Home > The Lean Post> Ask Art: Why Does Boosting Inventory Turns Matter So Much?
The Lean Post
Sharing how the world is making things better through lean.

Ask Art: Why Does Boosting Inventory Turns Matter So Much?

by Art Byrne
September 13, 2018

Ask Art: Why Does Boosting Inventory Turns Matter So Much?

by Art Byrne
September 13, 2018 | Comments (10)

Well, to me inventory turns should be a focus for every company because it a key to creating flow. This is true in both manufacturing and non-manufacturing companies. A non-manufacturing company might say, “hey wait a minute I don’t really have inventory in the traditional sense so this can’t apply to me.” Sure they may not have warehouses full of various parts like a manufacturing company does but they still have what I would call “inventory” but it just takes on a different form. For example, a hospital’s inventory consists of a bunch of human bodies that you are trying to move through various processes as quickly and efficiently as possible. A bank or insurance company always has an “inventory of say loan applications or requests for insurance that are waiting to be underwritten. The faster you turn this inventory the shorter your lead times and the better your customer service.

For manufacturing companies I see inventory turns as one of the most important measures you can have. In fact, if you wanted to simplify things and run your manufacturing company using only two measurements then the two that I would pick would be 100% on time customer service and inventory turns. These two measurements will drive everything else. If they both are increasing then everything else will inevitably be improving as well. Productivity, quality, sales growth, lead times, earnings, return on capital and most importantly enterprise value will improve. In fact, you cannot increase both customer service and inventory turns and not see improvements in all these metrics as a result.

But why inventory turns? Isn’t that just some minor “manufacturing thing?” And how can I have good customer service without carrying enough inventory to meet customer demands? I think we have all been taught that having extra inventory “just -in-case” something goes wrong is essential for any business. The traditional management approach is to produce things in big batches. Our finance gurus tell us that is how to have the lowest cost. “Sell one- make 10,000” is one way to think about this. Add to that the fact that Sales and Marketing don’t really trust Operations (i.e. the  manufacturing department) and are always arguing for more inventory to be safe. As a result, you will find the idea that the best way to increase customer service is to increase inventory turns (i.e. reduce inventory) to be a very hard sell in most traditionally run companies.

So why even think about this and why does it work? Let’s remember that lean practice keeps the main focus always on removing waste from your own operations in order to deliver move value to your customers. In his list of 7 types of waste, Taiichi Ohno cites Overproduction (i.e. the building of excess inventory) as the number one waste. So if we want to deliver more value to our customers such that we can grow and gain market share then we should certainly focus on reducing inventory. This will free up a lot of cash that is currently being wasted (I call it “sleeping money”) that can be reinvested in new products, new equipment, or acquisitions that will help expand our market share. And of course as you reduce inventory you also free up a lot of floor space (where the “sleeping money” used to be) that can be used to accommodate this increased growth. Your costs will come down, as carrying this excess inventory is expensive both from a capital invested point of view and from all the hidden operational costs involved. I say hidden because most accounting systems can’t capture the cost of moving, storing, moving again, damaging and eventually writing off a portion of this excess inventory.

Now that seems pretty straightforward and should make sense to most people. But like everything else in lean, the fact that it is the polar opposite of everything the traditional management team has been taught, means it will face resistance. In this case fear might even be a better word. “Heck, we’re struggling now with the inventory levels we currently have, how will we survive with even less inventory? You must be nuts.”

So, how you go about this is extremely important. In the past many companies have followed the path of “lower the water (i.e. the inventory) to find the rocks, then fix the rocks and lower it some more.” This usually leads to disaster in a very short time which leads to, you guessed it…management reverting back to building even more inventory.

To avoid this you need to understand why most companies build in big batches in the first place. It is normally because they take their setup times for granted, as in…wait for it…“Nothing we can do about that.” The lean manager, however, knows this is not true. Setup times can be drastically reduced without much capital spending. Our experience at Wiremold, for example, on many different types of equipment showed that we could cut set up times by about 90% during a one week kaizen. You can’t spend much money in a week. Even with that we kept going back again and again to reduce setup even further. Some examples; injection molding machines from 2.5 hours to 2 minutes, a 150 ton, coil fed punch press from 3 hours 10 minutes to 1 minute, a rolling mill from 14 hours to 6 minutes etc. etc. etc. Machines that we used to change 3 times per week were changing 20-30 times per day. As a result our batch sizes got smaller and smaller. Inventory turns increased dramatically and along with it our customer service was vastly improved. Our lead times went from 4-6 weeks to 1-2 days. We were growing and gaining market share. We freed up over half our floor space and used the cash from the inventory reduction to purchase 21 companies over the course of about 9 years.

But the lean leader understands that getting better inventory turns is not just about setup reduction. Changing from a functional organizational structure which by necessity batches everything to a value stream structure with a cellular layout where product can flow at the pull of the customer is another necessary step. Doing this will further reduce floor space, improve quality and productivity, shorten lead times and give a big boost to inventory turns. In addition, it allows the entire workforce to see the product built complete from raw material to in the box whereas in the functional structure most people only saw component parts and not the finished product. The same is true for say a bank where a loan application may have to travel through seven different departments all located on different floors or a hospital where you have to go through 9-12 separate fiefdoms to get treated for even minor injuries.

Increasing inventory turns is also a key to creating the learning environment which is so key to becoming a lean enterprise. Every time you remove more inventory learning occurs as the team has to figure out how to serve the customer without the crutch of excess inventory that was present in the prior functional batch state. The excess inventory was just there to hide the waste that existed. As the inventory comes out this waste must be eliminated and everyone is responsible for making this happen so lots of learning occurs every day.

So, ok, increasing inventory turns can drive a lot of growth and profitability. We increased operating income by 13.4x and enterprise value by just under 2,500% over these same 9+ years, but what should be the measure or target? When we started our inventory turns were 3.4x. We set an initial target of 20x. “Say what, 20x, why not 5x or something realistic?” Well, and I think this is very important, we set 20x because we wanted to change the conversation. We needed people to start thinking outside the box. Maybe you could get to 5x from 3.4 buy just working a little smarter but mostly staying the same. Getting to 20x on the other hand requires a total change in mindset. Once you get there however everything will have changed. Your customer service will be better and your lead-times will be way shorter than your competitors, weeks to days. Your costs will also be much lower and your quality much higher. But don’t stop there. Once you get to 20x then set a new target of 30x. This will be much harder but you need to keep driving out the waste. Even at 20x the company who is still at 3-4x inventory turns can’t compete. The changes that had to be made to get to 20x put that company in a totally different category than the company that stayed in the 3-4x state.

For acquisitions this focus on increasing inventory turns was a home run. Not only did we free up the cash to do the acquisitions in the first place but most of them were only turning inventory about 3x. Boy this was yummy. We knew we would be able to get those turns up to 6x by the end of the first year and to about 10x by the end of the third year. Combined with the rest of our lean implementation we were able, for the most part, to get all of our purchase price back in cash by the end of the 3rd year and then these companies were contributing cash towards the next new product or the next acquisition. The gift that keeps on giving, so to speak.

We of course had our suppliers in lock step with daily deliveries so that we always had raw material to respond to customer demand. As an example our biggest raw material was steel. We went from carrying 4 months worth to about 2 days worth and never ran out. We got 6-8 truckloads of steel per day but a good pull system can make this work very smoothly. So contrary to the traditional approach increasing inventory turns is the best way to increase your customer service and overall financial results. Put your focus on tracking the two measures I outlined above,100% customer service and increasing inventory turns. I’m sure you will get great results as well.

The views expressed in this post do not necessarily represent the views or policies of The Lean Enterprise Institute.
Was this post... Click all that apply
25 people say YES
33 people say YES
27 people say YES
20 people say YES
10 Comments | Post a Comment
Phil Coy September 14, 2018

Art, another excellent post as usual.

I come at lean from a quantitative perspective and get to the inventory turns that a value stream is able to support rather than just focusing attention there.  This may be more applicable just for HMLV where changeover cannot be ignored as you mention in your post.

EPEI is again my key metric.  The length of the interval is the size of the inventory buffer needed to prevent stockouts.  So I calculate EPEI and use  that result to calculate my ideal inventory level.  This is math not just focus.

If I know the amount of inventory that I should carry vs. what I actually have then I can strategically identify kaizens to see why I am holding more than I need.

With the right software, the value stream model can then calculate the turns in advance and you can design value streams that acheive your goals for turns rather than just waiting to till you see where inventory is building up.  This is done in the design of the vlaue stream itself.  I find that design of lean value streams is sorely missing today.



Reply »

art byrne September 14, 2018

Phil, thanks for your comments, I'm glad you enjoyed the article. If you can get high inventory turns using your quantitative software driven approach then great. Count me out however as this type of approach is based on a bunch of assumptions of what is possible. For example if you assume a 3 hour changeover time I'm sure you can make a calculation about how much inventory you need plus some added cushion to avoid stockouts. That is nice but it tends to lock in the 3 hour changeover when in reality it is possible to get that changeover to 1 minute. I think it is better to drive for inventory turn improvements and let the obstacles to this, for example a 3 hour changeover, pop up and get highlighted so they can be removed.

Reply »

Phil Coy September 17, 2018

Art, If your current reality is a 3-hour changeover when you look at capacity don't you need to reflect that reality?  Otherwise how do you know in the moment what resources are needed today.  I would agree that there must be room for continuous improvement and would use the 5 whys to find out why they have a 3 hour changeover in their planning.  

Reply »

Phil Coy September 18, 2018

Art, maybe you are assuming that software is inherently slower than the production going on out on the floor.  If that's so, then I would agree with you, a time lag between planning for capacity and the actual capacity in the moment would make my planning approach unrealistic. 

However that's not the case.  Our integration of value stream design where the capacity is run can be updated from the execution system (ERP) in very close to real-time.  So capacity planning can be easily made for a day with plenty of time to incorporate the capacity analysis and update the result back into the execution system. 

My earliest example to adjust plans for actual capacity for the next day has been running daily since 2006.

In our most dramatic example we are calculating capacity and releasing orders for configured products with a batch size of one and a fixed lead time of 2 hours for a plant producing 25,000 units a day potentially every order unique.  And with an approach that is fully TPS in all aspects.  Our design tool is fully integrated with the execution (ERP), so you can make changes in your value stream design and have deployment-ready software in your execution system in a few minutes.  This includes the ability to design complex value streams with conditional branches to pick up dynamic changes in quality sampling rates on the fly. 

In some ways, we are turning the paradigm upside down.  I get concerned that our solutions may outrun the people.  That's never the case when people are already cross-trained have the flexiblity and responsiveness built in to their processes via playbooks so they can shift their work around dynamically to respond to changing mix and volume. And there's no substitute for that hard work.  You can't dodge the design work but you can make it a lot easier and faster with a lot less non-value added time for crunching the numbers.

Visual management is best however there times when visual management is just not practical with high mix and volume with dynamic changes.  Normally schedules are limited to just once a day, but that's a limit more on people's ability to absorb the changes and not by some artificial time delay.

I respect your choice to skip the software but you should be aware of the impact that technology is beginning to have on lean operations.

Bringing software to lean is just the natural extension of bringing software to every element of business. 


Reply »

art byrne September 21, 2018
1 Person AGREES with this reply

Phil, thanks for the impressive examples of what new software can do to support lean operations. I have nothing against that.  My point is that focusing on reducing inventory is one of the most important things you can do to become a lean enterprise as doing so reduces lead time and alows you to deliver more value to your customers, frees up cash and floor space that can be invested in new products and growth while at the same time improving quality and productivity. Anything that can help do this faster and easier is a plus. But eliminating inventory should be the priority not calculating how much inventory you need. The traditionally managed company will be happy to have a program that can tell them exactly how much inventory they need given certain assumptions about there current state. They will give you 20 great reasons why their set ups can't be reduced below 3 hours. You may help them understand how much inventory they need but they won't make much progress in becoming lean or getting much overall inventory reduction. The company focusing on inventory reduction on the other hand will start with set up reduction. At Wiremold for example we took a 150 ton punch press from 3 hours and ten minutes to one minute. At that level why would I need a soft ware program to tell me how much inventory I needed? We averaged an 80-90% reduction in changeover time over the course of a one week kaizen. So why would we focus on calculating how much inventory we needed instead of just trying to get rid of it? At the same time some of the software programs you talk about could be a big help once the set ups were reduced. We had a cyclical business for example and calculating how many extra kanban cards we had to add to the system as we went into the system as we entered our high season was very important and could have benefitted from the type of software you talked about.

Adam Howarth September 16, 2018

Great srticle - thanks for sharing. 

Reply »

Andrew September 21, 2018


Thanks for this - I've passed it along to some people who need a little help "getting it". Connects nicely with Michael Ballé's piece on kanban a while back, making the point that kanban is about improvement, not just "order and fill".

It's rather minor with respect to the objective of your post, but I do have one bone to pick: While patients accumulating in a waiting room (or elsewhere) are indeed a signal of problems, they are not inventory! Patients are customers for heathcare services. One of the errors in the "tools" approach to lean in healthcare has been using valuestream mapping in a less than thougtful way, leading to concepts of "patient flow", when in fact, what we should be focused on is the flow of services and information - are patients getting what they need, when they need it? It may seem like an academic distinction, but it really affects the way we approach problem solving and system design in the complex, interdisciplinary world of high-value retail service that is healthcare.


Reply »

art byrne September 21, 2018

Andrew, thanks for your comments. I wasn't trying to belittle your patients to make them akin to pieces of steel or plastic. I was just suggesting that if we could think of them a little differently then perhaps it would help people understand ways to improve the flow of how patients are moved through the various processes needed to treat them. For hospitals this has a lot do do with easing or eliminating of the interfaces between the various fiefdoms that they have to pass through to get treated. For example what would you have to do to allow the patient to litterally flow through all the processes without stopping? I know that sounds a bit nutty at first but is the type of out side the box thinking needed  to get real improvement.

Reply »

Alex Lerch September 21, 2018

Hello Art,

Thank you for this post.  I have always thought of inventory in the traditional sense.  I manage an Engineering department of a manufacturing company so we do a lot of purchasing and building traditional inventory.  Your first paragraph however made me think about how inventory can likely be broken down on a departmental level (for Example my inventory might be design projects, or requests for clarification from the shop or Engineering change requests) So figuring out ways to increase the speed or the response to these should be a high priority for us.  I found this quite facinating...  Now the question for my team becomes what to do to speed this up :)  Anyone out there have any experience with working on this kind of thing?

Thanks again


Reply »

art byrne September 21, 2018

Alex, thanks for your comments. I'm glad you liked the article. Yes, I think you can look at the various types of requests that you have in the que as a form of "inventory".Reducing it will occur as you speed up your response time. The best way to do this is to run a lot of kaizen events focusing on why it takes the time it currently does. Value steam mapping will be important here as it will allow you to "see" where the problems are. Your focus should be on creating flow and eliminating all the things that currently are done in some form of batch. I think you will be amazed at what you'll find. My guess is that you should be able to reduce the current time for your response to any requests by 70%+

Reply »

Please include links as plain text URLs only. Do not copy and paste directly from a web page or other document. Doing so may pick up additional HTML that will not function here.
URLs will be converted to functioning links when your comment is displayed on the site.
Here's an example:
See this article for more details: https://www.lean.org/whatslean