There's no question that every company needs a solid pipeline of business -- that's what gives you predictability, profitability, and consistency. The \n\nquestion is, "how do you get more pipeline?"And if you want profitability, businesses should be focused on the cost of customer acquisition. The problem is, leads are pretty far way away \n\nfrom pipeline. To ask questions about the cost of leads is focusing on the wrong thing: it doesn't matter what your answers are, they're \n\nwrong.Why?While it's easy to state the cost of leads, the value of any specific lead is indeterminate. Ask the sales guys, and they'll say that the leads they \n\nhave are worthless. In many (maybe even most) organizations, 50 percent of the leads are not followed up on soon enough to make a \n\ndifference...if at all. But if you assign a lead to somebody else, the losing rep will scream bloody murder that he can no longer make quota because \n\nof losing that one lead.\n\nThis isn't a snarky joke. This is a classic problem that's dealt with in behavioral economics and game theory.When you have an item that has a low probability of having a very large value, it's not rational to run a weighted average. Think about lottery \n\ntickets: you have a 1-in-49 million chance of winning $75 million. The value of your ticket is $1.50, but only in the abstract. As a practical matter, \n\nyour ticket is worth $0. In business school, we teach MBAs to value startup stock options that way. And that's how I teach my students to value \n\nleads.Of course the probability of a lead converting is way better than a lottery ticket. But measuring raw leads just doesn't ... err ... lead to good \n\ndecisions. Job One: Get the Semantics Under ControlThe main problem with discussing the value and quality of leads is that the conversation is based on a blurred and confused definition of the \n\nword "leads." Sales reps will often confuse a lead with an opportunity, implying that leads are very valuable. Your CRM system doesn't work that \n\nway.So let's start by clearing that up (note, these semantics are for the enterprise sales model -- your firm's sales model will drive somewhat different \n\nsemantics):\u2022 Names are at the bottom of the food chain, because they aren't really leads at all. They are simply people whose names and \n\ncontact information we've obtained through list purchases, e-mail blast responses, Web site registration, or other means. While they may have \n\nindicated "I want to know more," they have given us nothing about their intention of doing business with us. The goal is to keep these "not yet \n\nleads" away from the sales force, and to use marketing automation to nurture these leads at near zero cost.\u2022 Once a name has started to take actions that indicate some serious interest, they are promoted to leads. While these leads are \n\nsometimes called Marketing Qualified Leads (MQL), in most B2B markets this is an oxymoron. In my opinion, it is not worth the effort to measure \n\nthese closely because the MQL statistics are inherently too noisy. In some markets, the probability of an MQL becoming a new customer is well \n\nbelow 2 percent.\u2022 Once a lead has been interviewed by the telesales organization (who may be called inside sales, market development, or related titles), it \n\ncan be qualified and converted to a contact. These Sales Qualified Leads are worth measuring for quality, quantity, and cost, even \n\nthough the final probability of becoming a new customer can often be below 10 percent.\u2022 After conversion, the sales organization begins to apply more effort to developing the contact, and at some point an opportunity \n\nwill be created. A new opportunity is the definitive signal that a sales cycle has begun, and it needs to be carefully measured as a B2B sales cycle is \n\na significant investment of the company's resources. Even though the win rate of open sales cycles ought to be better than 25 percent, detailed \n\nanalysis of your pipeline may show that a large portion of those opportunities that do close are not profitable. This is OK only if your organization \n\nknows this and is pursuing these deals with strategic objectives in mind. That said, one of the staples of management consulting is the discovery \n\nand weeding out of unprofitable customers.What Should You Measure, and More Important What Should You Optimize?The key to understanding the profitability of your sales engine is to measure these things: The cost of closing an initial customer\n The "first year value" of that customer\n The likelihood of getting repeat business from your customers\n The cost of closing repeat business\n The "repeat business value" of that customerAll five of these can -- and should -- be measured through your CRM system. But only the first one is really related to leads. And if you analyze \n\nit closely, you'll see that the direct cost of leads is only a small portion of the total required to close a new customer. The big issue with leads isn't \n\nthe direct cost, it's the quality (as measured by their likelihood of closing).If there were one thing to optimize in the whole sales cycle, it's the number and quality of sales cycles started (as measured by number and \n\nvalue of opportunities opened). You don't really care how many leads it takes to create that much pipeline, so don't focus on simple lead metrics. \n\nInstead, focus on optimizing which leads convert to pipeline.David Taber is the author of the new Prentice Hall book, "Salesforce.com Secrets of \n\nSuccess" and is the CEO of SalesLogistix, a certified Salesforce.com consultancy \n\nfocused on business process improvement through use of CRM systems. SalesLogistix clients are in North America, Europe, Israel, and India, and \n\nDavid has over 25 years experience in high tech, including 10 years at the VP level or above.Follow everything from CIO.com on Twitter @CIOonline.