There have been some recent press about the computer game Fortnite that has caused concern for parents of children that are spending excessive time playing due to its seemingly addictive effect on players (many of whom are young children).

Most of the design patterns and interactions built into games and applications with a focus on entertainment is aimed at increasing engagement, which is often inferred from the amount of onscreen time. While it can be easier to work out the difference between engagement and addiction using more qualitative methods, the behavioural patterns between an engaged vs. addicted user might be harder to determine using quantitative methods.

Is it assumed that engagement precedes addiction, and if so then is it possible to measure the behavioural change over time to identify at risk users?

If engagement and addiction are completely different behaviours, what are the differences that allows you to distinguish one group from the other?

For those that want to look more into the psychology research aspects of this, a similar question has been posted on Psychology and Neuroscience SE: https://psychology.stackexchange.com/questions/20981/how-is-addiction-distinguished-from-engagement-when-making-a-determination-about

  • I doubt there is any metric for this, because addiction is a binary state: either you have an addiction condition, or you do not. There is no threshold from which it can be defined that a behavior is characteristic of an addiction or not. That is why in the treatment of addictions the person is always aware of this condition and never denies it. Quite the contrary, the person is in a constant treatment or therapy (either idle or active)
    – Devin
    Oct 23, 2018 at 22:18
  • on a side note, this question belongs to psychology.stackexchange.com
    – Devin
    Oct 23, 2018 at 22:21
  • @Devin don't you think that addiction could be seen relative as well? turning the question from "is x addicted" to "how addicted is x" where is 0 = not addicted at all, i do agree with the part that there isn't a metric for it, not that i know for sure but i cannot see a metric working on subjective matters
    – UX Labs
    Oct 23, 2018 at 23:13
  • and the last note, this is what i've been debating for the past month or so, there are so many questions that belong to psychology, sales, and other fields that do intersect with ux, does it mean these topics don't belong here? i would like someone to settle this matter once and for all so that we know how ux is perceived in this community
    – UX Labs
    Oct 23, 2018 at 23:14
  • 1
    @UXLabs, there was quite some debates about this subject through the years. The general consensus always was to move questions to the proper SE sub-site, which happens on a daily basis. I didn't take part on those discussions and I don't remember the arguments to keep it that way, but I can think of 3 reasons: one, because it helps sub-sites grow (specially those in Beta that needs to be promoted, such as Psy.SE); 2: because you'll find more experts on those sites; and 3, because, as you say, almost everything intersects with UX, which doesn't mean everything is UX
    – Devin
    Oct 24, 2018 at 18:03

2 Answers 2


It should be possible to come up with come up with an amount of "engagement time" that is objectively a cause for concern. Although people have different lives and responsibilities, there is only a limited amount of time in a day. We can start there and work backwards. Addiction usually means that other behaviors needed for living a healthy life are adversely impacted. If we subtract the essential time needed for other activities that allow us to stay physically and mentally healthy and functional in society:

  • Sleeping
  • Eating
  • Work / school commitments
  • Social / outdoor physical activities (we are social animals)

Then we can begin to converge on a maximum amount of "engagement" that, if crossed, is statistically very likely to indicate an addiction problem. That is not to say that those with less engagement time than this do not have detrimental effects, but it would provide a good rule of thumb. This could also be tweaked to tailor it to the target demographic (e.g. children of school age).

We can also look at the time of day that this activity is taking place, and infer from this if the engagement times create a pattern that is clashing with times that the user should probably be doing something else. E.g. a school child with 1 hour of engagement time at 3 AM is likely to have an issue.

Another interesting metric (particularly for games/apps available on a mobile device) is how often users just "check in" to an app to see if there are any updates. True addicts will check in regularly even if they get notifications that mean they don't have to open the app to see if something new happened. Addicts are also more likely to respond instantly to notifications, no matter the time of day.

All of the above can used to infer the likelihood that activity required for a healthy life is being displaced by engagement with your app.

  • +1 This is a very good approach/strategy for establishing a baseline and working out the likelihood that someone is addicted. However, I think that getting that 'pure' baseline is very difficult because we are already preoccupied with many distractions these days.
    – Michael Lai
    Oct 26, 2018 at 22:35
  • 1
    @MichaelLai I think that applying the above is very likely to uncover other useful information that we haven't thought of yet. In tech I have found that the mindset tends towards believing that we can converge to a solution, but this only works with some problems. Sometimes you just have to strike out on a different path, with no expectations of what you might find along the way.
    – Franchesca
    Oct 29, 2018 at 8:13
  • And I would agree with that, hence I accepted it as the answer, but awarded the bounty because I think as a starting point it is important to consider what we know before tackling what we don't know :)
    – Michael Lai
    Oct 29, 2018 at 12:10

From Wikipedia Customer Engagement:

Customer engagement is a business communication connection between an external stakeholder (consumer) and an organization (company or brand) through various channels of correspondence. This connection can be a reaction, interaction, effect or overall customer experience, which takes place online and offline.

Or more in digital products context, user engagement it is a metric that reflects the amount of interaction the user performs with a product within a certain time period.

Does this necessarily mean that user engagement would lead to addiction? The short answer is No.

Take this horrible scenario, if someone points a gun towards one user - and i hope this will never ever happen - and threatens the user to shoot if the user does not use the product, the user will likely use the product, the engagement metric will likely increase, the user will very unlikely be addicted.

The purpose of this scenario is to present a case where engagement increases and addiction doesn't occur making it impossible for them to be be mutually inclusive.

From psychology today

Addiction is a condition in which a person engages in use of a substance or in a behavior for which the rewarding effects provide a compelling incentive to repeatedly pursue the behavior despite detrimental consequences.

So what about mutual exclusivity, it is also impossible since by definition of addiction, repetition of usage will also push up the user engagement metric, therefore the more addiction the more engagement.

So this is how it would look like:

enter image description here

From this breakdown we can reach to the following; addiction is not the only way to increase user engagement, but whenever addiction occurs user engagement increases.


If we refer back to Addiction definition, i will quote this part

the rewarding effects provide a compelling incentive to repeatedly pursue the behavior

This quote can be rendered as a tactic where if implemented effectively it can drive addiction therefore drive user engagement - and i cannot think of a single case where i would consider it a moral practice so please be my guest if you have other views in the comments.

And to answer the last part of your question

If engagement and addiction are completely different behaviours, what are the differences that allows you to distinguish one group from the other?

We cannot say they are separate behaviors, as one is a behavior and the other is a metric, but if you want to measure the addiction caused by your product the answer is also is part within the definition of addiction

You would want to measure to which extent is the user going to

repeatedly pursue the behavior despite detrimental consequences.


To answer

How do you define what a detrimental consequence is for the user?

It is a dillema

When trying to answer how do we decide on what are considered to be "Detrimental Consequences" and what are not, we often find ourselves back to the subjectivity dilemma, what we agree upon others disagree upon and vica versa.

Just as any other ethical dilemma, there is no definite answer rather opinion, this how i would handle it:

  1. The Law (the obvious)
  2. Code of Ethics (the underrated)
  3. Understand your users (the classic)
  4. The Experts (..and the doctor says no more monkeys jumping on the bed)
  5. You will not always succeed (the reality)

I just would like to elaborate on a couple of points of them:

Code of ethics

Often organizations don't understand or underestimate the value of having a code of ethics, this is a almost-framework that can correct such controversies, and can play the role of guiding judge in concluding uncertain decisions.

Understand your users

We always say understand your users especially in human-centered fields, to the extent where "ask your users" is almost becoming a cliché, but here in this sense it is about trying to understand what is perceived good vs bad by your users what are their cultures, beliefs, and attitudes which would support your decision.

You will not always succeed

It is important to be realistic that what is right and wrong, good and bad are all subjective matters the answer lies in the perspectives of whom you ask which can differ and conflict with others, there is no one rule to rule them all, no matter how hard we try we will fail at some point, but what matters is that we actually try.

  • 1
    +1 I like the way you have dissected the question, but it leads to another interesting question: "how do you define what a detrimental consequence is for the user?"
    – Michael Lai
    Oct 20, 2018 at 1:54
  • i'll update the question this evening to share my perspective of it
    – UX Labs
    Oct 20, 2018 at 10:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.