costs due to a more streamlined app experience
- Increased retention from customers who get a better experience
- Higher NPS scores (Mixpanel reported their NPS score tripled in 18 months after shifting their product focus from quantity to quality!)
- More organic acquisition through word-of-mouth referrals
Don’t let the long-term benefits of deprioritization go unnoticed!
Sunsetting a product may feel like the end, but the sun always rises
Apple wasn’t lying when they said removing the headphone jack from their newest phone took ‘courage.’ But we all know what happened once the online rabble quieted down.
By the end of the first quarter of 2017, the iPhone 7 had sold nearly 40 million units and was the best-selling smartphone in the world!
Just like it takes time to decide what to build, you should be purposeful and careful when removing features or products. Follow the guide above to uncover features and products you can prune and then swiftly make your cut.
There might be some momentary discomfort, but you’ll be setting your product up for the long term.