I follow a couple of diabetes-related groups on the internet, and if you're so minded, it's possible to have heated debates with people on the question of how often you should test. I mean, test yourself as often as you want, but bear in mind that for each test, you have to prick your finger, so you really have to ask yourself whether pricking yourself too frequently does you any harm. The reason the question is impossible to answer is that in many ways it is just like eating itself - eat as much of what you like right now, because in all likelihood it'll take years to have any effect. Plus there are lots of other factors that come into play during that time.
Anyway, I thought I'd describe my routine. Bear in mind that as I've blogged many times, I take insulin, so if I take too much of it it's possible for my sugar to go too low (potentially, if that happens, I go into a coma and die. But don't worry - the effect is recognizable and correctable before this happens.)
My testing regime is just to satisfy myself that I'm taking the right amount of insulin. To achieve this, I normally just check myself just once per day, before any food or meds, just to get an idea of what my baseline blood sugar is.
If the value is higher than I'd like, I make a note of it for future reference. I try and think back to yesterday - what did I eat which could have caused this? It's invariably something high in carbs. A bag of crisps, maybe? Not necessarily anything sweet. Sugar is a carb, but it is a mistake to think that sugar is the only carb.
Following this, I'm very strict for a day or so. If the high value were due to food, then the test result should go down. If the test doesn't go down, and it's been that way for a number of days (I don't have a fixed number of days), then I think about increasing my dose. In theory, if the number were consistently low, I'd similarly decrease my dose. But it doesn't tend to happen to me.
If I do decide to increase my dose, I'm back to testing once again. Basically, to get a picture of what my sugar is doing throughout the day. I take insulin twice a day (after breakfast and after supper) so I'd like to get an idea of which injection to increase. At this point, the here-and-now characteristic is a pain in the neck. It'd be far easier (although a darned site more expensive) to be able to look at something continuous. Anyway...... But once I have made a decision, I will increase the dose by a small amount at a time (2 units, if that means anything to you). The process then starts over.
Oh, it's probably worth mentioning as an aside that the glucometer that I use doubles as a USB stick, and so every week or so I walk it over to my computer, which obediently sucks the values in. So I have readings from...forever. I decided long ago that keeping track of these numbers was more important than cost (I can get another glucometer, and its test strips, on prescription, but don't bother). It attaches a date to every reading and therefore eliminates the mind playing tricks.
Individual tests I take at least 2 hours after food, just so that it doesn't skew the result. I've also heard that 90 minutes is an acceptable interval, and don't really think it's something worth arguing about. The finger should be clean before testing, just from the perspective of contamination.