A habit runs on autopilot and stops fairly easily once you decide to stop it. A compulsion resists the decision, comes back with a pull you have to sit through, and keeps going after it has started costing you. The cleanest test is a two-week attempt to stop. What happens in those two weeks tells you more than any checklist.
You do not need a diagnosis to use this, and for most behaviours there is not one available.
Why you are asking at all
Nobody looks this up about something that is fine. The question usually arrives with a specific memory attached: the hour you lost, the money you cannot account for, the thing you said you would not do again on Tuesday and did on Wednesday.
It also arrives with a fear of overreacting. Everybody does the thing you do, to some extent. People check, buy, scroll, pick, tidy, watch, eat, work, count, plan, reread, refresh. So there is no obvious line, and without a line you end up either dismissing it or catastrophising it, sometimes both in a week.
The four questions below give you a line that does not depend on what the behaviour is.
The four questions
Read each one against the specific thing you do, not against the general category.
1. Who decides when it starts and stops?
A habit starts when the cue arrives and ends when the task is finished. You put your keys in the same place, you bite your nails during a meeting, you check your phone on the bus. It runs without you and it also ends without drama.
A compulsion starts on its own and ends when something interrupts it. The battery dies, someone comes into the room, the shop shuts, it gets light outside. Ask yourself when the last five episodes ended and what ended them. If the answer is almost never "I decided to stop", that is the answer.
2. Has it moved up the list?
Habits sit alongside your life. Compulsions push things out of the way. Sleep first, usually. Then meals, then people, then the things you used to want to do and now say you have no time for.
The useful version of this question is not "how many hours". It is what you have quietly stopped doing since it started.
3. Has it continued after it started costing you?
This is the strongest single question, and it is why the amount of time matters less than people think.
Somebody who spends two hours a night on something that costs them nothing is doing a hobby. Somebody who spends forty minutes on something that has cost them money, sleep, a relationship or their skin, and who has not been able to stop despite knowing that, is describing something else entirely.
You already know what it has cost. Write it down rather than keeping it vague.
4. What happens when you try to stop?
The test rather than the question. Stop for two weeks and watch.
A habit: mildly annoying for two or three days, then largely gone. You forget about it.
A compulsion: restlessness, irritability, a pull that arrives in waves and gets stronger before it fades, bargaining with yourself, a period around day three to five that feels disproportionate, and a strong urge to conclude that this was never a problem so you can stop testing it.
That last one is worth naming. The most common outcome of a two-week test is not failure. It is quietly deciding halfway through that the question was silly.
The scaffold behind those questions
They are not made up. They are the shape the World Health Organization uses to define gaming disorder, which is one of only two behaviours with addiction status in the international diagnostic manual: impaired control, increasing priority given to the behaviour over other activities, and continuation or escalation despite negative consequences, along with significant impairment and a pattern that would "normally have been evident for at least 12 months".
None of those criteria mention gaming, which is why they transfer to whatever you are doing.
What counts as a diagnosis, and what does not
Be clear about this, because you may well be told your problem "is not a thing".
Only two behaviours have addiction status in ICD-11: gambling disorder and gaming disorder. In the UK, gambling is the only behaviour with its own NICE guideline. The American Psychiatric Association says it about as plainly as it can be said, about the gaming proposal in its own manual: "This proposed condition is limited to gaming and does not include problems with general use of the internet, online gambling, or use of social media or smartphones."
So compulsive shopping, checking, picking, watching, working, eating, scrolling and everything else with no name at all has no diagnostic status. Researchers are also actively arguing about whether it should, and that argument is worth knowing: a 2015 paper by Billieux and colleagues warns that "recent publications have suggested that nearly all daily life activities might lead to a genuine addiction", and that loose research "may result in the identification of an unlimited list of 'new' behavioral addictions".
That is a fair warning about how disorders get named. It says nothing about whether your life is being eaten.
The NHS covers the gap in one sentence: addiction is "not having control over doing, taking or using something to the point where it could be harmful to you", and "it's possible to be addicted to just about anything". The absence of a diagnosis is not the absence of a problem.
Three answers that are not the two in the title
Worth checking before you settle on compulsion.
**It might be anxiety.** Checking, rereading, counting and reassurance-seeking that is driven by dread rather than by pull looks similar from outside and is treated differently. If the behaviour is mostly aimed at preventing something bad rather than getting something good, say that to a GP in those words.
**It might be a coping strategy that is currently load-bearing.** Sometimes the behaviour is holding down grief, pain, a bad relationship or a period of genuine crisis. Removing it without addressing what it is holding tends not to go well. That does not mean leave it, it means expect to need something in its place, which is what [replacing it with something real](/guides/other/replacing-it-with-something-real) is about.
**It might genuinely be fine.** Some people arrive at this question because someone else disapproves. If the behaviour costs you nothing, has displaced nothing, and stops when you decide it stops, then the honest answer is that it is a habit and somebody else's opinion is the problem.
If the answer is compulsion
The label does not change the first move, which is why you do not have to be certain. [Four moves that work whatever the behaviour is](/guides/other/how-to-stop-a-compulsion-you-cant-name) is the starting point, and the fastest of them is friction: putting steps between the cue and the act while you are calm.
For a compulsion with no name, SMART Recovery UK is the most natural route, because its meetings are scoped to problem behaviours generally rather than to one substance. Meetings are free and run online and in person, on 0330 053 6022. A GP is the other route, and a precise written description of the behaviour is what makes that appointment work.
Renovyn lets you track a compulsion you define yourself, which helps when nothing on a standard list matches. It is a tool for keeping count, not treatment.
Common questions
How much is too much?
There is no hour count that settles it, for any behaviour, and anyone quoting you one is inventing it. The measures that mean something are control, priority and continuation despite cost. Forty minutes that you cannot stop matters more than four hours you chose.
Does it count if I can stop for a week when I really try?
Being able to stop with a large effort is not the same as being able to stop. The question is what the week costs you, whether the pull is still there at the end of it, and what happens in week three when you are tired and nobody is watching. Many people can do a week and very few can do it without noticing.
Nobody has a name for what I do. Does that make it less serious?
No. Only gambling and gaming have behavioural addiction status internationally, which reflects how slowly diagnostic manuals move and how contested the research is. It says nothing about how much of your life the behaviour is taking, and the NHS explicitly says it is possible to be addicted to just about anything.
What if two people in my life disagree about whether it is a problem?
Then use the four questions rather than the argument. Other people can see the cost and cannot see the control, so they tend to be right about consequences and wrong about how it feels from the inside. You are the only one who knows what happens when you try to stop.
Sources
- [NHS, Addiction: what is it?](https://www.nhs.uk/live-well/addiction-support/addiction-what-is-it/). The control definition and the statement that it is possible to be addicted to just about anything.
- [WHO, Addictive behaviours: gaming disorder](https://www.who.int/news-room/questions-and-answers/item/addictive-behaviours-gaming-disorder). Impaired control, increasing priority and continuation despite consequences, plus the twelve-month duration.
- [American Psychiatric Association, Internet gaming](https://www.psychiatry.org/patients-families/internet-gaming). That the proposed condition covers gaming only, and not general internet, social media or smartphone use.
- [Billieux et al., Are we overpathologizing everyday life? A tenable blueprint for behavioral addiction research](https://pmc.ncbi.nlm.nih.gov/articles/PMC4627665/). The caution against naming an unlimited list of new behavioural addictions.
