1. get everyone to click
2. get only converters to click
I've been frustrated by #1 costing too much money with too few conversions, so I'm delving into #2 and I see a lot of potential to make it work. With all the tools at our disposal, a lot can be done to weed out non-converters before they click and cost money. I would think that kind of optimization would eventually yield a conversion rate much higher than, say, organic traffic, and I'm wondering how much higher it can get.
How much better does your AdWords traffic convert than your organic traffic?
But that's far and away the exception amongst my client base.
I spend a lot of time trying write really really really tight ads for these products, building up the product information on the landing pages, and making sure those ads go directly to the right pages.
There are some niches were I write ads to discourage clicks from consumers. I lose at least a few of those keywords because it is hard to maintain a high CTR when you are only targeting a segment of the people searching that term. In cases like this, negative keywords can help, but they don't solve all problems.
To my knowledge, there aren't any guidelines on CTR anymore. It's merely one factor in Quality Score and that is what Google uses for pricing, placement, and if your ad can even show.
If I were you, I would do some testing with ads to discourage casual browsers. You can probably get to a point where you are lowering the CTR, but still maintaining a high-enough Quality Score to be active. Make sure to include a sizeable list of negative keywords (free, etc.)
A good starting place is the Google Keyword Suggestion tool. Type your high-traffic broad-match terms into there and use negatives for anything that doesn't apply.