For some reason, there’s a huge attraction to applying “manufacturing techniques” to selling. I suspect it’s the perceived orderliness to manufacturing processes and the predictability of the outcome. It’s easy to understand that attraction. The lean approaches applied to manufacturing create a hyper efficient process. Second, it always produces the same outcome (manufacturing experts will quibble, but we do design manufacturing lines to produce zero defects.) While most lean manufacturing is a “pull” process, we know that whatever we feed into the beginning of that manufacturing line will produce a very precise outcome/output.
Experienced manufacturing people, please forgive me, I’ll take some shortcuts and very small liberties with lean manufacturing approaches.
How de we design the manufacturing line? For simplicity sake, let’s assume it’s a single product we are producing.
We start with the idea of “what is it we are going to produce?” or “what’s the final product that comes off the end of the product line.” The answer to this includes the full specifications, quality expectations, pricing/cost expectations, etc. There is no sense manufacturing a product unless it meets the customer needs.
Then we look at, how we build the product. We start looking at all the steps, materials, things that we have to do to create the desired product. We start developing some ideas of, we need these raw materials at the start, we need these machines to help shape the materials, we need to add other materials at various points in the line. We start developing workstations, each dedicated to one part of the manufacturing and assembly of the product.
As we start defining those steps, we first ask, “Does this produce the product we want, at the quality, etc. that we want?” We keep tuning the process until we get the exact product, every time. If it isn’t quite what is needed, we start working backwards. Did the last step get the right partially manufactured produce they expected, were we doing the right things in that last step to produce the desired outcome?
If the answer to this is “Yes,” then we go to the step before that, asking the same questions. And we keep that process going backwards through all those steps, until we have found an corrected the manufacturing problem. Every time we have a problem, we go through the same process.
In lean manufacturing there’s the concept that each downstream workstation/process is the customer of the previous step. So as we look at the work we do for each step, we make sure it produces the “product” that customer is expecting, 100% of the time.
With this, we have the basic steps that we have to go through in building a manufacturing line.
Then in lean, we look at making that line as efficient as possible (we’ve already built it to create the quality output we want). Ideally in an efficient line, their is no build up of inventory between steps. That makes it very inefficient, things can get backed up at those steps. For the steps upstream of that, they may have to stop building, until the workers at that one step catch up. For the steps downstream, the workers are waiting for the product so they can execute their work and pass it on.
So there’s the concept that there should be zero inventory/backup between steps. This is where the concept of pull comes in. As the customer “pulls” product off the end of the assembly line, the people in the last step pull product from the previous step so they can to their work, producing what the customer wants. And this concept ripples back through the manufacturing line. When each step completes their work, they look upstream to pull the new work from that upstream process.
To prevent inventory buildups and inefficiency in the manufacturing line, the concept that each step takes the exact same amount of time to complete as every other step. If we can get this to work, we have a perfectly efficient line, as one station finishes their work and passes it downstream, the station immediately in front has finished their work and passes it to that station. Everything is perfectly synchronized through the entire line. Lean nurds call this TAKT time. Think of it as the entire line is working in a synchronized rhythm. Very often, in order to make this happen, when we are designing manufacturing lines, we have to break down certain manufacturing steps into smaller steps so we can optimize the TAKT time, making the line as efficient as possible.
When we finish the design, we have the ideal manufacturing line. Each step has the same TAKT time as every other step. Each step, produces, flawlessly, the product that is needed for the next step. The work done at each step is the same for everything going through the line, with absolutely no variation. And this is how manufacturing plants in lean manufacturing processes work.
Sometimes, certain parts of the manufacturing process couldn’t be made to fit this very clean, well defined flow. Lean manufacturers, created a number of “off-line” work cells or clusters. These would typically be teams doing as much standard work as possible, but they could adjust things or the work, slightly to produce the output. These teams created sub assemblies and those sub assemblies could be brought to the appropriate point in the line, for a very efficient manufacturing process.
A way manufacturers talk about these concepts is in the total abolition of waste. The seven forms of waste included: Correction, motion, overproduction, conveyance, inventory, processing, and waiting. Google or GPT it to learn more.
It’s amazing when you see these highly engineered and tuned lines working. I once worked with a manufacturer of mobile phones. They “owned” the phone for less that 30 minutes. That was the time it took them to convert the raw materials/parts assembling and testing a phone. They didn’t pay for the materials until they entered the manufacturing line, at the end of the manufacturing line the finished products were loaded on a truck and invoiced. So the only time they had ownership of the devices was the less than 30 minutes it took to manufacture them.
When things go wrong:
Sometimes, something goes wrong, defective products start appearing. These defects mean the line isn’t meeting the customer expectations. In these manufacturing lines, everything is stopped. The source of the problem is located. It could be a some step, someone was doing things incorrectly, or a machine got out of adjustment. All the workers congregate at that workstation looking at what went wrong, then they fix it. Getting things so the output at that step produced what the downstream step expects. These errors happen when there is a deviation in the work that is done at that workstation.
Sometimes, everything is being done correctly by the workers and machines, but defects are being produced. This means some of the raw materials may not be “in spec.” Suppliers are critical in lean manufacturing. They have to provide the raw materials that perfectly fit the specs required. There can be no deviation, no variation. If there is, another supplier who can meet the specs are found.
It turns out the dominant killer to lean manufacturing is variability. If the way the work is done in each step has very high variation, the work isn’t getting done in the way it is expected to be done. It impacts every other step in the manufacturing line.
And variability in the inputs/raw materials is deadly. If the quality of the materials change from lot to lot, if they don’t arrive on time. If they vary in any sort of way, out ability to have a high throughput, highly effective, highly efficient manufacturing process, producing the desired output goes down the tubes.
The selling manufacturing line:
It’s become fashionable in some sectors to model everything sellers do off of lean manufacturing principles. They seek the same predictability and flow that is sought in manufacturing.
As part of this, they started segmenting and specializing jobs. Marketing/demand/lead gen had their jobs, handing work to SDRs who handed their work to AEs who handed their work to Demo Geeks, and so on. Each job was highly specialized on very specific selling functions. At each selling workstation, we could precisely define the work, we created scripts, and all sorts of things to define exactly how the work at that step was done.
Assuming you could get everyone to follow the scripts with no variation you could standardize the work through the whole process. But people struggled with this.
Then there were TAKT time problems. Doing a discovery call took longer than a prospecting call to set up a meeting. Doing a demo took longer times. What we expected to accomplish in the well defined workflow couldn’t be achieved in reality. “Inventory piled up, leads weren’t being addressed as quickly as they were being produced. But demoer’s were left waiting for their next demo, so they twiddled their thumbs. Or not enough leads were being generated, so everyone downstream didn’t have work to do. Or one discovery call could be completed in 15 minutes, another 30 minutes, another an hour spread over 4 calls with different people.
So we can see in the design of the basic process flow, there were a lot of things that kept changing about getting the work that needed to be done, completed. What we designed to be perfectly predictable was no longer predictable.
The big killer of lean manufacturing, regularly raises its ugly head. What we thought we could standardize turned out to be highly variable.
Then an even worse form of variability happens. And there is nothing we can do to control this variability. In lean manufacturing, we send standard widgets down the assembly line. If I’m a manufacturing manager at an Iphone assembly line, and my suppliers send me screens or covers that are slightly off spec, those are rejected and returned to the vendor. Everything must be “in spec.”
But the widgets going down our selling assembly lines are people! Sometimes, if we are smart and disciplined, we try to minimize the variability by focusing on our ICP–but that it rare. The problem with people widgets is each one is different. While they might have similar needs and requirements, those needs and how they feel about those needs are different from customer to customer. So we are injecting, at the very beginning of our selling assembly line extremely high variability.
And as any lean manufacturing person would tell you, the moment you have high variability your scrap rate skyrockets. With high scrap rates, you don’t have enough coming out the back end of the manufacturing line to meet your goals. So to meet your goals, you shove more into the front end, generating even more scrap and waste. And all this adds to cost.
Then there’s the other deadly killer for our selling manufacturing line. While we may try to eliminate customer variation when we start the customer widgets down the assembly line, these human widgets demonstrate the most odd behavior. They change their minds, they shift directions, they shift priorities, they stop, they may have another human widget with other characteristics take their place.
So we are left with infinite, unpredictable variability at all stages of sending customer widgets down our sales manufacturing line.
We are also running into some other problems. Let’s call them supply chain problems. In lean manufacturing, we depend on high quality vendors supplying us high quality products in the volume we need, at the time we need it. Any interruption in the supply chain shuts down the line.
In our selling manufacturing line we have huge supply chain problems. Let’s say we need 100 qualified opportunities going into our selling assembly line every month. It used to be we could get that from 1000 leads. But then it got to be 2000, than 5000, then 10000. We are struggling to get enough customer widgets to put into our sales assembly line. And it’s not only us, it’s everyone that is struggling with these supply chain problems.
Hopefully, you are now getting the idea that the lean manufacturing model has so many flaws in applying it to our selling manufacturing lines. And one of the biggest problems is variability. We can control, to some degree, the variability of how the work gets done, but we have absolutely no control of the variability of the customer widgets we send down those lines. As a result we generate huge amounts of waste, which translates into skyrocketing cost and failures to achieve the manufacturing output we expected.
Some of you might say, “Dave you are being too tough, surely we can accommodate waste.” Well Yes and No. Talk to lean manufacturing managers and their mission is not to get waste to an acceptable level, it’s to eliminate it. Also, they recognize some processes don’t lend themselves precisely to lean methods, and they’ve adapted clustering and other tools to help bridge those into lean, but waste is anathema to them.
So part of me says, if that’s what the manufacturing experts do, who are we to say, “close enough is good.”
But when you look at the failure rates we are reaching in the manufacturing approach to selling, it’s insanity. With win rates of 15-20% it means we are designing our systems to produce 80-85% scrap! Think of the waste in opportunity and cost! To maintain constant response rates, we’ve had to go from 200-400 outreaches, to 1200-1500, to more in the future. Such high levels of scrap and wasted effort cannot be acceptable.
Our inability to control and minimize variability, means the lean manufacturing models we choose to model our selling approaches to will continue to fail at much higher levels and costs. There are some lean and agile principles, mainly around concepts of clustering work and adaptable work. Or principles manufacturer’s use in problem solving. Some of these can be adapted well.
We have to do better! And we can do better, and there are things we can learn from manufacturing, as well as other areas.
Leave a Reply