It seems that it takes more than one shot with the laser. The article stated that the treated "spot" was brighter. There is no report on the size of the spot, could be anywhere from 1/3 of the filiment (meeting the largest size I could call a spot) to the size of the pointed end on a pin or smaller depending on what they were using to observe the spot.
This was simply the laser they were using in the lab- I'm sure they were using the laser that was available. Now that they know this impacts brightness, they can toss some seed money to a few grad students to do some more intensive studies on what power levels work best, if wavelength matters, etc. From the sounds of it, this is affecting only the very very top surface of the surface, maybe a few atoms down- enough to change the wavelength of light emitted by the filament, but not actually damage the filament.
Commercial installations would undoubtedly use a more powerful laser on an assembly line, blasting away at the bulbs as they pass by at 50mph. We have cheap solid-state lasers powerful enough to cut 3/8" thick place steel to micron tolerances and blast artillery shells from the sky, so this isn't anything all that difficult. If it means improving incandescent bulbs in applications that still use them, that's great.