Why Operational improvement and change initiatives usually fail.

Why Operational improvement and change initiatives usually fail.

How do you make short term operational and process improvements ‘stick’ for the long term?

Most change initiatives fail to deliver on their early promise. You get some short term improvement, some changes made, but the effectiveness of the process dwindles with time.

I often see failed improvement initiatives, usually labelled ‘Lean” or ‘6 Sigma’ by those involved, that leave a pile of paper, some awareness and knowledge, and from time to time some useful results, but nothing like the promises of the expensive consultants as they signed you up.

Why is that?

Nobody goes into a change process expecting it to fail

In my observation, the single most common reason these initiatives fail is because they ignore one of the basic tenets of Lean: respect for people.

Lean gets a start because management sees problems they have failed to solve, or do not know how to solve. So they bring in some Lean consultants who reach into the tool box and come out with some of the common tools, go through an education process, implement, and get some quick and sometimes impressive wins, and victory is declared. After that declaration, the focus moves elsewhere,  and the process slowly deteriorates.

Why is that?

Everyone was so committed, excited at the early results, the consultants were paid a shedload, so it should have worked.

In 30 years of doing this stuff, there is always one dominant reason they fail.

The initiative is top down, not bottom up.

Those at the top see problems manifest in the P&L. Their motivations are financial, operational and strategic. They talk about alignment, and people being the most valuable asset, then ignore them.

By contrast, building initiative from the bottom, asking those doing the work how to improve it, then giving them the tools to improve, and rewarding them with acknowledgement as well as a more secure job and maybe a pay rise, is where the action is.

However, for managers, they are trained to see their job as managing. Having some stuff bubbling up from the factory that has not gone through the formal approval processes and subjected to the discipline of  the accountants mandatory NPV  and ROI analysis is uncomfortable and challenging to their authority as managers.

This is where the distinction between managers and leaders comes in.

Managers, usually unwittingly, kill off the grass roots enthusiasm to make their workplace safer, more interesting, and more productive because it makes them uncomfortable, less in control.  By imposing rules, they interrupt the productive flow evident in successful initiatives. By contrast,  leaders encourage and promote the ambiguity that sometimes results, and works with it.

Which are you, Manager or Leader?



The simple 4 letter word that underpins every improvement initiative.

The simple 4 letter word that underpins every improvement initiative.

Improving the performance of businesses is often like being set loose in a commercial kitchen without a recipe. Random ingredients, absence of some staples, disaffected staff, erratic processes, and severe cost pressures, but still being expected to produce an experience people are prepared to pay for.

Not easy

However, every time I look back on a project, the common factor that has made the most difference is not what you would expect.

It is not the financials, or the marketing plan, or how well the sales force performed, it is more basic than all that, and enables all those things:


Simple word, and an idea at the core of all performance improvement.

The concept of flow emerged from the work done to improve manufacturing processes by W. Edwards Deeming, Joseph Juran, and others, and was first widely implemented and documented by Toyota, then spread around the world as ‘Lean thinking’ and the ‘Toyota Production System’.

At the core of Lean is Flow, and at the core of any improvement in any process, physical or otherwise,  in any context, is flow.

The basic confusion is between being busy, and being productive. Jumping up and down in one spot may be  busy, but it is hardly productive unless you are killing ants.

Optimising flow in manufacturing operations requires the configuration of all the lines such that work passes unobstructed from job centre to job centre through to completion. The faster and more uninterrupted the flow, the higher the output.

Flow optimisation always requires the counter intuitive decision to leave unused capacity at points in the process, to avoid building Work in progress inventory, which act as ‘rapids’ in the flow metaphor. It usually feels wrong to leave available capacity unused, but the slowest work centre will be the limiting factor for  the whole process, and to keep the flow steady, the flow rate is limited by that slowest point. In addition, shit always happens, something breaks, an item spec ‘wanders, ingredient fails to come in as required, so there is always downtime of some sort. This means that  some spare capacity in the system is a requirement for  the flow to be matched to demand, or ‘Pull’ in Lean parlance.

A key component of flow is the orderly release of work into the process. A schedule is written based on priority and optimal flow, and is then executed without change. Queue jumping, to meet unscheduled customer expectations, is a common distraction from the plan that multiplies, disrupts everything, and often results in total turmoil in the flow. It is deadly to process optimisation.

These days, not as much manufacturing is done, after all many of us are told we are now knowledge workers.

Exactly the same principals apply. While it may be harder to see because there is no physical product moving down a production line, the thought process is identical.

The trouble with these non physical tasks, is that they come at us from every direction, often with little warning and lead time, and with ambiguous importance and priority. Unscheduled demands on our time.

How do you sort through the mess to optimise your productivity?

A now standard method is the scaling of Importance and urgency into quadrants. When analysing how our time is spent, most of us find that too much is spent in the not important/urgent quadrant, when we should be focusing on the important items, urgent or otherwise. It is almost always the important/not urgent tasks that get shuffled aside, and it is these items that have the greatest long term impact on the performance of an enterprise.

An alternative means to allocate time is on an ‘Impact/Effort’ continuum. Tasks that are high impact, low effort are the quick wins so beloved of consultants, by contrast, high effort, low impact tasks are just thankless tasks, and not worth doing.

Everyone is in charge of managing their time to some extent, the further away you are from a time driven physical process, the greater the amount if discretion you will have. It behoves you to work the tasks in front of you in order of priority. Responding to that email may seem important, after all it has come in, the ‘new email bleep’ (Pavlov would love this one were he still alive) has sounded, there is a sense of urgency generated, but in 99% of cases, what does it really matter of the email goes unopened.

In everything you do, consider the impact and benefits of optimising Flow.

Photo credit  Dirk Veltkamp: Thredbo river.


The differences between Takt and cycle time, and why they are important.

The differences between Takt and cycle time, and why they are important.


‘Job shops’ have particular challenges in production planning and capacity utilisation.

No two jobs are exactly the same, so the sequencing of jobs to optimise factory utilisation takes on even more importance.

In the middle of an operational improvement project in a job shop environment with multiple possible routes for a given job, I found myself having a regular conversation with the factory management and machine operators about creating a sense of ‘flow’ through the factory, and how that related to machine cycle time, total process time,  and Takt time.

There were multiple definitions of cycle, depending on who I was talking to, but none had any idea of what Takt time was.

Therefore I followed my own advice to ensure that as a first step, the language in any factory was absolutely common. This ensures that the meaning meant to be conveyed by specific words were exactly the way they were received.

Cycle time.

Cycle time has many definitions depending on circumstances. In effect it is the time taken to do one cycle of a specified task. The cycle time of production of a Boeing 747 is many months, whereas the cycle time for production of a specific part may be hours. Both are valid definitions, so ensuring that the definitions being used in your context are exactly the same is essential.

In my job shop, there are many routes for a completed job via a range of different machinery, each with its own limitations.

A job may be done on a machine that produces  5/minute, a cycle time of 12 seconds, or a machine that does 60/minute, a cycle time of 1 second. Make sure you are talking of the same machine when discussing the routing.

The cycle time of a particular completed job may have a number of separate processes that require both a specific order and potentially variable routing, so knowing the cycle times of each machine option, and other limitations, such as manning, that may apply is essential.

Takt time.  (derived from the German ‘Taktzeit’ referring to a musical beat measured by a metronome)

Takt time is similarly subject to differing interpretations, but less so than cycle time.

The classic definition of Takt time is:

Available time for production/Required units for production.

This is exactly right, but the confusion usually occurs in the definition of ‘available time’

Assuming a normal 8 hour day, we start with 8 hours X 60 minutes = 480 minutes.

However, there will also be standard times during the day when the machine is not available. For example, it may take 15 minutes to be set up in the morning, then there are 2 x 10 minute ‘smoko’ breaks during the day, a 30 minute lunch break, and a 40 minute wash up at the end of the day.

The available time then becomes: 480 – 15 – (2 x 10)-30-40 = 375 minutes/day.

This ‘down time’ might be managed by having split times for lunch and start-up/wash up, and would change the Takt time calculation, but essentially to be simple there are 375 minutes available in the day. This becomes the base for the Takt time calculations.

Let’s assume that the customer demand was for 50 units. You then have 375 minutes to produce 50 completed products, or an available time/unit of 7.5 minutes.

Let’s further assume that the product takes  8 minutes /unit to produce, you are therefore  30 seconds short per unit, or 25 minutes over the course of a normal day. This is typically made up with overtime used to produce the 50th unit, or producing only 49 units, which annoys the customer of the 50th who are short or late delivered.

The additional problem is the backlog of back orders builds up, resulting in customers cancelling, going elsewhere, or just losing confidence in you.

None are outcomes you would wish for.

In addition, there are always unexpected things that happen, a machine goes down, a well-meaning manager walks through talking to operators and slows down the speed, a productivity improvement meeting is called, and so on. All this impacts on the available time, but the customer demand does not vary from 50 units/day.

The operational improvement task, often referred to as ‘Lean Thinking’ is to apply continuous improvement to both the available time, and time required to produce the unit, so that it eventually matches or exceeds the Takt time available.

Most factories I have seen have no idea of takt time, and many see no need for it, but it has several benefits.

  • It makes capacity calculations relatively easier, even through a complex set of sub processes. By calculating the capacity and cycle time of each process, and the alternative routes that may be available, the best fit to the demand is exposed.
  • It enables the calculation of the best batch sizes to be done.
  • It gives team members an idea of what ‘well done’ looks like for every bit of production, an enormously valuable outcome.
  • It acts as an early warning system of an emerging problem.

Understanding the two times, Cycle and Takt enables a clear sense of priority that can be applied to planning the sequences of jobs, and an understanding the total and local (individual process) capacity utilisation necessary  to be of maximum service to customers.

When you have that clarity, you can address the opportunities for improvement, leading to a greater sense of ‘flow’ through the factory.


The greatest failure of marketing management.

The greatest failure of marketing management.


The headline is a big call, competing as it does with some real doozies.

Remember “New Coke” or the Ford ‘Edsel‘ or perhaps Thomas Watsons declaration that there was a world market of no more than 5 for computers?

This one is a general observation on the nature of marketing people generally and their project management skills, at least in my experience.

They tend to spend too little time really defining a problem, but then jump effortlessly to a conclusion, leaving a pile of crap in their wake for others to clean up, and sub optimal outcomes from projects.

The explosion of marketing technology is not just making the shortcomings more obvious, it is delivering the means to measure it.

Holy cow Batman: Accountability!

The smoke and mirrors are being removed, leaving many self declared marketing gurus naked.

Leaving aside the question of individual capability, the root cause of this can usually be pinned down to a failure of project planning.

Specifically the failure to recognise the nature of critical activities that are sequential, building on the one before. For example, only a fool would lock into a creative approach and copy before the persona of the target market was absolutely crystal clear.

This notion is entirely different to the usual ‘critical path’ which can be a moveable feast as timetables move around.

Critical activities are just that, and they do not move around, at all. Project planning should always acknowledge the time necessary to complete each critical activity, and the specific sequence that is necessary.

Marketing project planning is no different to planning any other context, although the questions to be answered are usually less black and white, which simply means that the planning process needs to be rigorous and scientific if it is to be any good.

Marketers have a lot to learn from the manufacturing end of the lean movement.

Marketing’s lean future

Marketing’s lean future

The best marketers I employed in days past as a honcho in a corporate environment were pretty much always from a professional background of some type. Those with engineering, science, architecture training, and with a bit of sales thrown in were almost always better marketers than those with marketing degrees.

Having just a 40 year old (had anybody heard of marketing 40 years ago) marketing degree, this both intrigued and disturbed me, and it took a while to understand “why” it was so. It became clear that those trained in the professions had an instinctive approach to problem solving, commonly called the ‘scientific method’. By contrast, those with just marketing degrees tended to jump from problem definition straight to a final solution, missing all the nuances and understanding to be gained from the interim steps, and often getting it horribly wrong as a result.

“Lean thinking” 35 years ago was just a babe, an approach to operational improvement that has transformed the way manufacturing operations around the world are run, and delivered huge benefits to our way of life. It is absolutely based on the scientific method.

More recently however, it has been realised that lean thinking has applications far wider than just operations, and office work flows are rapidly transforming as a result. The next obvious cab off the rank is marketing, traditionally a qualitative, smoke and mirrors part of a business, and marketers have been their own worst enemy.

It often seems marketers sit around, drink coffee, and go to lunch a lot, and as a result of allowing that perception to survive, (and it has been true in an unfortunate number of cases) , we get what we deserve.

If people do not understand the role of marketing, particularly those who allocate resources and measure performance, no wonder there are problems, but there are some pretty simple solutions taken from lean thinking.

Have an objective

Map the processes

Form hypothesis

Test and measure

Rinse and repeat.

From the perspective of managing this process, it is essential that two criteria be met:

  1. There is great transparency of the whole process across functions, everyone these days is “in marketing”. Leveraging the resources available by the collaboration enabled by technology will evolve as standard practice, for which transparency is essential.
  2. It is OK to be wrong, you can learn more for your mistakes than from what works well, but to learn  there must be understanding of the flaws in the tested hypothesis coming from the failure. In addition, there needs to be a robust process of due diligence both before and after the experiments are done. Saying it is OK to be wrong is not a license to be sloppy, as I have seen from time to time.

Marketing has been very late to the technology table, and as a result the size of the gap between the ‘leading edge’ and what most small and medium businesses are doing is huge, and getting progressively ‘huger’. There is a real danger of just leaving it in the too hard basket, but that would be a mistake.

Marketing technology offers the opportunity to leverage resources and widen the impact, the core principals of marketing remain unchanged, indeed become more important as we increase the leverage that can be applied. For small and medium businesses these technologies are the competitive tool-box they have been seeking.

It is my prediction that inside a decade technology decisions will be driven by marketing. If you cannot locate, understand, engage, and deliver value to your customers in a digital world, you will not survive. This post from the ‘Martech’ guru Scott Brinker detailing the 2016 marketing technology landscape says it all.

Lets see how that goes as the technology landscape continues to evolve.

The least understood large cost impacting small business: Transaction cost.

Image courtesy of ddpavumba at FreeDigitalPhotos.net

Image courtesy of ddpavumba at FreeDigitalPhotos.net

This post is the sixth in the series that sets out the means by which small businesses can take advantage of their small scale, and be successful competing against the industry giants for expensive supermarket shelf space.

Remove transaction costs.  Easy to say, hard to do.

The concept of transactions costs is generally attributed to British Nobel prize winning economist Ronald Coase, and the publication of his 1937 paper “The nature of the firm”

Transaction costs will always be present, they are the enablers of an organisation. The challenge is squeezing the maximum productivity out of the transaction costs you will inevitably incur.

Like all costs, transaction costs fall into three categories:

      1. Those that are necessary for the sale, and that add value to the customer, so they would be willing, if you asked them (and this is the big test) to pay for it. Things like delivery of physical products fall here, and we all know  there is no such thing as “cost free delivery”. ,
      2. Those that are necessary, but do not add value to the customer. Costs associated with compliance, your training and innovation programs, taxes and charges all fall here .
      3. Those costs incurred that do not add value in any way, just consume time and money, such as rework, picking up wrong deliveries, or correcting wrong invoices. You generally do not need an activity costing initiative to know that this third category is usually uncomfortably large, and should be eliminated.

The bloating of transaction costs has three basic causes:

      1.  Not getting “it right first time” requiring rework to correct the mistake. For small businesses, the costs of mistakes are relatively much harder to absorb than they are for a large enterprise.
      2. The penalty of small scale, expressed in the variable operational costs incurred, and the productivity per   dollar of overhead spent. The flip side is that small operations can be far more agile than large ones, as the distance between a decision being made and actually getting something done, is much shorter.
      3. Less than optimum processes, or the ways that businesses manage the things that need to be done to support and document a transaction.

If you chose to take a deeper look at these three causes, they are all rooted in the way people go about doing their jobs on a daily basis, and for small businesses, with less people, and far easier personal communication, this is where the leverage can be applied by continuous improvement.

It costs the same to raise and process an invoice of $1,000 as it does for an invoice of $100,000. Therefore the transaction cost % of the invoice value is far greater for the smaller invoice. This relationship is reflected throughout the supply and distribution chain, and even minor improvements can deliver substantial savings. Technology offers the opportunity to reduce the absolute cost of processing to almost nothing, making the transaction cost irrelevant either way, but once people are added to manage the exceptions that cannot be handled automatically, the costs soar.

The source of Woolworths superior performance over the last decade compared to Coles has been the impact of their reductions in transaction costs that have dropped straight to the profit line. Wal-Mart became the biggest retailer in the world by focusing on the reduction of transaction costs of all types, and passing the savings on to consumers as lower prices to attract the volume creating a virtuous circle. Less obviously, they passed many costs back to suppliers, then continued to insist on and successfully extract cost reductions from those same suppliers in spite of increasing their costs, simply because of the scale of their sales potential for suppliers.

It seems to me there are two parameters to transaction costs:

      • The absolute amount of the costs in a whole process
      • The productivity of the costs in the process.

Most systems just look at the quantum, and set out to cut corners, work the current system harder, but by looking at the detail of the things that generate the costs, you can eliminate those that do not add value. However, moving a transaction cost on to another link in the supply chain does little to eliminate the cost, it just moves it. Retailers generally have been expert at this moving of transaction costs, while often creating them as a source of revenue. Practices such as making minor claims on a supplier, and holding up payment of a complete invoice until the claim is dealt with, then making the dealing with the claim a minefield for small suppliers abound. A source of the success of Aldi in Australia has been their focus on the reduction of transaction costs, but in return they get their “pounds worth” at the invoiced price point.

In dealing with supermarket retailers over many years, a number of transaction cost types have become evident:

      • Cost of searching, storing, processing & managing information. Category management is a prime suspect here. Suppliers engage in a costly, data intensive exercise in the expectation (hope in most cases) that there will be returns from the collaboration that is hoped to occur, and from the opportunities good category management can unearth. While the costs of the data transactions themselves may have dropped precipitously over the last 20 years, the costs of the overheads to manage them have not.
      • Cost of negotiation. In almost any negotiation where one party has the power, and is happy to use it, the outcome is virtually pre-ordained, it is just the quantum of the cost that is in question. Knowing, and sticking to your “Walk away” point is an absolute must.
      • Cost of time. A vastly under measured cost in most businesses. We tend to have people on staff because there is a job to be done, and we pay them competitive rates to ensure we get the best people we can for  the job, but we tend not to measure the value delivered by the doing of the job, its cost is just a part of the fixed overhead. Every minute spent costs a business, but apart from VC operators who use “burn rate” as a key measure, we tend to ignore it.
      • Cost of certification. The range of certifications that are supposedly “needed”  from HACCP to OH&S, to quality verification of components in a product to various religious and quality standards are legion. Each costs time, money, effort, and carry heavy opportunity costs. A bit of effort to isolate those that are really needed, and to manage those that are with automated or at least consistent processes can save a significant amount of time and money
      • Cost of influence. People deal with people, not corporations, no matter how automated and impersonal our communications systems become. Getting to know people , building relationships and trust takes time and effort. It is time and effort well spent, to a point, and finding the point at which the costs outweigh the benefits is a management challenge most fail.
      • Costs of cock-ups and rework. This is probably the biggest, most pervasive  source of transaction costs. From the wrong invoice to a truckload pf product turning up to be rejected, and turned around dumped or put into rework.  It is not just the cost of the product, but the added time, lost sales, loss of reputation, and needless consumption of capacity that really hurts. “Lean” processes target waste, and this one is the biggest waste that occurs, and is often made up of a lot of low hanging fruit if you go looking for it, and know where and how to look.

Small businesses are in a great position to reduce their transaction costs, simply by being good at everything they do, and being “close to the action” can make the wrinkles that can be ironed out that more obvious.

The original post that started the series is here, followed by the more detailed posts, 1, 2, 3, 4, 5.