The Product Manager's Guide to Non-Addictive Design

For a Product Manager, data is the difference between flying blind and seeing clearly. It strips away our assumptions and shows us exactly how people interact with what we’ve built. We have dozens of ways to measure this, but there is one metric that we all chase: the one that proves users are finding value. 

That metric is engagement.

For many businesses, that’s because engagement has a direct impact on revenue: think about every application that gathers and sells your data and every website that shows you ads. Increasing engagement does increase value - from the business’s point of view. Whether or not the product user is getting much out of their engagement might not be considered at all.

But it’s not just user-hostile businesses; we all fall into the trap sooner or later.

When we build products to maximize time-on-platform, we’re doing so at the direct expense of our users’ time, attention, and overall quality of life, and they're starting to notice. Users are tired of being manipulated into spending more time than they want or need on their apps and devices. A whole cottage industry has emerged around helping them disconnect from devices, ranging from apps like minimalist and Brick to timed lockboxes. These tools are useful, but they’re an abdication of responsibility by the organizations that build these products and devices. 

It’s time - past time, really - for product development teams to take responsibility for the impact of their work beyond the metrics, and find ways to foster sustainable engagement that is genuinely good for our organizations and our users. 

What makes design addictive?

I choose to believe that most product development teams aren’t intentionally creating addictive products, but our benchmarks for how our products should function need recalibration. 

We have adopted certain design patterns simply because they are industry standard, and they feel familiar and expected, without realizing that these conventional features are often behavioral traps:

• Variable reward schedules. When you never know when your next reward is coming, it’s really hard to stop paying attention to the reward source. This is why it’s so hard to stop checking your email or social media apps - your brain is primed to encourage you to keep checking back for that new message, like, or comment.
• Infinite scroll. If every time you reach the bottom of the page, more content is served to you, there’s no natural place for you to check in with yourself to decide whether you want to keep reading or not. This means it takes more willpower to walk away. 
• Noisy notifications. Notifications can be both too numerous and too invasive when they’re sent in the application, via email, and pushed to your phone or wearables like a smartwatch. The combination makes it hard to disconnect from the app. 
• Guilt-based gamification. At its best, gamification rewards healthy engagement and encourages users to achieve a meaningful goal. In some products, rewarding users for returning is relevant (e.g., when there is a learning component where repetition matters), but it can create real anxiety for users who want or need to take a break and fear losing status if they break a usage streak.
• Algorithmic content streams. When used to serve inflammatory or sensational content, algorithmic content might maximize the time users spend in the product, but at the expense of quality, value, and well-being. 

Many of these elements are well-intentioned. Infinite scrolling makes things frictionless. Notifications inform you when something you may find interesting happens. Algorithms can help match you with content that is interesting or meaningful to you. We don’t need to necessarily ban all these elements outright, but as product developers, we should advocate for design choices that make it safe and easy for users to step away, and help our businesses find other, healthier measures of success for our products. 

Measuring success beyond time-in-app

If you’re not swayed by the humanist ideals, there are practical reasons why this should matter to you and your business. Folks are becoming more aware of how technology companies profit from sapping our attention and of the ways these design choices compromise our mental health. This is leading to increased scrutiny of these companies, both in terms of their reputations and trust, and in regulatory action. To protect your users and your bottom line, it’s time to find a better measure of success.

If we’re not meant to focus solely on increasing the amount of time people are spending on our apps, what should product managers be measuring? The specifics will vary, but here are a few ideas we’re exploring at Discourse:

• How might you measure the quality of time spent in your product? Is time spent scrolling a content section as valuable as time spent reading an article? Is there a “sweet spot” of time that you might target, which shows that someone interacted meaningfully with the product, but not obsessively?
• Why do users use your product? Are you helping them meet that goal? There’s usually an underlying reason behind someone’s behavior. When someone opens your app, what are they setting out to do? What can you do to help them reach that goal, and then safely disconnect once it’s done? 
• How do your users feel after they use your product? This can be tricky to measure, but at least starting with this in your mind as a design goal can be very useful. I imagine most product managers don’t want people to feel stressed out or depleted after using their products - so how might you design your product to actually elicit the more positive feelings that you do want for your users?

Designing for sustainable, empowering engagement

Judicious notifications 

We disable automatically, so users must opt in. We give users maximum control over what triggers notifications, and our default settings are quite conservative. If users want lots of notifications, it’s possible, but they have to explicitly choose it. Discourse makes it easy to pause notifications for a short time (in fact, I’m using that now so I can focus on this writing!) or set up a notification schedule to keep things quiet in the evening or on weekends. 

Create natural stopping points

Think about how you might design natural stopping points where a user can check in with themselves and then decide whether to continue using your app or leave it. 

On Discourse, new content feeds have clear endpoints and can be further customized to limit what content users might feel they need to catch up on. We also provide affordances like bookmarks, mark as unread, and reminders so that users can leave the app safe in the knowledge that they can pick back up where they left off next time. 

Design for asynchronous participation

When possible, let your users interact with your product at their convenience, without FOMO or guilt.  

Discourse is designed with asynchronous communication as a core tenet. If you don’t see content as soon as it’s posted, that’s no problem - we give users a variety of ways to find meaningful content of any age, such as search, filtered lists, and settings that control when a topic is considered new. We also don’t reward synchronous interactions (e.g., gamifying the “first like” on a new post). 

Content discovery that respects attention

Content feeds are black boxes that users can neither understand nor influence directly. Respect the valuable time and attention your users spend on your product with transparency and control. 

Discourse gives communities topic lists that show content by category, depending on whether the user has seen it. End users have the ultimate control over which topics they want to read, get notified about, or mute entirely, so they can prioritize what’s interesting over what could be engaging, but ultimately unfulfilling. 

Intentional friction

It sounds counterintuitive, since we work so hard to make our products easier to use, but a little friction can be a good thing. We don’t want users to struggle, but sometimes slowing them down can lead to a better overall experience. 

At Discourse, we do this with features that slow down posting to encourage considered responses, particularly in heated topics. Our drafts system makes it easy for users to slow themselves down, too, so they can take time for considered responses without worrying about losing their work. 

It's up to us...

Start with an honest audit of where your product is today. Interrogate your product and business metrics and identify opportunities to redefine engagement so your business’s growth doesn’t come at the user’s expense. 

Engage your users to understand how your product fits into their lives, or where it might be taking on an outsized role in their day. From there, share your findings with stakeholders and teammates to help them understand your product's current situation and the viable alternatives. Wherever possible, prioritize changes that restore user agency (or build it in from the get-go) and encourage sustainable engagement with your product.

There are plenty of successes in products you’re already using: forums that thrive asynchronously, newsletters that come less often but are read with more enthusiasm than endless feeds, productivity tools that support work without dominating it, and learning platforms that are built around completion, not consumption. These products succeed because they embrace the obvious truth that people want better outcomes, not more software. 

As product builders, we ultimately decide whether we’re in the business of mining for attention or creating real value for our users. 

It's up to us...