How to build a high-throughput Sales Process

Applying Sales Process Engineering (including TOC) to the design, resourcing and management of the sales process The modern sales function resembles production, as it would have been, prior to the industrial revolution. This paper presents a method for engineering the sales function into a measurable, manageable and scalable process — the key characteristic of which is an extraordinarily high volume of business-development activity (salespeople consistently perform at least four appointments a day, five days a week). As well as presenting the theoretical basis for the new approach, this paper provides detailed insights into the practical implications of Sales Process Engineering. AbstrAct How to build a high-throughput Sales Process SaleS ProceSS engineering Justin Roff-Marsh © Copyright: Ballistix 2009 2 of 21 Introduction The traditional sales function is hard to manage and all but impossible to scale. This paper introduces a radical new approach to sales process design, resourcing and management. A method called Sales Process Engineering. The result of this Sales Process Engineering is a process where: 1. Salespeople consistently perform four appointments a day, five days a week. 2. A queue of sales opportunities is generated and maintained, without salespeople’s involvement. 3. Budgets, targets, bonuses and commissions are eliminated and all activities are synchronized (in real-time) with the goal of the organization. 4. Throughput is increased, without a significant increase in operating expenses I’m sure you’ll agree that the phrase ‘high-throughput sales process’ has a pleasant feel to it. We’ll benefit, however, by clarifying the meaning of two critical words that appear in the title of this paper: throughput and process. Throughput In the TOC world, throughput means something very specific. Throughput is the rate at which a business generates money (where money refers to the difference between the sale price of a product and its raw- material cost)1. It’s convenient to think of a business as a Throughput pump. The business consumes raw- materials (investment), adds value using some kind of machinery (maintained by operating expenses) and outputs Throughput. When a business pumps Throughput faster than it consumes operating expenses, then it’s profitable. So, with this in mind high-throughput is obviously good! But, in what sense can we claim that a sales process pumps Throughput? Well, a sales process doesn’t pump actual Throughput, but it does pump probable Throughput. And in the context of the sales process, a unit of probable Throughput is what’s called a sales opportunity. This paper introduces Sales Process Engineering, the goal of which is to maximize the flow of sales opportunities. It should be clear now that this is congruent with the goal of the organization — which is to maximize the flow of Throughput. You may be surprised to discover, however, that most sales processes are not designed to maximize opportunity flow. The opposite, in fact, is typically the case! 1 Technically, Throughput is the difference between (irrevocable) sales revenue and totally-variable costs. Totally- variable costs are only those costs that vary in direct proportion with revenue in the time horizon under consideration. You may be surprised to discover, however, that most sales processes are not designed to maximize opportunity flow. The opposite, in fact, is typically the case! Justin Roff-Marsh © Copyright: Ballistix 2009 3 of 21 Process The term sales process is commonly used in organizations. However it is quite uncommon for an organization to contain anything that even vaguely resembles a process in or around the sales function. At least in a production context, the word process conjures up images of a production line: a series of tightly-coordinated activities — each performed by a specialist — that, collectively, converts raw materials into finished goods. But a typical sales function hardly fits this description. A typical sales function consists of a number of individuals (salespeople), each of whom is responsible for all sales activities (and quite a number of non-sales activities). These individuals operate in parallel — as autonomous entities. Even if some individuals follow standard procedures, there is huge variation between individuals. It’s worth highlighting that the sales function, today, looks remarkably like production would have done, prior to the dawn of the industrial revolution! A modern salesperson can be accurately characterized as an artisan. This paper describes how to convert the sales function from its pre- industrial-revolution form into a process, in the true, industrial-age, sense of the word. And yes, this paper describes how to create a high-throughput process — where maximizing opportunity flow is the goal. What’s wrong with the sales function? If you were to survey managers from a number of disparate organizations and ask each to compile a list of the issues that bother them about their sales functions, you’d almost certainly discover that the resulting lists contained similar items, in a similar order: 1. Salespeople spend only a small percentage of their time selling (see diagram below) 2. Deal-flow (sales) is lumpy and unpredictable 3. There always seems to be conflict between sales and other functions 4. Salespeople are all but impossible to manage It’s worth highlighting that the sales function, today, looks remarkably like production would have done, prior to the dawn of the industrial revolution! Justin Roff-Marsh © Copyright: Ballistix 2009 4 of 21 If we were to track these issues back to their ultimate causes we would discover something interesting. All are the consequence of management equivocating between: 1. Treating salespeople as autonomous agents (sub-contractors) 2. Treating salespeople as team members (employees) Now, each of these approaches, in isolation, is acceptable: It is quite acceptable for management to outsource the sales function — as it does when it sells through a reseller (or dealer) network. It is also acceptable (as we’ll discover in due course) for management to treat salespeople as employees. The problems occur when management attempts to equivocate (or compromise) between these two approaches. This is because, while it is possible for a person to be an autonomous agent or an employee, it is not possible for that person to be both. The two are exact opposites (an autonomous agent marches to his own drum; a team member, to the team’s). The diagram below presents this conflict, along with the reasons why management pursues these mutually-exclusive approaches — as well as the assumptions that underpin the reasoning. As the diagram suggests, management’s objective is to maximize sales. On the one hand, in order to achieve this, management must maximize conversion — which requires that salespeople are treated as autonomous agents. But, on the other, management must exercise control — which requires that salespeople are part of the team. Both appear to be reasonable enough lines of reasoning. If we take a moment to review the assumptions presented in this diagram, the assumptions also seem reasonable enough also. There is one assumption, however — one that is firmly embedded in the fabric of every sales team — that simply does not survive contact with reality! Justin Roff-Marsh © Copyright: Ballistix 2009 5 of 21 A new assumption: a new sales reality The erroneous assumption is this one: In order to maximize sales, we must maximize conversion. The fact is, in most organizations, conversion rate is already too high! Management typically assumes that the ideal conversion rate is 100%, but nothing could be further from the truth. In practice there’s an inverse relationship between conversion rate and the amount of time salespeople devote to business-development tasks. Not only is this an inverse relationship, but it’s a non- linear one: incremental increases in conversion invariably come at the expense of an exponential decrease in business-development capacity. The reason for this is simple. Salespeople achieve increases in conversion by assuming responsibility for more and more tasks. What’s more, because most salespeople receive equal credit for repeat purchases and for new-client acquisition — and because most sales are sales to existing clients — salespeople spend increasing amounts of time servicing existing clients. So, in summary, in a typical organization, when conversion goes up, sales volumes actually go down! Once we accept that the assumption that conversion drives sales is erroneous, an interesting thing happens: we recognize that it’s no longer necessary to strive to maximize conversion — and, consequently, that it’s no longer necessary for management to treat salespeople as autonomous agents. This second realization will come in handy shortly. If conversion does not — in practice — drive sales, it’s obvious that opportunity flow does. Accordingly, if we want to increase sales we must switch our focus from maximizing conversion to maximizing opportunity flow. This change of focus will have radical and far-reaching implications. Maximizing opportunity flow So how, in practice, do we go about increasing opportunity flow in a typical sales function? We stumbled across a clue when we noted that a typical sales function, today, looks remarkably like production would have done, prior to the dawn of the industrial revolution. Now we know that the productivity of production (measured against any reasonable standard) has increased by several orders of magnitude over the last couple of hundred years. And we don’t have to look far to identify the primary reason for this increase in productivity: division-of-labor2. It’s certainly fair to say that the primary difference between a modern production (or project) environment and sales is division-of-labor. After all, it’s the absence of division-of-labor that resulted in our previously characterizing the modern salesperson as an artisan. Therefore, if we wish to significantly increase opportunity flow, division of labor would appear to be a sensible method. 2 It’s often assumed that automation is the primary reason for this increase in productivity. Automation is actually a sub-set of division-of-labour (a discrete task is handed to a machine, rather than a person). So, in summary, in a typical organization, when conversion goes up, sales volumes actually go down! Justin Roff-Marsh © Copyright: Ballistix 2009 6 of 21 However, before we pursue this line of reasoning, we must stop to consider why, in the 21st century, the sales function looks the way that it does. Is there not a good reason why division-of-labor has not been applied to this sales process to date? The relationship between opportunity flow and conversion rate There are two reasons why division-of labor has not been applied to the sales process: 1. Quite simply, it has not been necessary. For most of the last two centuries the ability of businesses to make money has been constrained by production, not sales. For this reason, improvement initiatives have been focused, quite correctly, where they have produced the best return (production and, more recently, projects). However, in the last fifty years (or so) we’ve seen the constraint shift from production to sales. Today, many organizations have more production capacity than they can sell. It’s now time to apply what we have learned about process improvement in the production environment to the sales function. 2. Those who currently work in and around the sales function are naturally resistant to the application of process-engineering principles — just as an artisan might have been resistant to division-of-labor and statistical-process-control two hundred years ago! It’s worth briefly considering the primary objection we are likely to receive from those who currently work in sales when we broach these ideas with them. If we suggest that we switch our focus from maximizing conversion to maximizing opportunity flow, we will inevitably be told that any gains in flow will come at the expense of conversion: “What’s the point of performing more appointments if we fail to make any sales?” The assumption here is that opportunity flow and conversion are mutually exclusive — that they’re opposite ends of the one continuum. We’ve already acknowledged that there is a relationship between flow and conversion, but is it true that they are mutually exclusive? In other words, will our gains in flow necessarily come at the expense of conversion? We can gain insight into this question if we reflect again on production. If we assume that conversion (in sales) is analogous to waste (in production) we immediately make an interesting discovery. Over the last two hundred years, production flow (volume) has increased by several orders of magnitude — and, we’ve seen a comparable reduction in waste. The reason for this is simple. The primary driver of waste is production variability. And a pre-requisite for increasing production flow is the reduction of variability. Therefore, when management focuses on maximizing production flow, a reduction is waste is an inevitable consequence. Now, this is not to say that the optimal level of waste is 0%. The only way to achieve this would be to slow production to zero! The point is that, while there is some kind of relationship between flow and waste, the two are not mutually exclusive. In the context of sales, the same applies (and for the same reasons) to the relationship between opportunity flow and conversion rate. In practice, the method presented here is likely to result in an initial reduction in conversion as opportunity flow increases (remember, in most organizations, the current conversion rate is simply too high). However, as flow continues to increases, conversion naturally finds its optimal rate, and stabilizes there. Justin Roff-Marsh © Copyright: Ballistix 2009 7 of 21 Division-of-labor With this second common (but erroneous) assumption dispensed with, let’s return to the method we identified previously for the multiplication of opportunity flow: division-of-labor. We know that the sales function is a perfect candidate for division-of-labor. A typical salesperson performs a large number of disparate tasks. We could categories these tasks in two ways, as per the table below. Function Task Type Opportunity Generation Opportunity Management Fulfillment Clerical Direct Mail & Appointment Setting Proposal Generation & Appointment Setting Customer Service Selling (persuasion) Prospecting Business-development Appointments NA Technical NA Solution Design Project Leadership The obvious question we should ask, at this point, is: to what task is the salesperson ideally suited? If we had to choose just one task, it would be the performance of business-development appointments. It should be possible to allocate fulfillment tasks to the production function (although it may be necessary to review production processes to ensure that salespeople’s withdrawal from these processes does not cause problems). It is also quite possible, in our experience, to re-allocate all opportunity-generation tasks, as well as those clerical and technical tasks that fall within the opportunity-management function. This will be discussed in due course. Assuming this is possible, it’s obvious that the salesperson’s capacity will increase significantly when he or she performs nothing other than business-development appointments. In practice, we consistently see salespeople increasing their capacity from two business-development appointments a week to 20 (an order of magnitude improvement). Two missing ingredients If the reader were to have the ability to ask a question at this point, it’s easy to imagine what that question might be. Almost certainly, the reader would ask: “What you are suggesting may indeed be possible, but is it even remotely practical?” The answer, at least at this point in our journey, is no: what we are suggesting is not remotely practical! In the absence of two critical pre-requisites, the application of division-of-labor to sales will result in chaos, conflict, a rapid decline in customer satisfaction and plummeting conversion rates. Justin Roff-Marsh © Copyright: Ballistix 2009 8 of 21 These two critical pre-requisites are well understood (at least, implicitly) by production people. However, because the sales function has no experience with division-of-labor we must explicitly recognize their importance. These pre-requisites are pictured below.     Standardization As is the case in production, division-of-labor cannot be applied unless the procedure employed to complete each task is the same for every iteration of the process. Obviously, in a typical sales function, there is a huge amount of procedural variation from opportunity to opportunity and (particularly) from salesperson to salesperson. This must be eliminated. It should be noted that standardization, in and of itself, should not be expected to increase opportunity flow.3 Rather, standardization is necessary because it’s a pre-requisite for division-of-labor (the ultimate cause of increased flow). We should consider standardization with respect to the sales process as a whole and with respect to the opportunity-management process (a sub-set of the sales process). (Because we are in the process of applying division-of-labor to the sales function, it’s now reasonable to begin to refer to it as the sales process.) The sales process The diagram below is a representation of the sales process as a whole. In its simplest form, the sales process consists of two activities that convert promotional expenditure (raw material cost) into Throughput. We’ve already proposed that salespeople should operate only within the opportunity-management process. Opportunities should be generated using promotional campaigns. This will be discussed in due course. 3 This is another common (but erroneous) assumption that explains why many CRM implementations fail to produce their expected benefits. Justin Roff-Marsh © Copyright: Ballistix 2009 9 of 21 The opportunity-management process The purpose of the opportunity-management process is to convert a sales opportunity into a sale (actual Throughput). As the diagram below suggests, an opportunity is essentially a project. (This diagram represents a typical opportunity-management process in a minor-sales environment)                In a minor-sales environment, it is likely that all activities — with the exception of the generation of the proposal — are performed by the salesperson (all are business-development appointments). The proposal is likely to be generated by the Customer Service team. In a major-sales environment, it is likely that: 1. The opportunity-management process contains many more activities (perhaps even a small consulting project) 2. Some appointments are attended by a technical expert (who is likely to be responsible for solution design, proposal generation and perhaps even project leadership if the opportunity is subsequently won) Irrespective of the structure of the opportunity-management process, it is essential that all opportunities traverse the one process. (It is permissible for some opportunities to skip activities.) Centralized scheduling The second pre-requisite for division-of-labor is centralized scheduling. Obviously it is difficult to imagine a production environment without a production scheduler (or a project environment without a project manager). The job of the scheduling function is to synchronize the rate of work of the various resources within the process. In a conventional sales process, there is no need for centralized scheduling and it is the responsibility of the salesperson to sequence tasks as he or she sees fit. This situation cannot, under any circumstances, be allowed to persist as we move towards division-of- labor. In a sales environment, the scheduling function consists of two components: 1. A scheduler (this person is typically called the sales coordinator) 2. Scheduling software (the scheduling software determines the sequence in which tasks should be performed by the various resources within the opportunity-management process) Justin Roff-Marsh © Copyright: Ballistix 2009 10 of 21 Putting it all together We’ve now reached the point where we’ve canvassed (at lease, at a conceptual level) the essential components of this new approach to the sales process. It’s time, now, to put all the pieces together. We know that our objective is to maximize sales. And we’ve recognized that the key to achieving this is to maximize opportunity flow. We understand that a radical increase in opportunity flow will necessitate division-of-labor, and that the pre-requisites for division-of-labor are standardization and centralized scheduling. If we assume that our business (like many today) has plenty of production capacity (or the ability to scale production rapidly), it follows that the output of the business will be determined by sales process capacity. We know that the capacity of the sales process will ultimately be determined by the process constraint: the resource with the least capacity. Because we are reengineering the sales process, we have the ability to choose the resource that we would like to be the constraint. If we consider that the constraint is the only resource that can be consistently fully activated, then in most cases we would like the salesperson (or people) to be the constraint. (Because the salesperson is typically the most expensive sales process resource, it makes sense to keep this person fully loaded with work). The activity in which the salesperson is engaged is appointments. Therefore, the salesperson’s unit of capacity is an appointment slot. (In practice, it is acceptable for some appointments to consume more than one slot.) Typically, average appointment duration is 90 minutes. In most cases it is easy for a salesperson to perform four (sensibly planned) appointments a day. If we want to maximize flow — and if the salesperson is the process constraint — it follows that we should: 1. Ensure that every available appointment slot is filled 2. Sequence appointments so as to maximize the yield on each slot (Throughput/Constraint Unit) The diagram below shows the beginnings of our new sales process.         We see that the sales coordinator (scheduler) is maintaining a queue of work upstream from the salesperson. This queue (the appointment buffer) consists of forward-booked appointments. Justin Roff-Marsh © Copyright: Ballistix 2009 11 of 21 In most cases this queue will contain about 10 days worth of forward-booked appointments (this will ensure that the salesperson is always solidly-booked about five days in advance). It should be obvious, at this point, that the sales coordinator has ownership of the salesperson’s diary. As is the case in a production (or project) environment, responsibility for scheduling cannot be shared. As well as ensuring that the salesperson is fully loaded with appointments at all times, the sales coordinator is also responsible for ensuring that the technical team subordinates effectively to the opportunity flow. I am using the term technical team to refer to any resources (other than the salesperson) that are responsible for activities within the opportunity-management process. (These resources might include customer service personnel, technical specialists, project leaders, etc.) These resources must fulfill their responsibilities as they fall due to avoid opportunity flow being compromised. (It is management’s responsibility to ensure that the technical team has sufficient protective capacity at all times.) Now that we can be confident that every available appointment slot will be filled, we must sequence appointments so as to maximize the yield on each. The following diagram completes our picture of the new sales process.   Upstream from the sales coordinator, we see a queue of sales opportunities. This queue is maintained by the promotional coordinator: the person responsible for opportunity generation. The opportunity queue is indexed (sorted) dynamically by the scheduling software. The result is that, when the sales coordinator draws opportunities from the top of this queue and programs them into the salesperson’s diary, the yield on the salesperson’s finite capacity is maximized. It’s helpful to visualize each opportunity as a series of (probable) appointments. In indexing the opportunity queue, the software must estimate: What Throughput contribution each opportunity is likely to make if it is programmed into the salesperson’s next-available appointment slot. The sales coordinator has ownership of the salesperson’s diary … responsibility for scheduling cannot be shared. Justin Roff-Marsh © Copyright: Ballistix 2009 12 of 21 Now that the salesperson is fully activated, performing appointments sequenced in order of expected contribution, we can now honestly claim that we have maximized opportunity flow (or, technically, the flow of probable Throughput). Practical considerations Let’s return, now, to fill some of the obvious gaps in our new model of the sales process. Is there really a shortage of sales opportunities? Opportunity-generation is a problem for many organizations. But, in most cases, it is nowhere near the magnitude of problem that management believes it to be. In order to understand that statement, we must explore the significance of a concept we’ve neglected to discuss thus far: qualification. In most sales environments, significant attention is paid to this concept, based upon the innocent (but erroneous) assumption that qualification adds some kind of value. In theory, the purpose of qualification is to ensure that salespeople do not devote units of finite capacity to individuals (or companies) that have negligible sales potential. In practice, however, qualification consists of salespeople discarding all opportunities other than those where the potential customer admits that they have both the budget and the intention to purchase within the near term. When we reflect upon disparate range of tasks that a typical salesperson has to perform — and, consequently, the minimal amount of capacity available for performing business-development appointments — it’s easy to understand why salespeople engage in this behavior. But, while qualification may be necessary, it’s certainly not beneficial! In discarding opportunities without a budget and the intention to purchase in the near term, salespeople are discarding the more valuable opportunities! The fact is, if a potential customer has already allocated a budget and a procurement date, then they have already designed and specified a solution (perhaps with the assistant of a consulting firm). Now, all they are looking for is bids. This has two obvious consequences: 1. The salesperson will end up competing for the opportunity on the basis of price (and fulfillment lead-time) 2. Because the salesperson always works in this environment, he or she will lose the ability to sell conceptually We must now accept that qualification (at least, as it’s currently practiced) is a value-destroying activity! A consequence of Sales Process Engineering is that salespeople’s (effective) capacity has increased by an order of magnitude. We no longer have a need for qualification. (In actuality, we’ve replaced qualification — which is categorical — with the indexing of the opportunity queue — which is incremental.) With the requirement to qualify opportunities eliminated, management will often discover something extraordinary: there is now an abundance of sales opportunities! Management is also likely to discover that, while the conversion rate will drop in the absence of qualification, the value of an average sale will rise. We must now accept that qualification (at least, as it’s currently practiced) is a value-destroying activity Justin Roff-Marsh © Copyright: Ballistix 2009 13 of 21 Opportunity generation If not now, at some time in the future it will be necessary to generate additional opportunities (to supplement the organic flow from existing customers and referrals). A pre-requisite for opportunity generation is a compelling (irresistible) offer. Such an offer can only be created if the product (or service) in question is meaningfully differentiated from those of competitors’. In the absence of meaningful product differentiation, the organization must recognize that its constraint has shifted to either production or new-product- development and devote resources to either of these areas immediately. The sales process (including the promotional function) should not be expected to manufacture demand for a product that is essentially unappealing to the market. Assuming that an appealing offer is in existence, simple direct-marketing campaigns can be used to generate sales opportunities. These campaigns should promote either: 1. In a minor-sales environment, an information package (containing product information, samples, case studies, etc) 2. In a major-sales environment, a whitepaper (or similar) — where the whitepaper is designed to sell a methodology that transcends (but stimulates interest in) the organization’s products and services It must be remembered that the volume of opportunities generated using this method is directly proportional to the degree of differentiation of the organization’s product. Organizations devote massive amounts of resources to marketing, when these resources should really be invested in production or new- product-development. On the subject of marketing; because this word refers to such a broad range of concepts, it is wise to discontinue its usage in any critical discussion. Certainly, in this discussion our interest is only in the promotional component of marketing. It makes sense to define promotion as: any activity that is designed to generate sales opportunities. This line of reasoning makes it apparent that promotion is a component of the sales process. If an organization has a marketing department — and if this department’s personnel are engaged primarily in promotional activities — then these people must come to accept that they are now part of the sales process (as opposed to members of an unrelated department). Indexing opportunities Earlier, we recognized the requirement to index the opportunity queue — and the general method that must be employed (the scheduling software must estimate): What Throughput contribution each opportunity is likely to make if it is programmed into the salesperson’s next-available appointment slot. The sales process (including the promotional function) should not be expected to manufacture demand for a product that is essentially unappealing to the market Justin Roff-Marsh © Copyright: Ballistix 2009 14 of 21 In most circumstances (including major-sales environments) the following formula (or a variation of it) will do a surprisingly effective job of indexing the opportunity queue: )'1( dueDaysOngApptsPendi %T lueRelativeVa +× × = It’s worthwhile developing an understanding of the components of this formula. Relative Value: the result of this formula is used only to index opportunities. T (Throughput): this is the expected value of the opportunity if indeed it is won. Every opportunity must be allocated a value (even if it is a best-guess). If (as will typically be the case) salespeople are not responsible for securing repeat purchases from existing customers, then the value of the opportunity should be grossed-up to include the value of future transactions. % (Probability): this is an estimate of the odds of the opportunity ultimately being won. Probability must always be determined by the current milestone (see below) within the opportunity-management process, not by the salesperson’s opinion. Appointments Pending: this is an estimate of the number of additional appointment slots that are likely to be consumed if the opportunity is ultimately won. Again, this is determined by the current milestone. Obviously this value will always be greater than, or equal to, one. Overdue Days: this is a measure of whether or not the opportunity has been current for longer than should reasonably be expected. Obviously, opportunities can atrophy over time. To determine Overdue Days, we subtract Maximum Days from Actual Days (and return only positive results). Maximum Days is also determined by the current milestone. As mentioned, the values for Probability, Appointments Pending and Maximum Days are determined by the current milestone within the opportunity-management process. These values are determined by interrogating historical data (or, in the first instance, with estimates). These values must not be volunteered by salespeople (or sales coordinators). The table below displays typical values for the opportunity-management process pictured earlier. Milestone Probability (%) Appts Pending Max Days 1. Best-practice Briefing Pending 8 3 28 2. Executive-briefing Pending 19 2 42 3. Proposal Pending 42 1 56 4. Proposal-customization Meeting Pending 67 1 70 5. Instruction-to-proceed Pending 86 1 84 Justin Roff-Marsh © Copyright: Ballistix 2009 15 of 21 Opportunity queue size It is easy to estimate the optimal size of the opportunity queue. The formula follows (again, with an explanation of terms): Safety OppAppts CycleTimeCapacity zeOptQueueSi × × = / Optimal Queue Size: the number of opportunities we expect the promotional function to maintain within the opportunity queue. Capacity: this is the daily capacity of the process constraint: the salesperson (typically 4 appointments a day). Cycle Time: this is the average cycle-time for a sales opportunity (in work days). Appointments/Opportunity: this is the average number of appointments performed for each opportunity. When we remember that many opportunities will be lost after the first appointment, it is unlikely that this value will be more than 1.5. Safety: it is necessary to factor some contingency into this calculation to allow for the variability that is associated with promotional activities (as well as a number of other more esoteric factors). A typical value for Safety is 1.5. Retaining existing customers In most organizations, it is assumed that salespeople should be responsible for customer-retention. The fact that salespeople are often called account managers is evidence of this. Because of this assumption, salespeople are often expected to: 1. Maintain an involvement in the delivery of the products or services that they have sold 2. Perform routine customer service tasks 3. Engage in a number of tasks that are more appropriate for personal than commercial relationships (golf days and the like) As well as distracting salespeople from selling, it’s important to recognize the hidden cost associated with salespeople performing these first two tasks. Because salespeople are not resourced to perform fulfillment or customer service tasks, in practice the quality of these tasks suffers, relative to the quality that could be delivered by: 1. Dedicated, telephone-based, customer service personnel 2. Dedicated, project leaders For this reason, we should question whether salespeople should, in fact, be held responsible for customer retention. Justin Roff-Marsh © Copyright: Ballistix 2009 16 of 21 It should be obvious that, if we divide the organization into three functions, fulfillment (including production), new-product-development and sales, the responsibility for retention should rest with the first two functions: Fulfillment: the primary determinate of retention is whether products (or services) are delivered on time, in full and free of transactional errors. This must be the responsibility of fulfillment. New-product-development: the secondary determinate of retention is the desirability of the organization’s products. It’s the responsibility of new-product-development to ensure that — at all times — the organization’s products are cheaper than, or superior to, competitors’. If fulfillment and new-product-development are doing their jobs properly, the organization will not have a retention problem. If this is the case, the only contribution that salespeople can (and should) make to retention is to re-visit existing customers and sell them new product- or service-lines (thereby increasing customers’ switching costs). The role of the sales coordinator In a lot of ways, a sales coordinator is like an executive’s personal assistant. The sales coordinator plans the salesperson’s diary and performs all of the clerical tasks associated with the opportunity-management process. The salesperson telephones the sales coordinator at the conclusion of every appointment and reports the appointment outcome. The sales coordinator does all data entry. If the salesperson has access to the CRM, it should be read-only access. It’s important to note that the sales coordinator is not a telemarketer. He or she does not generate sales opportunities; this is the responsibility of the promotional function. If it transpires that a competent sales coordinator cannot readily schedule appointments from the opportunity queue, then this indicates a problem with the offer (or the promotional campaigns). One sales coordinator can support one (and sometimes two) salespeople; never more. To appreciate the volume of work a sales coordinator must perform, reflect on the fact that the salespeople he or she supports are performing at least 20 appointments a week each. The technical team If the salesperson is focused exclusively on business-development appointments, it’s likely that other resources will need to perform activities within the opportunity-management process. We have been temporarily referring to these resources as the technical team. In practice, such a team is likely to include some of the following resources: Customer service: as suggested previously, responsibility for customer service tasks should be transferred from salespeople to a central, telephone-based team. This team should also be responsible for repeat purchases from existing customers. This means that team members must make outbound calls to secure orders, as well as receiving inbound orders. If the existing relationship with customers does not facilitate this, it is the responsibility of management to design a new relationship, and of salespeople to then sell it. Estimator: both estimation and the generation of proposals (including tender documentation) should be centralized. In many cases, the estimator will be the technical specialist (below). Justin Roff-Marsh © Copyright: Ballistix 2009 17 of 21 Technical specialist: it is often assumed that technical sales environments require salespeople to be technical specialists. This is not true. In technical environments, salespeople should be generalists. In other words, they should have the ability to recognize a wide range of opportunity types. But they must then be supported by technical specialists, who can assist with solution design, estimating, document preparation, etc. Project leader: as will be discussed below, in a major-sales environment, it is often beneficial to have experienced personnel who can bridge the gap between sales and fulfillment. In many cases, the project leader will also be the technical specialist. Major sales A common objection to this new sales process model is that it must only be applicable to minor-sales environments. To expose the foolishness of this objection, all we have to do is consider a parallel objection, this time, in a project environment. Imagine that we had two projects on which we were about to start work, the assembly of a hang-glider and the assembly of a helicopter. If we have limited project-management resources, consider which project would benefit more from these resources. Obviously, it’s the latter! If one of our staff were to object that, because of the complexity of the helicopter, standardized procedures and scientific-management methods were unnecessary, this objection would (I hope) be quickly dismissed. As was mentioned earlier, there is one important consideration in the major- sales environment. That is, the hand-over between sales and fulfillment. In a typical sales environment, the salesperson is also a technical specialist. This person takes the customer’s brief, designs the solution and then discovers that he or she must maintain an involvement in the fulfillment process because no other person has the necessary understanding of the customer’s requirements or of the proposed solution. In the new sales process model, the sales and technical responsibilities must be both divided and dovetailed. Not only must we have a salesperson and a technical specialist but, for much of the opportunity- management process, they must work closely together. The key to making this relationship work is that the technical specialist — not the salesperson — must take the customer’s brief. The salesperson should be in attendance, but it should be clear to all parties that the person who is taking the brief — and who will ultimately design the solution — is the technical specialist. In summary, then, the responsibility of the technical specialist is to ensure that a deliverable solution is designed (after all, as mentioned previously, it’s likely that this person will lead the delivery). And the responsibility of the salespersons is to ensure that the solution gets sold. As well as multiplying the salesperson’s capacity, this division-of-labor eliminates much of the conflict that otherwise occurs between sales and fulfillment. There is one important consideration in the major- sales environment. That is, the hand-over between sales and fulfillment. Justin Roff-Marsh © Copyright: Ballistix 2009 18 of 21 Measurements Management should be delighted to discover that this new sales process model results in a sales process that is easy to measure, easy to manage and, consequently, easy to scale. Division-of-labor has resulted in each resource having a single responsibility and, therefore, one measurement to which to be held accountable. It will be beneficial to review (and briefly discuss) each resource’s key measurement: Salesperson (T/ASC): the salesperson’s responsibility is simply to convert appointment slots into Throughput. Accordingly, the salesperson’s measurement is Throughput / Appointment-slot Consumed. It’s important to note that the salesperson is held responsible for neither total sales (repeat purchases are the responsibility of customer service), nor monthly Throughput (the salesperson’s unit of capacity is an appointment slot, not a month). If a sales team contains people who are selling different products (or operating in different regions) this figure may need to be normalized. Sales Coordinator (Forward-booked Days): the sales coordinator is responsible for ensuring that the appointment buffer is full at all times. Buffer size is measured in forward-booked days. Promotional Coordinator (Pending-appointment Days): the promotional coordinator is responsible for ensuring that the opportunity queue is full at all times. Queue size is measured in pending-appointment days (this is an estimate of the number of days-worth of appointments within the opportunity queue). Technical Team (Activity Cycle-time): the primary reason for measuring members of the technical team is to ensure that this team has sufficient protective capacity. For this reason, it makes sense to monitor key activities’ cycle-times. A diminishment in protective capacity will be reflected by an increase in average cycle-time. Each of these measurements should be plotted on a simple run chart. These run charts will make it easy for management to diagnose problems and make rational decisions. They will also provide team members with a valuable feedback loop. Resourcing implications If the reader has one pressing concern at this point, it’s likely to be the cost associated with the additional sales support personnel that this model requires. (Sales support refers to all sales process resources, excluding the salesperson, but including the technical team.) This is a valid concern. Even though this Sales Process Engineering will increase salespeople’s productivity by an order of magnitude, it will, in practice, take quite some time for the benefit of this increase to appear. Fortunately, operating expenses can be controlled if management follows one simple implementation rule: Initially, reduce sales capacity to match organic opportunity flow. (Under no circumstances, do the opposite.) It is entirely unrealistic for most organizations to suddenly increase opportunity flow by 10 times to match the increase in sales capacity. Justin Roff-Marsh © Copyright: Ballistix 2009 19 of 21 Accordingly, if management attempts to match opportunity flow to existing sales capacity, the following chain of events will occur: 1. Operating expenses will increase significantly (due to the additional sales support staff) 2. Significant promotional expenditure will be required (to stimulate opportunity flow) 3. Management is likely to discover that its existing offers are unappealing (possibly meaning work will need to be done on production and new-product-development processes) 4. The resulting shortage of sales opportunities will cause the constraint to shift away from salespeople, leaving them sitting idle for much of their time 5. The expected increase in Throughput will not materialize 6. Management concludes that Sales Process Engineering is a flawed concept Contrast this (undesirable) chain of events with what is likely to transpire if management applies the rule above: 1. Management measures the existing (organic) opportunity flow (all opportunities, not just qualified ones!) 2. Management calculates the number of appointment slots that this organic flow will consume 3. Management reduces the size of the sales team until it is sure that organic opportunity flow will result in the team being fully loaded on day one 4. Management transfers most members of the sales team into technical specialist, project leadership, or customer service roles, retaining only those who are genuinely prepared to perform at least 20 appointments a week (week in and week out) 5. Management only has to add enough sales coordinators to support the remaining salespeople 6. Initially, management is unlikely to have requirement for additional technical resources or promotional expenditure 7. Due to (a) the elimination of qualification and (b) more capable salespeople performing all appointments, there is likely to be a rapid increase in Throughput (in spite of the fact that conversion has dropped) 8. Management only adds salespeople when it has proven that it has the capability to ensure that they are fully-loaded with appointments from their very first week on the job Accordingly, if this simple implementation rule is followed, the initial impact on both headcount and operating expenses should be negligible. Furthermore, management will only add personnel in the future when it is convinced it will receive a rapid pay-back. In practice, the re-assignment of salespeople tends not to be met with a lot of resistance. The reality, in most sales teams, is that the greater majority of salespeople’s time is spent on non-business-development activities. For this reason, many salespeople are intimidated by the prospect of performing 20 business- development appointments a week! Salesperson compensation This, the last of the practical considerations, is likely to be the most contentious. But there’s really no reason why it should. In most organizations salespeople are compensated quite differently from other employees. But this is understandable, because — as we discovered earlier — they are not managed like normal employees. Justin Roff-Marsh © Copyright: Ballistix 2009 20 of 21 Traditionally, management tries to compromise between treating salespeople as employees and as autonomous agents (sub-contractors). Salespeople’s compensation (part salary and part commission) is entirely consistent with this. But, in this new model, the compromise is eliminated. The sales function is converted into a formal process (division-of-labor) and salespeople become employees, in every sense of the word. From this point forward, not only is performance pay (commission) unnecessary, but if it is retained, it will result in a number of undesirable effects. In this new model, the salesperson is no longer the sales function. The salesperson has become a cog — albeit an important one — in a much bigger machine. This doesn’t mean that salespeople should receive a reduction in pay (if anything, the opposite tends to be the case). The bottom line is that salespeople — like all other employees — should simply be paid their market value. Typical results The potential of this new sales process model are obvious. And there are many organizations that have — and continue — to exploit this potential. However, because this model requires a radical change to how the sales process is designed, resourced and managed, there are also organizations that fail. Following, in order of importance, are the three critical prerequisites for a successful implementation: Management maturity: management must be prepared to make a commitment and stick to it. Management must also demonstrate to team members that — one way or another — the organization will adopt the new model. Operational excellence: the organization must have strong operations. This means, it must have excellent on-time-delivery and a healthy new-product-development pipeline. As has been mentioned, a more effective sales process is not an antidote to operational problems. Growth potential: the organization must have growth potential. While this new model may produce some cost savings for organizations that do not intend to grow, it’s questionable whether these savings are significant enough to justify the degree of change required. With these prerequisites noted, let’s explore the consequences of this method for those organizations that successfully make the transition: A consistent, scalable flow of Throughput Earlier, we characterized the business as a Throughput pump. This new model enables the sales process to be properly integrated with the rest of the organization. The flow of opportunities (probable Throughput) can now be synchronized with the flow of actual Throughput in the production and distribution process. This new sales process can be readily scaled. Because the division-of-labor has simplified roles, management can add people when required, confident that they will be productive within weeks (if not days). Justin Roff-Marsh © Copyright: Ballistix 2009 21 of 21 Management can manage Without control over which resources perform what tasks in what sequence, management cannot manage (at least, not in any sensible use of the word). This new model gives management that control, in the form of centralized scheduling. Furthermore, the centralization of all clerical tasks (including data entry) ensures that management always has access to current and accurate information. Customer service improves Only the most naive of managers would truly believe that customer service is optimized by having salespeople perform customer service and fulfillment tasks (in addition to sales, solution design, tender preparation and clerical tasks). Obviously it isn’t. Organizations that implement Sales Process Engineering consistently report immediate and significant improvements in customer satisfaction. Expansion is easy Because Sales Process Engineering inevitably results in the centralization of the sales support function, many organizations have been able to expand into regions where, previously, this expansion would not have been affordable (or even economically viable). This is because the requirement for branch offices in regional areas is eliminated. Salespeople’s diaries are managed by head-office-based sales coordinators; customers are supported by the head-office-based customer-support team; technical specialists fly into regional areas only when they are required and logistics is outsourced to third-party providers. What’s more, in this situation, salespeople report that they actually feel less remote from the head office. This is because the team that plans regional salespeople’s diaries is the same team that plans the diaries of head-office-based salespeople. Throughput increases Across a range of industries, in both minor- and major-sales environments, organizations report that salespeople consistently generate between $850 and $2,000 of Throughput, per appointment-slot consumed. If we assume a conservative $900, this means each salesperson consistently generates $18,000 of Throughput each week. For most organizations, this is a significant increase in the productivity of salespeople — and an increase that has been realized with a negligible increase in operating expenses. Next steps For more information on Sales Process Engineering, please: 1. Visit (where you can browse a extensive library of articles, download a multimedia presentation or view an online video of this article’s author) 2. Join an online forum dedicated to this subject: 3. E-mail the author direct at: > >