Do not chase a valuation. Run a process that finds it.

Travis Kalanick, the founder of Uber, the ride hailing company, learned to raise money by refusing to name his price. Nine months after a round that valued Uber at $3.5 billion, Bill Gurley, a partner at the venture firm Benchmark Capital who sat on Uber's board, urged him to take the first term sheet at a suggested $6 billion valuation. Kalanick waited, ran his process, and closed two months later at a $17.5 billion pre-money valuation.

His rule, as Kalanick puts it, is "never get attached to a price, get attached to a process." A founder who takes the first written offer to invest, called a term sheet, has no alternatives, he says. Then the investor can keep pushing for more terms, the negotiation drags, and the deal is more likely to blow up. With competing bids, the round closes faster.

The process has two parts: a QED story, then an auction. In normal times, he tells a detailed narrative with numbers woven through it that proves he has the winning formula. QED means proven, from the Latin for which was to be shown, and the pitch aims to leave no question open.

In a very bullish market, which he says describes the present, he shortens the same pitch from two hours to 45 minutes and shifts to what he calls the theory of the case. A detailed forecast three to five years out cannot prove a 100-fold payoff, and too much detail, he says, will talk you out of the deal.

At peak Uber, the auction meant five rooms running at once for an entire week. There was a room for investors writing roughly $250 million checks, then rooms for $100 million, $50 million, and $25 million, with Kalanick and his team telling the story in several rooms at a time as a kind of performance art. It takes staff and preparation to sustain, and he measures it against the resources available.

The bidding starts low. Instead of defending a high valuation, he tells investors he does not know where the price will end but knows it is at least this level. Once one firm leans forward, he has a bidder and can move to the next.

In a winner takes all round, he carries each bid to the next room at the prior bid plus five, telling earlier bidders the price is going up until bidding tops out. Then it is going once, going twice, sold. Most large rounds today, he says, use demand curves. Each investor writes down how much it would invest at each valuation. He adds up total demand at every price, picks the price where demand matches the amount he wants to raise, cuts investors who bid too low, and lets them submit again. He reruns the curve and repeats until it is clean.

To start, line up a crowded calendar in a single week so you can open low and let competing bids do the work.