My kids are on screens most of the day and I am not even sure what counts as positive use anymore. Watching videos, gaming, scrolling — it all blurs together. Is there a way to actually shape how they use technology rather than just limiting time? Would love some practical thoughts. 
First thing I track is the ratio of consuming versus creating. 
If a kid spends 90% of screen time watching and 10% making something, that split is worth addressing. Doesn’t matter what the total hours are as much as what those hours contain. Typing, drawing digitally, editing a short clip — anything where they are producing shifts the balance fast.
That consume vs create framing is interesting. What I noticed with my own kids is that creation only happens when there is no pressure attached to it. 
The moment it feels like a task, they disconnect. Leaving tools around (a simple video editor, a drawing app) with zero expectation tends to pull curious kids in on their own timeline, not yours.
I set one rule and stuck to it: no passive screen time before 5pm on school days. 
Passive meaning YouTube, reels, anything you just sit and watch without doing anything. Gaming counts as active if there is problem-solving involved. Reading online counts. That single boundary did more than any screen time limit app I tried. Simple line, clear outcome.
TripodMax, that passive versus active split is worth comparing across age groups though. 
For younger kids (6 to 9), almost any screen time needs to be supervised or at least structured. For 10 to 13, self-directed active use becomes more reasonable. The same rule applied to both groups tends to either over-restrict younger ones or under-guide older ones. Context does matter here.
Building a structured framework for this:
Tier 1 (Always acceptable): Coding, writing, digital art, educational content with output
Tier 2 (Time-limited): Gaming with social or cognitive elements, structured video content
Tier 3 (Restricted): Passive scrolling, autoplay content, no-goal gaming
Assigning apps to tiers and setting device rules around those tiers takes about one afternoon to set up. 
RigidDatum’s tiers are solid but I want to add one thing: kids push back hard when the system feels imposed on them. 
Letting them help sort the apps into categories made a surprising difference in our house. My son argued that Minecraft belonged in Tier 1 for twenty minutes and honestly made a pretty good case. Ownership of the system changes how seriously they take it.
Nobody is talking about the elephant right in front of us, which is that most parents are also on their phones constantly. 
Kids are watching. If the rule is no screens at dinner but you check notifications between bites, the rule is already dead. Modelling the behaviour you want is less exciting to talk about than apps and frameworks, but it is doing most of the actual work.
GorillaBlink is right and it connects to a technical setup worth doing. 
Most routers now let you create device schedules and usage reports per device. Set this up once: your router logs screen time, you review weekly, kids know the data exists. It removes the argument from the conversation because you are both looking at the same numbers rather than guessing.
The underlying model here is habit formation, not restriction. 
A habit needs a cue, a routine, and a reward. If positive digital behaviour (coding, writing, learning) has no consistent cue to trigger it and no satisfying reward after, it will not stick regardless of rules. Designing the environment so that the good habit is the path of least resistance works better than willpower-dependent restriction.
Pulling the threads together so far: structure matters (RigidDatum), ownership matters (SofterWorld), modelling matters (GorillaBlink), environment design matters (Primeset). 
None of these conflict. They are just different entry points. Most parents try one and give up when it does not work alone. Combining two or three of these at the same time is where the actual change tends to happen.
Forget gradual. Set the system up in one weekend and run it for 30 days straight. 
Week one is resistance. Week two they start adapting. By week four it is just how things work at home. The parents who report failure usually tried something for a week, saw pushback, and adjusted. Consistency in the first month is the whole game. Adjust after you have data, not before.
DroidPro is right about the 30-day window. One thing I added in week two was a Friday review. 
Five minutes, we go over what they used that week and whether anything should move between categories. Keeps it from feeling like a permanent sentence and gives kids a legitimate way to make a case for something. Has worked better than I expected.
The Friday review idea is good and pairs well with something I tested: letting kids pick one new app per month to trial in a higher tier. 
If they can explain what they will do with it (not just play it but what they will make or learn), it earns a two-week trial. Most kids rose to that bar once they understood the criteria.
All of this is practical and worth trying. The thing I keep coming back to though is that kids read the atmosphere around technology more than the rules. 
If screens in your home carry anxiety or conflict, that is what they absorb. If screens are treated as tools with clear roles, that settles in too. Rules work better when the general feeling around them is calm rather than combative.