Paleontologists studying the process industry are positively giddy. For once, they have live specimens to examine. No dusting off tiny bones with toothbrushes this time. They can finally study living relics—process approaches with exotic names like six sigmasaurus, leanosaurus, constraintipod and controliform. Manufacturing-based process methodologies that failed to evolve as the bulk of our employment shifted from manufacturing to much more variable, office-based and service environments.

Seriously, you can look out your office door today and see what process was like 20 plus years ago—when process was often notable for its absence. And if you do see process, you’ll be looking back at least as far as 1985, when Motorola first implemented six sigma, the last popular process approach to emerge. Of course, six sigma and Lean live on in manufacturing. And yes, they’ve both grown incrementally over the years. Practitioners have even taken crowbars to them, trying to adapt process methods designed for manufacturing to fit into “variable” environments, as we’ll call the whole spectrum of non-production settings. But for all their efforts, they’re still pounding square pegs into round holes. Working from scratch, you’d never design a process approach for use in variable environments on a six sigma or Lean platform.

Moreover, the evolution in workplace settings is hardly the only change that left 1980s process approaches behind. Think back to 1985. Little chance you had a PC. But if you did, you probably had but one application—word-processing. If you were lucky, a spread sheet, too. How many icons show on your desktop today? And guaranteed you weren’t connected to a network. Although we did have sneakernets back then for spreading viruses. The internet? It was still a brain fart of rocket scientists wanting to collaborate and share data with other rocket scientists. Windows? Forget it. My company had just switched from CPM to DOS the year prior. And the staff had to drag me kicking and screaming because CPM was so much better. Now let’s stride out onto the factory floor. Robots? Only in comic books. Buffer parts inventories? How do you spell that? Managing variances in terms of infinitesimal deviations from the mean rather than whole percentage points? What planet are we talking about? And shifting over to the warehouse, automated warehouse management systems? What’s wrong with quills and scrolls?

I hope I’m not being overly subtle here, but do you perhaps get the sense that we don’t work the way we did in 1985? And might you suspect that circa mid-1980s process approaches, especially manufacturing-based approaches, might not factor in all our new, variable environment tools and capabilities?

Outside of manufacturing, the work world of 1985 bears virtually no resemblance to our variable work environments today. Application software, the Internet, sophisticated data networks, collaborative data sharing and an adult dose of demographic and psychographic work force changes have revolutionized how we work. And to further differentiate yesterday from today, today we understand—at least those of us not pledging allegiance to one process approach under God understand—that designing business process for variable environments takes very different techniques and skill sets than designing the manufacturing side. The work of one bears little resemblance to the other. And neither do the workers.

That’s why it’s long past time for us to design, from the ground up, process approaches capable of addressing variable process needs. Not only will we create the capability to cut costs, increase efficiency, raise effectiveness and reduce cycle times throughout companies—but perhaps even more importantly, we’ll have the opportunity to properly align variable environment processes with the new, customer-focused business strategies many companies are scrambling to adopt to stay competitive.

Any takers?

By Dick Lee, High-Yield Methods
[This is the second in a series of posts I’ll be making using edited excerpts from my new book: “Process to the People: delivering process wellness outside of manufacturing.” My goal is to complete writing by mid-2008.]
Several years ago, I was contacted by division of a Fortune 100 global manufacturer and asked to roll out Visual Workflow to solve some nasty variable (non-manufacturing) process problems. The company, which crawled with black belts and green belts, had been applying six sigma for several years to find and fix these process defects, but to no avail. Regardless, we still had to pass muster with this division’s representative on the global six sigma council before we could bring a relatively new and unknown (to them) process technique  into their six sigma environment.

So I met with this stern looking woman who took me aback by asking me directly, without any preceding pleasantries, “Why should we let you in here with this Visual Workflow stuff we’ve never heard of, when we’re the poster company for six sigma?” I had to scramble for words, but I recovered and said: “Because Visual Workflow is like an MRI, while six sigma is like surgery, and you don’t go looking for broken bones with surgery.” 

After the words spilled out of my mouth, I almost swallowed my tongue in fear of having crossed the line. Until she burst out laughing and said, “Okay, I get it.” From there we had an excellent conversation about our experience finding most variable process defects embedded in how work and information flow from person to person, function to function, and between the company and customers, suppliers and the like. Most variable process defects are in the handoffs of work and information, or as we call it “the seams.” That’s a 180° apart from six sigma looking for manufacturing defects in the way individuals perform their own work.

Driving almost all these differences is the nature of variable work—which is almost all information-based, making information management as significant a process element as work design. Plus, while manufacturing works on a “do it, pass it along” system, accomplishing most variable work requires cross-functional cooperation and collaboration. You can’t design variable work one function at a time. And you can’t measure it one function at a time.

Still, we haven’t yet touched on the most fundamental difference between manufacturing process and variable process. The need for alignment.

Manufacturing process operates quite independently of the rest of the company. When business strategy changes, manufacturing doesn’t have to change in response—unless it has to make new products or better products or cheaper products. But these adjustments don’t affect core manufacturing process. And if manufacturing process does change, the changes rarely affect strategy. Manufacturing-specific technology may have to change to compensate—but enterprise technology driving the company? Not a problem. So alignment with other functions is almost a moot point.

In contrast, variable process is highly interdependent. When business strategy changes, as so often happens today in the scramble by companies to stay competitive by becoming more customer-centric, process must change accordingly. Otherwise, companies wind up acting out Freud’s definition of insanity:

Doing the same things over and over again but expecting different outcomes.
Variable process should work like a corporate transmission to convert business strategy into work—the right work to produce the desired strategic outcomes. That ties variable process effectiveness to alignment with business strategies. And when strategy changes trigger process changes, the process changes in turn trigger the need to change technology support. This ties technology effectiveness to alignment with process —both for systems architecture, which controls information flow, information accessibility and data integration, and for application software.

How can companies achieve alignment using manufacturing approaches that treat technology as an afterthought? Beats me. And beats them, too, because they don’t get to alignment. Aligning the “moving parts” that make up the variable components of companies is a fundamental variable process design requirement. And trying to accomplish variable environment alignment with manufacturing process tools is like looking for broken bones with a hammer.

Would you do that to a company?

Many companies make a fundamental error at the start of redesigning business process to support customer-alignment. They call over to manufacturing. Or they may call a six sigma or Lean process consultant. Or, they may even call IT. Regardless, the results are the same. Manufacturing (or systems) process denizens wind up applying inappropriate process design approaches in customer-facing areas—which often does more harm than good, and at best produces mediocre results.

How do you know if this scenario is about to do a tap dance on your head? Look for these ten key elements the process design approach you’re about to apply to customer-facing functions—in fact, to all variable (non-manufacturing) work environments.

1. Start with customer-aligned strategies.  “Customer-aligned process” means business process fully supports customer-aligned business strategies. If you don’t align strategies with process, by default you’ll wind up designing for efficiency, which almost always means taking value away from customers rather than adding value. Ironically, because taking the customer-aligned approach creates the freedom to restructure rather than just trim around the edges, it almost always produces more efficiency than designing for efficiency.
   
2. You need a workflow scanning tool.  Workflow describes the movement of work and information from work station to work station, function to function, and between functions and external stakeholders—like customers. Compared to variable work environments, including front office functions, manufacturing has relatively few key workflows—so few that there’s no need for a scanning tool. But assessing variable work environments often including hundreds of workflows requires a powerful and fast workflow assessment tool.
   
3. Workflow redesign must be coupled with information flow redesign.  In variable environments such as sales, marketing and customer care, work means managing and communicating information. Work and information are tied at the umbilical cord. So if you optimize workflow, it only makes sense that you have to change information flow with it. Manufacturing process methods don’t need to factor in information flow because work is performed on physical objects, not information.
   
4. Workflow must be uncoupled from individual work process.  In variable environments, the vast majority of work impediments, time loss, work errors and other defects—especially failure to deliver value to customers—occur at the workflow level, not at the individual work process level (which describes how individuals do their own work). So why start off looking at very detailed work process, as Six Sigma does? You’ll only lose the forest for the trees. Further, you’ll proceed at a snail’s pace, which you can’t afford because variable environments include so many flows.
   
5. The drill-down to work process occurs after workflow is redesigned.  Individual work process is a dependent variable in variable environments. It has to support the workflow above. In manufacturing, work process is more independent. And by the way, Lean has no mechanism to reengineer and document individual work process, which is essential for the front office and most other variable settings.
   
6. The process approach has to “specify” systems architecture changes.  Please don’t get thrown by the fancy “systems architecture” term. Systems architecture describes how the various technology components are arrayed and connected—which is vital to getting the right information to the right place at the right time. Of course, you don’t hand IT a piece of paper and say: “Do this.” IT should be involved in workflow redesign so they’re already up to speed on what’s happening. But information flow must follow workflow, not the other way around, making your redesigned workflow a blueprint for IT to follow.
   
7. The work process reengineering must lead to application software requirements.  Variable workers need application software tools for managing and communicating information. And if the work process design doesn’t produce software requirements, you’ll inevitably wind up with software that’s misaligned with work. Manufacturing process folks aren’t concerned with application software.
   
8. Business process redesign, including both workflow and work process, has to be participative.  On the shop floor, people are cogs in a wheel. They have little choice but to conform. Think about sales. Does sales “have to conform” to process changes? Not in our lifetime. And that’s true of almost all variable settings. Variable workers must be involved in designing their own changes. Otherwise, you’ll raise a brick wall of resistance.
   
9. Process design must include change management strategies.  While manufacturing workstations are “work islands,” variable functions have great interdependency. A change in one may rebound all over the company. Plus, because of the complexity of variable environments, process change inevitably triggers unintended consequences that must be carefully managed. Companies have to carefully plan and stage change—and the planning has to emanate from the process work.
   
10. The process methodology has to be accessible to all.  Because variable process change is a participative endeavor involving mostly “non-process” people, process symbology is out. So is specialized process vocabulary. And training a special cadre of non-process people in a process approach only creates “process police,” who try to take away variable worker empowerment—often with disastrous outcomes.

Making sure your process methodology observes these nine basic rules will prevent you from trying to drive square process pegs into round process environments.

Bad process costs companies big money. But until recently, business as a whole paid little heed to process improvement outside manufacturing. And the business community might still be ignoring front office process had CRM not arrived on the scene, with needs for front office process redesign that couldn’t be ignored (okay, sometimes they were ignored, but some companies aren’t too bright). But let’s not celebrate yet. Business may not be ignoring front office process these days, but it’s not paying all that much attention, either.

So, I thought reeling off a few common front office process gaffes we encounter and their costs might help create more focus out front—particularly on the B2B side where we do lots of our work

1.     Three-level customer service systems:  Microsoft, Intuit and a host of other high profile customer-service operations take the three-tier approach. Too bad, because while it sure alienates customers, it also costs more.
Here’s the three-tier drill. Customers call for service, let’s say tech support. Partially-trained, poorly paid first line agents take the call. But these agents can’t resolve a high percentage of problems. So callers either hang up disgruntled, or if they really need an answer, they demand to speak to someone more qualified.

If the level-one agent is willing to escalate, and some aren’t, callers are put through to “supervisors.” These blokes make callers repeat all the same information they already gave to first-line agents. And then (and I love this part), they just patch through the caller to a higher tech tech. Value-adding process, or what? Any time you see a traffic cop in the middle of a process, the red flag should run up the pole. Regardless, now callers who manage to get this far get to repeat everything for a third time—after which they learn that “higher tech” is only relative, and they still may not get a solution.

Ugly from the caller’s standpoint. But also ugly from the company’s standpoint because these efforts to save money by chintzing on service only increase labor costs, rather than reducing them. .

Contrast this scenario with Amazon, which very smartly emphasizes first call resolution. And the Qwest DSL folks, who may need to escalate, but handle a much higher percentage of calls without kicking them up. And when they are escalated, there’s no “supervisor” middleman, and all the caller input data is forwarded. Or Verizon wireless, where the only time I’ve ever needed an escalation turned out to be a Palm PDA problem, and the Verizon rep set up a three way conference with Palm and actually stayed on the line—a powerful reason why Verizon doesn’t need to spend $500 or so in marketing money to replace me..

This stuff is hardly rocket science. But to save money and make customers much happier, companies have to invest more in training plus empower their first level agents. Most do neither.

2.     Mismanaging sales lead campaigns:  Most sales lead campaigns generate more optimism than revenue. And when they do produce substantial revenue, most companies don’t know how much. No surprise, because lead campaign “management” normally consists of: marketing generating sales inquiries in one process; marketing dumping raw inquiries out to sales, a non-process; followed by sales following up (supposedly) in a disconnected process (although throwing inquiries in the can is hardly a process, either) . What an unfortunate place to use simplistic process approaches, because successful B2B sales lead campaigns succeed because of intricate, highly controlled and measured process approaches.

Recently, our long time client, lead management company Performark, shared with us the outcomes of over 100,000 sales inquiries followed up using a variety process approaches—ranging from the “three half-steps and a prayer” described above to campaigns that follow inquiry generation with inquiry qualification; then divide “qualified” inquiries into two pools, immediate need and future potential; send the immediate leads out to sales; hold the future potential leads for nurture marketing; with sales following up leads only when they’re ripe and ready to close; plus sales “closing the loop” by feeding back to marketing all sales call outcomes; and then (we’re not through yet) marketing or sales management constantly but gently monitoring sales activity to maintain accountability for following up leads as well as feeding back data. All this last little activity accomplishes is doubling the ROI on some camaigns.

Guess what? The “wing and a prayer” process produces but a tiny fraction of the sales and profits the more sophisticated process produces. And as it turns out, all the cost “saved” by skipping so many process steps is a pimple on an elephant compared to the wasted sales cost incurred by sales reps following up raw inquiries—and to the opportunity cost of reps throwing sales potential into the round file.

3.     Order-entry:  If I could have a free cup of coffee every time a data entry clerk has to reenter data that someone else could more readily enter somewhere else; or has to enter data that someone else could have readily entered upstream; or has to calculate data that could have been machine-calculated at the original data entry point—Starbucks would go under in two or three minutes. Same applies for pricing clerks entering or price-checking orders.
You’d think this issue is last millennium stuff. At least you’d hope it would be. Unfortunately it’s not. Well over half the time we assess order-entry and pricing processes, workflow maps are dotted with “red keyboards” (our visual symbol for redundant data entry). Too costly to fix? How about calculating the cost created by data entry errors?

4.     Credit approval:  Using online data and readily-available automation tools can turn credit approval into a virtually hands-free (for the company), customer self-service process. Instead, many companies have redundant redundancies piled on top of redundancies. They’re called “checks and balances.” But as often as not, they’re job protection.
One of these situations inspired me to include in a client management presentation a slide saying, “You guys are spending millions to save pennies.” The next day the COO posted enlarged copies all over the executive suite. It became the process team rallying cry. Risk aversion is good policy. But risk avoidance almost always costs way more than it saves.
5.     Sales proposal generation:  Most sales proposal processes make me want to cry, they’re so manual. Wonderful automation tools are readily available. And today, they’re cheap. But many companies haven’t even developed Word templates for frequently used text. Hey, it’s cheaper to have sales people recreate stuff from scratch, right?

We recently worked with a client with the most convoluted proposal process we’ve ever seen. Unfortunately, for regulatory reasons their initial proposals often grow to 600 to 700 pages. And then clients require multiple revisions—even before awarding the job. When you start changing items and item prices in complex proposals, changes in one place have to cascade all the way down line, at least they’d better cascade. Want to risk doing this manually?

During our engagement an RFP issuer called to ask which proposal figure to use—the final number or the sum of all the components, which was $250,000 higher. And that wasn’t all that unusual. They could have automated their proposal generation processes for 1% of that.

6.     The “all hands on deck” approach to key customer service:  This is a hard to resist temptation, in addition to sheer absence of process. When a key customer suffers a significant service problem, regardless of industry, customer-friendly companies are frequently tempted to throw all available resources at the problem. This makes the company feel like it’s “doing all it can;” plus it provides customers with immediate gratification from seeing sellers jump. However, when problems don’t get solved quickly, immediate gratification turns into long-term disenchantment, or even worse.
Hey, do you believe in solving urgent problems by committee? Great way to make a mess.

Companies that solve service problems quickly and efficiently usually do so because they carefully design service response processes. They carefully assess how best to solve emergency problems—and design response processes to resolve issues as quickly as possible, not to throw as many bodies as possible into the breach. Having everyone and their brother or sister, including C-level execs, crawling all over issues leads to mayhem and compounding original problems—not quick resolution.

I could go on forever identifying common, front-office process issues. But this is a blog, not a book, and jumping on whichever of these six issues apply is a good start.

But one final caveat. Unlike in manufacturing, front office process is highly cross-functional and inter-dependent. Using the manufacturing process approach of isolating specific process problems to fix them is disastrous in the front office. Whenever you change a front office process, you have to consider the entire context, including affected activities outside of functional boundaries.

Last month, I underwent planned but significant surgery. In the process, I “enjoyed” a bird’s eye view of the U.S. health care system, which in itself looks like a bird. A vulture. Makes me wonder, in fact, whether there’s room for healthcare and customer-care to co-exist, considering the aggregate narcissism and greed of many players in our healthcare system.

Here’s just a sampling of the recent goings-on.

§         I was referred to a “reputable” orthopedic surgeon in a multi-site practice because of pain in both shoulders. Following a five-minute “history” and a subsequent MRI, Doc McChopShop told me both shoulders urgently needed rotator cuff surgery, with the left one having priority. After five minutes explaining the supposed MRI outcomes (without me seeing them) he instructed me to go to his scheduling nurse’s office to “get on the calendar.” The only place I went was for a second opinion.

§         Next I saw Dr. McShoulder, who happens to minister to the delicate arms of a major league baseball team. Looking at the same MRIs, but walking me through them, he showed me that my right shoulder was a mess. However, considering my age, he didn’t see much value in surgery to correct a long-ago high school pitching injury to my left shoulder, since it doesn’t seem to be getting any worse over time and can be dealt with through physical therapy and occasional meds. But he wanted to do the right shoulder “yesterday.” Unfortunately, his next open surgery slot at the only hospital where he has privileges and I have insurance coverage was In three months. So “yesterday” turned into three months.

§         Prior to surgery, McShoulder’s PA (physician’s assistant) referred me to a PT (physical therapist) who also specializes in shoulders so I could loosen things up and manage pain before getting cut. When I arrived for my first session, Mr. PT-shoulder told me, “you’ll be working with APT (assistant physical therapist) so-and-so, after which I barely saw the PT except when the APT had to ask him what to do. Yah sure. Helluva recommendation.

§         When Dr. McShoulder finally operated, he discovered that things were in much worse shape than shown on Dr. McChopShop’s MRI (health insurance wouldn’t pay for a second MRI), very likely because my right shoulder had continued deteriorating over the three-month interval. Consequently, he had to perform a different and much more extensive surgery than anticipated. He informed my wife immediately after, and she caught some of what he was saying, but he didn’t talk to me until my post-op appointment 10 days later. That might have been okay, except in his haste he forgot to change my post-op care instructions. So I did potentially injurious stuff until I saw him. Fortunately, I survived (or rather my shoulder did).

§         And then there was the nursing. I did have several knowledgeable and helpful nurses during my stay. During the day shift. But overnight, nothing but NAs (nursing assistants) who were left on their own to dispense and adjust serious pain medication, despite an obvious lack of training or nursing temprament. And then there was the recovery room nurse who kept me in the post-op room for about an hour while chatting up everyone in sight. Too bad they don’t serve post-op pain killers there. And when I asked her when I was going to my luxurious suite, she responded, “When I have time.”

I’ll leave things there, without regaling you with stories like my previous physician declining to authorize a sleep apnea test because: “You just need to lose weight.” And C-Pap machines (the common remedy for sleep apnea) rarely work, anyway.” When he finally did authorize the test because I couldn’t get through the workday, the hospital started recording “normal” sleep patterns, only to stop in less than an hour because my blood oxygen level was “in the red zone,” putting me at risk of a stroke or heart attack. And when I got to the pulmonary clinic to get set up with a C-Pap machine, the pulmonary PA just exploded hearing the story, saying (literally): All the jerk had to do was point a flashlight into your mouth to see that this has nothing to do with weight.” All this after I’d had to put my consulting on the back burner for a considerable time while I tried to get healthy (I recovered completely in a week using the C-Pap machine – and fortunately they don’t count against carry-on baggage limits).

Okay, I’m regaling. But I’ll stop after saying that in our cost-cutting times when doctors and clinics are rewarded for keeping patients away from expensive tests, two different doctors told me my right shoulder problem stemmed from “arthritis.”

Okay, so why did I belch out all this sordid stuff? Because I’d like you to wear your patient hat while you try to answer two questions. First, “How would CRM software functionality (imbedded in medical practice systems, of course) fix any of this?”

Huuuuhhhh??

You betcha. We actually have CRM yahoos out there pushing use of CRM software functionality to improve relationships with patients But why should healthcare be any different. Isn’t it always about software?

Second, what, in your opinion, will it take to put the patient in the center of the health care circle (well-applied 2X4s aside)? Personally, I don’t believe our health care system is capable of behaving better of its own volition. I believe reform will only result from customer push-back – as in constantly reminding practitioners that we’re customers, not patients. And very vocally taking our business elsewhere when they can’t grasp that. Moreover, when some surgeon tries to pay for his daughter’s wedding or whatever by cutting open both my shoulders, starting with the wrong one (do you know that statistical research proves that Ob-Gyns are far more likely to deliver via c-section when they have a daughter getting married?), we’d better start filing complaints – with clinics, hospitals, insurance companies, HMOs and even state medical boards. Which is exactly what I’m doing.

Or do you think I’m being too cynical?