Many college students are procrastinators. The days before papers come due and exams arrive are stressful indeed as they scramble to catch up. And while this is far from optimal, in terms of a learning experience, most procrastinators somehow manage to squeak by and complete their courses successfully.
Farmers can't work this way. The farm owner has to plan for the coming growing season. On a predetermined schedule, adjusted as necessary to reflect the prevailing weather conditions, the farmer has to turn over the soil, plant, fertilize, weed, and harvest. There is no way to stall, or to compress the steps needed to bring a crop to harvest.
Unfortunately, most salespeople act more like college students than like farmers. They approach their year-end quota in much the same way procrastinating students approach semesters: confident that they can cram as needed if they fall behind their year-to-date (YTD) target. Is there a more optimistic person in the world than a salesperson sitting at 37 percent of quota going into the last quarter? (We have yet to meet such a person.) Most lagging salespeople convince themselves that somehow, the elusive numbers will be hit by the end of the year. The surge of hope stays alive until the last week or two, when they finally admit to themselves that this year is a lost cause, and that it's time to "sandbag" orders so that they can hit the ground running the following year.
Salespeople simulate electricity: They follow the path of least resistance. Procrastination is an easy trap to fall into. And procrastination with respect to YTD quota position is reinforced by a salesperson's desire to have as many opportunities in the funnel as possible. Yes, if you add up all the items listed, the resulting figure may approach the gross domestic product of a small Central American country. But a closer look reveals that many of the "opportunities" listed in most pipelines have little chance of closing. This underscores the need for managers to grade funnels with an eye toward (1) setting appropriate activity levels and (2) disqualifying low-probability items.
As noted in a previous chapter, we distinguish between pipeline and funnel management. Pipeline milestones are graded and used by sales management for the purpose of forecasting, while sales funnel milestones are graded and used by sales management to assess the quantity of selling activity and the quality of selling skills at the individual salesperson level.
Left to their own devices, many salespeople compete to stay busy. In our view, the role of sales managers is to help them compete to win. But the prospect of putting a salesperson on an improvement plan - or, worse, firing an underperformer and recruiting and training a replacement - is enough to make a sales manager procrastinate like a college student.
The goal of managing pipeline is predicting what will close - forecasting - but these predictions are driven as much by personal agendas as by reality. A salesperson who is below quota has a specific agenda when forecasting: get my manager to believe that there is sufficient activity so that I can keep my job.
One major challenge in forecasting is that companies lack consistent qualification elements. Managers with experience at different companies bring along their own approach to grading opportunities. Even when companies attempt to impose standard milestones, an "80 percent probability" actually varies from salesperson to salesperson, and from district to district. The problem is that these milestones rely on a subjective judgment: Has the opportunity progressed to that level, or not? Well, tell a traditional salesperson who is behind quota that by the end of next week, she must have five opportunities at a certain milestone. By the end of next week, she will have six (thus allowing a margin for error, if one is successfully challenged by the manager). You can set your watch by it, but would you want to forecast these opportunities?
Predicting the date that an opportunity will close is a challenge for even the best salespeople. It's simply not their call. Forecasting can be as meaningless as pushing a few dates back by 30 days and tweaking a few numbers (especially when there is little or nothing that has been added in a given month). And these summaries tend to blow away in the first wind, in any case. If a salesperson forecasts a transaction to close in September, then forecasts it in October, and it finally happens in November, the accuracy is 33 percent - but what's remembered is that the salesperson got the business.
Salespeople despise forecasting because, in most cases, they are being asked to lie in writing. They know all too well that the exercise of forecasting is likely to bear little or no resemblance to reality. In fact, the major value of forecasting is that it has the potential to give salespeople with inadequate pipelines a wake-up call 12 times a year - that is, a message from on high that there are inadequate opportunities in their pipeline, and that they must increase their business-development activities.
So let's look at ways to get to better forecasting, in part by removing salespeople from the process.
In order to break away from the insanity referred to euphemistically as forecasting, a number of complementary components are necessary:
Sales-Primed Communications, designed to position offerings specific to title, vertical industry, and goal, which provides less subjective input into the pipeline
Auditable correspondence between the seller and the buyer
Sequences of Events containing estimated close dates, as negotiated with the buyer
Companywide milestones, with defined ways to achieve and document them
Sales managers (not salespeople) who are willing and able to audit milestones, grade pipeline, disqualify low-probability opportunities, and predict what will close
Senior executive commitment to ride through potential "push back" from salespeople and sales managers who prefer less visibility and accountability
Items 1 through 3 have been described in previous chapters. So let's start with Customer Focused Sales milestones, which should be modified as needed and adopted as companywide standards. First, refer to Figure 15-1.
Inactive means that an account fits a company's target market and is assigned to a salesperson, but there is no current activity. A salesperson's business-development efforts should focus on getting a buying cycle to start by getting a targeted Key Player within the account to share a goal, which would advance it to the next milestone.
Active indicates that contact has been made and some form of interest on the part of the customer or prospect has been expressed.
Goal Shared is the initiation of a sales cycle in Customer Focused Sales; it indicates that a targeted Key Player has shared a desire to achieve at least one goal that is on the menu of business issues the seller can help them address.
Champion status can be granted only by the sales manager. It happens only after all the qualification elements have been achieved: the letter or email has been sent; the buyer has agreed to the content; and the buyer is willing to provide access to Key Players. The sales manager must have reviewed the customer document and graded it as a C, usually after a brief discussion with the salesperson.
The opportunity remains a C while all Key Players are being interviewed, either in person or on the phone. All calls should be documented with each buyer, and there should be agreement that the content reflects the conversation that the seller had with them. After all Key Players have been qualified, there is an attempt to qualify the opportunity. Some of our clients like to not only track the number of months an opportunity is at this level, but also get a sense for how many Key Players must be interviewed and how many have been called on. A greater number of Key Players most likely means that an opportunity will remain a C for a longer period of time.
Evaluating status is determined by the sales manager, but only after the salesperson has gained consensus from Key Players that further investigation of the salesperson's offering is called for, and a Sequence of Events has been accepted by the buying committee. At the end of that month, the salesperson provides a copy of the cover letter and Sequence of Events, and the sales manager can change the status to E (and update the Sales Force Automation/Customer Relationship Management (SFA/CRM) system) if it is deemed a qualified opportunity.
Opportunities remain at this stage until one of three things happens:
The buyer withdraws (and the opportunity becomes a loss).
The seller withdraws (and the opportunity becomes a walk).
The seller asks for the business after the Sequence of Events has been completed.
Once the seller has asked for the business, the opportunity goes into one of four grades:
W (Win) | An order with all necessary documents signed. |
L (Loss) | The buyer informs you that he or she will not be moving forward with you. |
V (Verbal) | The buyer has given you a verbal commitment to go forward. |
P (Proposal) | A proposal is provided to the prospect with a decision due.> |
A few notes about the chart. Many of our clients choose to assign probabilities to the G, C, and E grades. Until you accumulate your own historical data with your forecasting system, we suggest assigning initial probabilities of 25 percent for the Champion (C) level and 50 percent once the manager assigns the Evaluating (E) level to an opportunity. (Organizations with SFA or CRM systems may ultimately collect percentages for individual salespeople, enabling them to fine-tune forecasting accuracy.) For an E prospect, each checkpoint agreed to increase the chances of getting an order.
Capturing the number of months (m) that a line item in the pipeline has been at a particular stage allows the sales manager to identify opportunities that are not progressing. For E status, adding a percentage to reflect the percentage of steps in the Sequence of Events that have been completed can also provide a meaningful point of reference for the manager. For the grades of W and L, it may similarly be useful to track the total number of months (M) that an opportunity was in the pipeline before you won or lost. After losses, data indicating whether the winner was a named competitor or there was no decision (meaning no gold medal was awarded) should be entered.
By implementing the system shown in Figure 15-2, managers can detect early-warning signs that a given opportunity is stalled or in trouble. For each pipeline milestone shown, you'll see two examples: one that is on track, with normal parameters for a viable opportunity; and another that is beyond normal conditions. While there may be extenuating circumstances, the latter situation should prompt the manager to have a conversation with the salesperson. If necessary, both can agree on a course of corrective action to get the opportunity back on track. Alternatively, it may be appropriate for the manager to disqualify this line item from the pipeline, rather than allow the seller to toil with a low probability of getting the business. (Remember: Bad news early is good news.)
Even without an SFA/CRM system, sales managers can now forecast by asking each salesperson to email the latest copy of the Sequence of Events for each opportunity. By reviewing these documents, the sales manager can assess each opportunity with a reasonable degree of objectivity and consistency (in an apples-to-apples sense).
In order to home in on forecasting accuracy, we suggest that sales managers maintain three pipeline categories for each salesperson:
Add-on business with existing customers, having the highest probability of closing
New account business in which you were proactive and started as Column A
New account business where you were reactive and started as Column B, C, and so on
Forecasting is now a monthly review of pipeline, with sales managers choosing which opportunities are likely to close and the Sequences of Events providing lots of visibility and realistic close dates.
This monthly review also affords an opportunity to be proactive with salespeople who might be procrastinators. "Train wrecks" within a funnel do not happen overnight. Let's see how a manager can be proactive in anticipating a shortfall in a given funnel. A few variables are necessary to provide sales managers with visibility one sales cycle ahead, as shown in this example.
A salesperson's annual quota: $1.5 million
The average length of a sales cycle: 4 months
The probability that an E will close: 50 percent
One way to reduce procrastination is to break down a quota to a monthly number - in this case, $125,000/month. Multiplying this by an average sales cycle tells you that in 4 months, a salesperson will be expected to close $500,000. If, however, the salesperson closes 50 percent of the Sequences of Events that were put in place, the funnel target would be twice that figure, or $1 million. This means that at any given time, when you take a snapshot of a salesperson's funnel, you would like to see at least $1 million at E status, as graded by the sales manager. If you were halfway through the year and the salesperson was $100,000 behind the YTD number, the target would be the standard of $1 million plus ($100,000/50 percent), or $1.2 million.
This may seem like a mouthful, so let's look at the activity balancing example below, where the per month quota is $125,000, the average sell cycle is 4 months, and there is a 50 percent probability of closing a Grade E prospect.
By doing this calculation on a monthly basis, sales managers can help salespeople stay ahead of the curve. While managers should review changes to, and volume of, A, G, and C entries, E opportunities provide the best sense of how things will look one sales cycle ahead. If the E funnel is below target, business development activity must be increased. The conversation could sound like this:
Jane, you have been running on your numbers through April, but this month you only have $800,000 in E business. Over the next month, I want you to make sure that you increase your business-development activities. My suggestion would be to make a minimum of 10 new contacts per week. I'd also like to see you migrate two opportunities from C to E, and I will work with you on doing that. Hopefully, next month you'll have sufficient new opportunities in your funnel so that we'll both feel better about where you'll be YTD in September. We can then revisit how much of your time should be spent on business development.
This approach, in other words, emphasizes proactive rather than forensic sales management. If a salesperson competes for 8 months and loses, it is no longer solely his or her responsibility. Each month (or more frequently), the sales manager should have made a qualify/disqualify decision at all checkpoints in the Sequences of Events that were approved. The Sequences of Events also allow a sales manager to get further upstream on opportunities, to coach new or struggling salespeople all along the way, and to have control and ownership of the forecast he or she provides.
When these kinds of analyses are performed on a monthly basis, course corrections can be made to maximize the probability that salespeople will achieve their numbers - and thereby minimize turnover, either voluntary or involuntary. Involuntary turnover, as suggested above, is difficult for all parties involved. From the company standpoint, there is a significant expense inherent in recruiting and training a replacement, lost continuity within the territory, ramp-up time for new hires, and management time spent on the territory during the interim period. By the time many managers (and salespeople) realize they are in trouble, it is often too late to save that year. An analysis on a monthly basis, therefore, is well worth the time it takes to perform.
In a previous chapter, we distinguished between "will not" and "cannot" attitudes in a salesperson. "Will not" is, in most cases, a management issue. "Cannot" is a skill issue, and will serve as the basis of the next chapter.