For the bid price strategy, I tend to start a low as possible and work up towards what it will cost to show on page one. This has two main advantages, it's cheap and it gives me the chance to weed out high CTR poorly converting KPs as we go along. The disadvantage of this technique is that it's not as quick as going in at top bid price and working down and the lack of clicks compared to the expensive way of doing it will give you a slightly lower quality score which can negatively affect your CTR. Bit of a catch 22 but since my goal is minimum client initial expenditure I have to go with the slower method (We can always bump the budget up once they've seen some results and are happier about spending more on it).
So tactic 1, create X number of Ad groups, each named after the KP that they're focussed on and containing all the actual KPs,set the bid price at 5p across the board and then monitor. Increase bid prices as the average positions start to show to bump the Ads onto page one. Weed out poorly performing KPs before they become too expensive.
Tactic 2, throw caution to the wind, set all your bid prices to whatever Google recommends they should be for page one (always inflated in my experience), set a budget that will cope with the cost and then monitor it closely, knocking bid prices down to get them into positions 3 - 7 (less clicks, more conversions as people who work down the list are more serious about spending).