I faced a similar conundrum when I started playing. You have a great number of options at your disposal for handling this however. I'll list a few so you can see what I mean.
First you need to decide for yourself if your group is actually having issues with this. Some games might have many opportunities to improve, there isn't a right or wrong way to do it. Shorter stories often will necessitate faster improvement, whereas a slow long term game might need something a bit more drawn out.
The first common way to handle this is noted above. Simply restrict the amount of exp gains per scene. It works, although it also removes the incentive for that player to behave in more specifically hard/hot/cool/sharp/weird rolls. What I mean by that, is that if they have two stats marked, they can pick the same one each scene and ignore the other entirely while continuing to maximum their exp gains per session. Not that it's really all that problematic if they do so.
The second way to handle it is to have rolls mean more individually. For example, instead of having a player declare that they want to rush up and take the gun from the manic (roll seize by force), succeed you narrate that fiction, and then they declare they want to beat them to death with the butt of the gun (roll seize by force again). You could instead have them say their full intention, like I rush the guy, keeping out of his line of sight using the debris for cover until I'm close, then I leap out and grab for the gun, intending to beat him into the concrete with it. (roll one single seize by force roll). Not the best example, but the concept can be applied much more broadly. Many systems have round by round combat, and that instills a sense of micromanaging the action. AW is best when the action is full, meaningful, and the consequences move the scene forward in larger ways. Attack Hit Damage is not a recommended mindset (though extremely common).
Another option you have is simply increasing the number of experience required to gain every level. There is no rule saying 3 or 5 or 10 is more right then another. It should be whatever fits best within your group, the narrative's style, and the intended duration of the games. Longer game sessions (for those hardcore 10 hour session gamers) might really need a spike in exp in order to give the story a sense of time to developed before every box ends up checked.
Finally you can do a gradual experience, such as five bubbles for the first improvement, and then each time you improve the character add another bubble. Lets people with less keep pace, gives players faster essential moves for their concepts and personal desires, while still preserving the duration in a longer game. I've been testing this one more often then the others, and found it fits with my groups expectations of the game's improvements, while still making them really conscious about how they want to use all of those highlighted stats.
Anyhow those are just some thoughts. Hope it inspires you to find your own solution.