Since a company grows in value as it progresses, the founders can minimize their dilution by raising only as much money as necessary at each stage of growth. Ideally, you would raise money just as you need it, but that would require constant fund raising and preoccupy management with selling stock as opposed to building and selling product. Because investors tie the growth in the value of the business to the achievement of demonstrable milestones, increases in valuation can only be realized in a stepwise fashion.
So the answer to the question, "How much capital should we raise?", becomes apparent. You should raise as much capital as is necessary to get to the next major milestone that will justify a substantive increase in the company's stock price. When it comes to cash, the cost of under funding vastly exceeds the cost of over funding. It is therefore prudent to add a fudge factor to the estimate of how much capital is required to get to the next milestone.
What milestones justify successively higher prices? Typically they are the completion of a prototype, completion of the management team, conclusion of beta testing for a product, building a list of initial referenceable customers, getting customers to place repeat orders, reaching cash flow break-even and profitability, filling out a fuller product line, and completing a series of profitable, growing quarters on plan.
Prudent CEOs raise more capital than they think they'll need and rarely turn away capital in an oversubscribed round. There are two reasons why taking too little cash and running out is so costly. First, it puts the company in a very weak position when negotiating price with a new investor. More importantly, it reveals a lack of ability to forecast the future and therefore undermines new investors' confidence in management's plans. Most venture capitalists believe with good reason that there is an inverse correlation between bridge loans and a company's ultimate success. If it is available, Take The Money!